Coyote Tracks

"How are you leveraging AI in your technical writing?"

2026-04-06T00:00:00+00:00

I shared a LinkedIn post with that title last week. I’m going to share what I wrote there, but I’m going to ask a second question after that, which I may or may not share with the LinkedIn crowd later. If you’re thinking “but I thought you were an AI skeptic!”, hold on. Okay? Okay, here’s the post:

First and foremost, what we’re calling “AI” is not the AI of sci-fi movies. Generative pretrained transformers are statistical prediction engines; when they’re coupled with large language models, what they do is generate a statistically likely continuation of their input text. That’s it. Full stop. Now, it turns out that “that’s it” undersells how amazing that output can be. The statistically likely continuation of an input text that's a search query is a plausible answer to that query; the statistically most likely continuation of feeding a Github repository into an LLM with a command to write documentation for the code is plausible documentation.

But plausible is not the same as correct. The bigger and more complicated the task presented to an LLM is, the likelier it is to go off the rails—maybe just a little, maybe a lot. If you don’t know what the output should look like, you may not catch this. That’s the obvious peril of “vibe coding”; a similar issue arises with “vibe documenting.” The more constrained the task you give it, the better job it’s going to do at it. Claude will probably nail “write a Python script to transform the date formats in this text file” on the first try; “write a clone of Photoshop that runs as a web app,” not so much. Likewise, “write reference descriptions for this API, including input parameters, sample output, and error conditions” will get you a better first draft than “write a complete casual developer’s guide for this database.”

Second, the statistically most likely continuation of anything is definitionally the median. An LLM will be much faster at generation than a human, but at its absolute tippy-top best, it’ll be better than about fifty percent of humans at the same task…and worse than about fifty percent of them. The more difficult and expansive the problem you’ve given it is, the less likely it is to be at that absolute tippy-top best. It’ll make mistakes a human wouldn’t, because it will have made what appear to be incorrect decisions or assumptions: calling a library that doesn’t actually exist, or including a guide on Git basics in an overview of how to write game plugins. The thing is, it’s not actually making any decisions or assumptions. It’s just determined that statistically speaking, a lot of code in its corpus calls similar libraries in similar functions, and a lot of documentation in its corpus talks about version control after it talks about NPM modules (or whatever).

So, if you’re going to use AI in documentation:

Target constrained, clearly defined use cases that contain sufficient context the LLM will stick to the task at hand.</li>
Expect the output to be first draft quality, not production ready.</li>
Humans must review the output, both to verify the engineering aspects (do the samples work, does the API really have those calls, etc.) and to make sure it’s staying on topic.</li>
Do not expect the LLM to have your corporate voice or stick to your corporate style guide, even if you’ve prompted it to do so. It won’t.</li>
The human reviewer(s) should be using a Markdown linter. (There’s probably some automated way to handle this, but I think it’s a good idea regardless.)</li> </ul>

Okay. LinkedIn enough for you? Here’s the second question:
“Should you be leveraging AI in your technical writing?”
A tech writing group I’ve been loosely involved with the past few years, Write the Docs, has been consumed by the question. A writer in that group who lost his job but got a new one less than a month later rhapsodized about LLMing all the things to do it (rewriting his LinkedIn profile and his resume, editing cover letters, creating custom targeted resumes, etc.). People in the WTD Slack talk about rebranding themselves as “context engineers” or “knowledge engineers.” Technical writers, they say, have been writing clear instructional procedures for technical concepts for decades. That’s literally the job description. Ergo, we’re exactly who companies should be hiring to direct AI systems. We should be kings of the new world!
And, man, my urge to leave tech and open a tiki bar or coffee shop or take up woodworking has never been stronger.
I’ve talked about my</a> various</a> reasons</a> to be skeptical/wary of AI before, so I won’t rehash them. LLMs have become good code monkeys, but generated code is less likely to be algorithmic plagiarizing than images or, more broadly, any attempt at creative work. (Also, “less likely” is not the same as “never”.) Training models on copyrighted material almost certainly is a copyright violation if you don’t have a legal right to access it: it’s behind a paywall, it’s a non-freely-licensed ebook, possibly it’s just a blog article that’s only licensed for non-commercial use (like the Creative Commons “Attribution-NonCommercial-ShareAlike” license</a> I use). Literally all the current popular AI systems use corpora that violate copyright in that regard. So, you know, there’s that.
But even if we set that aside (which we shouldn’t), even if we grant the relatively rose-colored view of the first part of this post, there are two big questions.
First, for companies: do you want your documentation to be middle-of-the-road at best? Because that’s what you’re going to get if LLMs do most of the work. Companies that let AI write all of their documentation won’t even hit that mid-level bar. The AI drafts that I worked with were okay, in the way that a first draft written by a freshman intern hopped up on Red Bulls would be okay. Every section needed editing, and most sections needed to be entirely rewritten. Or moved. Or cut completely. The overcaffeinated robot intern doesn’t understand information architecture or ontologies. You may think you are going to get it there with sufficiently clever prompting; you won’t.
Did using an LLM save any time? In theory, it should. Even if the first draft isn’t that good, writing is almost always slower than editing. In practice, though, if I’m doing so much editing it basically is writing, it’s probably a draw. And if I’m not, the documentation isn’t as good as it could be. If the LLM adds any value here, it’s in giving me something to bounce off of, something to look at and say, That section doesn’t need to be there, it needs to be over here. The voice here is all wrong. This example is redundant and this other example doesn’t quite work, but I can find one that does.
Back in Write the Docs, there seems to be a consensus—a vibe, if you will—that users can’t tell the difference between AI-written and human-written documentation. As “documentarians,” we need to accept that and get with the times if we want to keep a job. To which I say: bullshit. Users can tell the difference between good documentation and bad documentation, just like they can tell the difference between good UX design and bad UX design, or good typography and bad typography. Maybe they can’t explain the difference, why this interface is good and that one is bad, why that poster looks beautiful and that one looks like ass, or Stripe’s API documentation is mostly terrific and Apple’s API documentation is mostly undercooked tripe. But that doesn’t mean they can’t tell.
So this leads me to the other question, for other writers: is this actually what you want?
Like I said, I can see ways to use LLMs and still be a writer. But prompt engineering is not writing. Could I do a job that involves just tweaking prompts to Claude until it emits something with the general appearance of product documentation? Sure, I guess. But I’m not convinced anyone will pay me enough to keep me in the amount of high-quality rum—we are not talking Captain Morgan here—I’ll require to deal with that.
So, should you be leveraging AI in your technical writing? Here’s a LinkedIn-friendly answer: If you wanna play around with it in the margins—using it for proofreading (not rewriting), or as a jumping-off point for initial drafts, summarizing, research—try it! But don’t take it for granted that it’s going to make you faster or smarter or better. Check its work constantly.
Here’s an honest answer: I am already so damn tired of thinking about this question.
I’m so tired of the fear that I’ll be penalized if I do not sound sufficiently enthusiastic about a tool which is already being used to replace people in my field despite being demonstrably incapable of doing so. (“Now that we have electric screwdrivers, we’ll never need trained carpenters again!”)
I’m so tired of the entire tech industry willfully ignoring the signs that AI is a bubble, that it is being wildly oversold and jammed into places where it is inappropriate or even dangerous.
And frankly, I’m not just tired, I’m angry seeing other technical writers buy into the idea that the only way they can stay in their career is by helping people who never took the field seriously to start with relentlessly devalue it. Maybe I’m a weirdo here, but I’m a writer because I enjoy writing. I’m a technical writer because I like the work of taking complex topics and figuring out best to present them to a specific audience. I take pride in my craft. I think there’s even some art to it.
I’m not saying it’s impossible to take pride in figuring just the right words to type to get Gemini to spit out the best possible mid-tier documentation it’s capable of, but—actually, scratch that. I think that’s what I’m saying.
Now, if you’ll excuse me, I’m going to go work on a tiki cocktail menu. You know, just in case.
To support my writing, consider a tip on Ko-fi.com.

What you're allowed to do

2026-03-15T00:00:00+00:00

John Gruber linked</a> to a piece by Sam Henri Gold entitled “This Is Not the Computer for You”</a>, which both is and isn’t an article about the MacBook Neo. It’s a reflection on how kids get started with computers, and how when he was a kid, “every limitation was just the edge of something I hadn’t figured out yet.” It’s lovely writing, but these few lines struck an unexpected chord of recognition:

The limits you hit on the Neo are resource limits—memory is finite, silicon has a clock speed, processes cost something. You are learning physics. A Chromebook doesn’t teach you that. A Chromebook’s ceiling is made of web browser, and the things you run into are not the edges of computing but the edges of a product category designed to save you from yourself. The kid who tries to run Blender on a Chromebook doesn’t learn that his machine can’t handle it. He learns that Google decided he’s not allowed to. Those are completely different lessons. </blockquote>
What I recognize in this is the most common charge I see against Apple products on places like Hacker News, and from HN-minded people: that the only thing you can do with them is what Apple allows you to do. That they’re all so locked down, you’ll be constantly fighting against arbitrary restrictions.
Yet, I’ve seen few good responses to the challenge, “What can’t you do on a Mac that’s impossible because of arbitrary Apple-imposed restrictions?” If someone responds at all, they often respond with something that is possible and simply isn’t as easy to do as they’d like, or can’t be done in the way they’re accustomed to. The rest mostly fall into three buckets: UI preferences (for instance, “no focus follows mouse”); limitations that aren’t deal-breakers for non-Apple products (non-user-replaceable batteries, anemic gaming performance); fact-free insistence that macOS either will be or already is “just as locked down” as iOS is. The most valid criticism is that Apple is ruthless than other companies in cutting backward compatibility; I have one remaining 32-bit Intel Mac application that I wish I could still run.
What Gold’s article made me reflect on is that, in a quarter-century of owning Macs, I’ve never felt “arbitrarily restricted” from doing anything meaningful. I have a plethora of scripting systems and utilities that hook into the OS and applications at deep levels. I can install tens of thousands of Unix apps via Homebrew. Sure, the tightening of various security models eliminated Unsanity’s Haxies</a>, but losing those weren’t dealbreakers.
As for the iPhone: of the Android users I know, few have ever installed an application from anywhere other than the Google Play Store. I doubt many others regularly do things they couldn’t do on an iPhone, even when they think they are. There are definitely areas where Android is ahead for “normal” users, like pairing with smart accessories that aren’t made by the phone manufacturer. Or as a friend who went into a rant about this some time ago would point out, Bluetooth file transfers. (Most iPhone users would likely have the same response to that I did: “they don’t support floppy disks, either; what year is this again?”) But in practice, I suspect most acquaintances of mine who specifically chose an Android phone because “it’s more open”—because of vibes—would have been just as happy with an iPhone. And bluntly, nobody should choose an Android tablet over an iPad unless they have a very specific need, like running an app that simply doesn’t and can’t exist on iOS, or because they want an e-ink display. (I’ve toyed with the idea of getting a Boox</a> and trying to install Emacs on it to get a nerdy “writing deck,” but it sounds like more of a capital-P Project than I want to take on.)
It does feel like I face an ever-increasing amount of restrictions in my computing life, but with few exceptions, they don’t come from Cupertino. They’re imposed by moving not just services but applications to subscription-based models, and by rising digital surveillance. It’s hard to simply own your media and your software now, to browse the web without suspecting all your movements are being harvested to give to marketers, to data brokers, to ICE. While I trust Linux distributions to be at least as privacy-respecting as Apple, I certainly wouldn’t extend that trust to Windows or Android—and if you’re browsing the web with Chrome and using Google web services, it doesn’t matter which OS you’re using.
The MacBook Neo isn’t the computer for me, but it may well be the best first computer for Sam Henri Gold’s imagined child. Maybe they will become a developer, or a designer, or a filmmaker. But we need them to also become a digital rights activist. It’s not the Mac that will box them in⁠—it’s the rest of the tech world that’s made that their business model.
To support my writing, consider a tip on Ko-fi.com.

“Trim Silence”: Threat or menace?

2026-02-18T00:00:00+00:00

While I don’t think Apple Podcasts itself has a “trim silence” feature, it’s popular in a bunch of podcast players. Overcast</a> probably kicked it off with its “Smart Speed” function, but it’s in Pocket Casts</a>, Castro</a>, and I’m sure a bunch of other ones. The idea is that the player “listens” for silences in the podcast and (hopefully subtly) shortens them, so you’re saving a little time even when you’re playing the podcast at normal speed.
I suspect many users always run with Trim Silence enabled. I did, until I realized that occasionally Overcast interpreted very, very soft sounds as trimmable, subtly screwing up the timing of music. No problem: you can disable it for specific podcasts, right?
Well, if you do that, you’ll learn that podcasts that I think of as “NPR style,” e.g., ones with a certain kind of production value like 99% Invisible</a> and Pop Culture Happy Hour</a>, actually sound pretty good without Trim Silence enabled. In fact, they sometimes sound better. They’re edited precisely—even though PCHH itself is, despite being an actual NPR production, very much the “several dudes talking” kind of show unique to the podcast form. (With the asterisk that PCHH’s rotating panel is extremely diverse across just about any demographic line you care to name.)
Now, this isn’t a unique observation; I remember chatter a year or two ago on social media that I follow from people discovering this (and making dramatic denunciations of the option). But what I didn’t expect is what happened when I listened to podcasts from folks who are part of what one might cheekily dub the 5by5 Diaspora without trim silence enabled.
(Brief aside: 5by5 was a network a decade and change ago run by Dan Benjamin, with shows featuring hosts that pod-listening Apple nerds know: Merlin Mann, John Siracusa, Myke Hurley, Marco Arment, John Gruber, and more. It was huge for a brief time, then fell apart in just as brief a time, with next to all its hosts either jumping to the nascent Relay</a> or going independent. Someday I’d love to know the story of what happened there; reading between the lines, it sounds as if Benjamin was either difficult to work with, didn’t give hosts generous business deals, or both.)
My impression of many of those conversational shows was that even though they were relatively lightly edited, they were still edited tightly enough to take out awkward pauses. But listening to a few shows that I otherwise love without the Smart Speed or Trim Silence function at work, though, makes me question that belief. Are their creators are, consciously or not, assuming that everyone’s listening with Trim Silence enabled?
This could be a style choice, to be sure, the idea that the conversation’s going to sound more natural if you keep in the two or three seconds while a host gathers their thoughts between sentences or phrases. In a sense, it does. But I don’t think it’s the kind of “natural” you actually want in a podcast any more than you want it in a scripted show, where any long silence is deliberately written into the script. When I listen to these podcasts as recorded, I think: this could have used a little more editing.
Is this bad? No, not in the grand scheme of things. I am re-evaluating my podcast listens, in part because I don’t have the free time to keep up any more, and in part because I’m burning out on tech news in general and on AI AI AI AI AI AI in particular. Even so, I’m not going to stop listening to podcasts that I still like because of too-long pauses; I’ll just keep Trim Silence enabled for them.
To support my writing, consider a tip on Ko-fi.com.

Is AI helping you to do work, or just to do stuff?

2026-02-10T00:00:00+00:00

For someone who’s neither all-in on AI nor a passionate anti-AI crusader, I think about it a lot. From 2018 through 2023, I worked for Viv Labs, a startup from the folks who built Apple’s Siri; they had been bought by Samsung in order to fix Bixby, their much-maligned assistant. (The punchline here is that, to a large degree, they did fix it, but nobody noticed.) And my novel Kismet has emergent AI as a subtheme.
I don’t want to belabor, yet again, the very valid complaints about AI—the horrendous energy consumption, the way it’s being jammed desperately into everything, how all current models are trained on illegally-obtained material, and so on. But let’s nonetheless consider that Anthropic, the company that’s paid millions to settle copyright violation lawsuits, may well be one of the least unethical AI companies compared to their competitors with suicide advice engines and child porn generators. The industry as a whole is plagued with moral issues that go above and beyond the rest of Silicon Valley’s, and that’s saying something.
To be clear: the lack of a high-quality model trained entirely on legally accessible content casts a shadow on even otherwise innocuous uses, and I’d personally be leery of building a business that depends on current AI models.</aside>
And yet, and yet, I find myself uncomfortable with a maximalist anti-AI position. Yes, current models incorporate pirated material, but there’s nothing intrinsic to the technology that suggests you couldn’t have a fine model trained entirely on legally accessible content. (I don’t buy the argument that the act of training itself is a copyright violation; if a search engine can legally get to it, a large language model should be able to, also. That LLM scrapers seem to be incredibly bad net citizens is an artifact of LLM companies being run by shitty people.) Yes, LLMs make mistakes when they parse natural language, screw up summaries in sometimes egregious ways, make imperfect transcriptions of audio—but they’re nonetheless a quantum leap past previous attempts.
But the sudden virality of ~~Clawdbot</s> ~~Moltbot</s> OpenClaw snapped something else into focus for me recently. In case you missed OpenClaw’s story, I’ll let TechCrunch explain it</a>:~~~~

According to its tagline, [OpenClaw] is the “AI that actually does things”—whether it’s managing your calendar, sending messages through your favorite apps, or checking you in for flights. This promise has drawn thousands of users willing to tackle the technical setup required, even though it started as a scrappy personal project built by one developer for his own use. </blockquote>
On one level, that is absolutely the kind of Star Trek stuff AI optimists dream of. MacStories’ Federico Viticci, who’s recently poured his long-standing love of automation and scriptability into a 24/7 bear hug for productivity-focused generative AI tools, enthused, “OpenClaw showed me what the future of personal AI assistants looks like</a>.” (I am convinced that an alternate universe version of Federico is a legendary Emacs guru, given that I doubt there’s a more scriptable editor in the world.)
On another level, it’s a massive risk to both your security and your wallet. That TechCrunch article notes an investor wryly pointing out, “‘actually doing things’ means ‘can execute arbitrary commands on your computer.’” If you have OpenMoltClawBot taking actions for you based on incoming mail and messages, it’s possible for someone to send a message to you that takes action you don’t want. As for the risk to the wallet, Viticci breezes past “I’ve been playing around with [OpenClaw] so much, I’ve burned through 180 million tokens on the Anthropic API”—but that’s over $500.
In part, these flaws stem from OpenMoltClawBot being dim in the way all LLMs are. If you ask it to run a task at sunrise, it may write a cron</code> job</a> that calls the Claude API every fifteen minutes with the question “is it sunrise yet?” By literal definition, an LLM’s “smarts” are the median of what it’s been trained on: they will never write like the mythical “10× programmer.” Now, there may be some other flaws stemming from OpenClaw itself being vibe-coded. Its author has literally said “I ship code I don’t read</a>.” I, for one, see no problem giving code no human has so much as glanced at access to my personal files and bank accounts. What could possibly go wrong?
But let’s assume that your personal installation of OpenMoltClawBot doesn’t wire your entire 401(k) to a script kiddie in Hackensack. Even with that relieving condition met, how much time are you spending with it that’s doing stuff rather than doing work?
See, this is an old joke/truism about productivity systems: it’s easy to mistake Getting Things Done™ for getting things done. Spending time trying out different task managers, defining tags and projects, setting up your Inbox Zero system—it feels productive! It feels like you’re making progress! Variants of this appeal to a certain kind of mindset: you’re always on a search for the perfect task manager, the perfect pen and paper notebook system, the perfect</a> text</a> editor</a>.
But a lot of your time is actually being spent in overhead, whether it’s managing the system you’ve chosen or switching between systems you’re trying out. Time spent organizing your tasks is time not spent doing tasks. Time spent fiddling with new text editors is time not spent writing. (As much as I’ve taken to Emacs, its endless fiddlability can be a double-edged sword.) A lot of the time you think you’re doing work, you’re actually just doing stuff. Stuff is not work.
If you’re spending a lot of time brainstorming things you can get OpenClaw to do for you, you’re spending more time than you think doing stuff. Maybe more time than you’d be spending actually just doing the work yourself. The examples of integration that Viticci shows in his article are, without a doubt, damn cool, but some have a distinct “I built it because it was possible” air to them. Is it neat that I could tell my virtual assistant in Telegram to add two things to my to-do list and then resume playing Spotify on the computer that I’m sitting in front of at the moment? Yes, but I can already press a global shortcut anywhere to pop up a window to add something to my to-do list. If I want to resume playing Apple Music on the computer that, again, I am sitting in front of at the moment, I can (ahem) press the play button.
Emacs Org Mode nerds would also point out that, if you live your writing life in Emacs, there’s a global capture key for adding not just to your to-do list, but almost anything—with customizable templates to boot.</aside> But what’s started to get a little crazy-making for me is the “AI will inevitably eat everything” attitude, whether in Viticci’s article or in Marco Arment’s prophesying in a recent episode</a> of the Accidental Tech Podcast</a>. To quote Viticci: When the major consumer LLMs become smart and intuitive enough to adapt to you on-demand for any given functionality—when you’ll eventually be able to ask Claude or ChatGPT to do or create anything on your computer with no Terminal UI—what will become of “apps” created by professional developers? I especially worry about standalone utility apps: if OpenClaw can create a virtual remote for my LG television (something I did) or give me a personalized report with voice every morning (another cron job I set up) that work exactly the way I want, why should I even bother going to the App Store to look for pre-built solutions made by someone else? </blockquote> Any given functionality? Really? Do you actually think you’ll eventually be able to ask Claude or ChatGPT to write a word processor for you? A photo editor? The next Zelda game? No, let me rephrase that; I mean, you can ask Claude or ChatGPT anything, and it’ll plunge ahead with the absolute confidence that comes from being a predictive text generator being sold as a robot brain in a jar. But do you think that what comes out is going to be good? Really good? Overcast</a> good? Halide</a> good? Scrivener</a> good? How long do you think it’d take you to even begin to approach any of those doing vibe coding? How much human interaction would be needed? How much actual programming and design knowledge would the creator need to arrive at anything remotely usable? Once again, the best an LLM can generate is by definition statistically median. This is not just a temporary limitation of current models; it’s a fundamental artifact of the vector mathematics that makes LLMs work. If you, the user, are patient and clever enough like Federico, you’ll be able to cajole better output. But cajoling an LLM into designing a large, consumer-focused application, from a clean internal architecture to all the nuances of good UX? I’m pretty skeptical. I know many people aren’t. I suspect those people are, over the next few years, going to prove me right in the most depressing way possible: flooding app stores with LLM-generated crap riddled with confusing UX choices, ugly design, memory leaks, and terrifying security holes. I don’t doubt that LLMs can become a primary interaction method we use to interact with certain apps for certain tasks, and yes, agentic AI could become a primary automation tool for those who care about that level of automation. But how many people is that? Viticci describes having set up a Zapier automation</a> that creates a new project in his to-do app for the next issue of his weekly newsletter by monitoring the newsletter’s RSS feed and creating the project after the current newsletter appears in it. Now, he’s replicated that in OpenClaw, which potentially saves money! (Accent on “potentially.”) But most people, faced with a similar problem, would have just created a repeating weekly reminder for “start new newsletter project” and called it good enough. This is a chimera nerds have been chasing for decades: surely, surely, the only reason the majority of computer users don’t write their own programs and automations is that it just isn’t easy enough to do it yet. When I say that I don’t think generative AI is going to finally crack that particular nut, it has nothing to do with AI: I don’t think there’s actually a nut there to crack. To most users, programming and automating is doing stuff, and they want to do work. (Or play the next Zelda game.) Again, I’m not rabidly anti-AI. But in the same way generative AI actively encourages us to ascribe powers to chatbots that they simply don’t have—a simulacra of absolute confidence at any given task is literally baked into LLMs—it encourages us to mistake doing stuff for doing work. Studies show that developers who use AI assistance consistently overestimate how fast it makes them, and I’m not surprised: it feels fast to use AI assistance. My own experiments using AI to draft technical documentation (at the behest of an all-in-on-AI employer) are great illustrations of that. The AI “writer” generated hundreds of lines of stuff that was mostly correct, most of it usable with a few tweaks here and there. But when I look at the diffs between the machine-generated first draft and my final copy, I wasn’t making “a few tweaks,” I was massively rewriting the whole thing. Can I swear the LLM didn’t help? No. If nothing else, it kickstarted the process. And I know that whether I’m coding, writing docs, or even writing fiction, I feel really damn productive when I’m refactoring or editing line after line instead of staring an empty, foreboding text window. But am I doing a lot of work, or am I just doing a lot of stuff? To support my writing, consider a tip on Ko-fi.com.
Emacs and Shortcuts 2025-12-07T18:41:41+00:00 It’s a match made in…frankly, I’m not sure where. But I’ve been experimenting with using org-journal</a> as a replacement for Day One, and while it’s pretty good, there are two things that I immediately missed: The way Day One records my location and the weather for every entry.</li> Being able to quickly insert photos that I’ve taken into an entry.</li> </ol> Emacs can’t do that directly, but there are Shortcut actions that do everything I need there, right? So there’s got to be a way to get Emacs to talk to Shortcuts. Well, there is a Shortcuts package</a> for Emacs, but a bit of poking at it revealed that it seems to be just one way communication: it lets Emacs fire off a Shortcut, but doesn’t let the Shortcut return anything back to Emacs. Not so useful. But you know what? There’s a shortcuts</code> command line utility which can not only run a Shortcut, but get results returned to it. As an example of how to use it, here’s the photo picker Shortcut I created (click on it for a bigger image): </a> The Photo Picker Shortcut </figcaption> </figure> The first action lets you select photos. Then, it resizes each one and saves them as JPEG files to ~Documents/journal/images/</code>. Lastly—and this is the part that needs to get back to Emacs—it turns each filename into an Org-style link and outputs it. So, what does Emacs do to call this (and receive it?) It is an extremely simple bit of Lisp code: (defun wm-journal-photos () "Select photos, export, and insert as org links." (interactive) (insert (shell-command-to-string "shortcuts run \"Photos to Org link\" | cat"))) </code></pre> The shell-command-to-string</code> function does what it says: runs a command and gets its output as a string. That string just gets insert</code>ed at point. The location and weather Shortcut is even simpler: it just calls the “Get weather at Current Location” action, then creates a text block with LOCATION and WEATHER as Org-mode properties. (Why, you might ask, does shortcuts run</code> have to be piped through cat</code> to work? You got me. It’s not an Emacs quirk, for the record, it’s an Apple quirk.) To support my writing, consider a tip on Ko-fi.com. Apple wins me over again, but it’s getting tougher 2025-11-21T00:00:00+00:00 Whenever I make a big purchase—a TV, a new computer, a car, an A/V receiver—I inevitably enter a long period of internal debate, arguments and counter-arguments, pros and cons. It doesn’t matter whether I’ve set money aside for this, how sure I am that yes, I do need/want/carnally desire the thing in question. Post-purchase, I can look forward to an immediate period of regret and self-doubt, although as with all things, it fades with time. For at least two years, I’ve pondered returning to a “laptop-first” computing approach. I haven’t done this since I went from a MacBook Pro to an iMac/MacBook Air combination in the late 2000s. At times, my iPad was my “real” portable computer, and I only took the laptop out with me when I absolutely needed to. Desktop/laptop works, as long as you sync all your working documents through Dropbox or iCloud or what have you. But it inevitably leaves you with some things on one machine and some on the other, and you end up with a lot of things in the cloud that you suspect you don’t need to keep there. And it also makes it hard to justify having more than one machine be “the big one”: if I have a desktop, that’s the big one, and getting a MacBook Pro just wouldn’t make sense. The MacBook Pro must become the desktop to justify itself. From summer 2025 on, I edged closer to buying a MacBook Pro with the M4 Pro CPU in it. When the M5 MacBook Pros dropped, they left me with a dilemma: wait for the M5 Pro CPU, or get a maxed-out M5 “non-Pro”? After long dithering, I settled on the latter. I knew I’d get a 14″, not a 16″ model, and the base M5 is still nearly twice as fast as my M1 Max Studio across the board in all things but GPU performance. I can “only” get 32GB of RAM, which is what the Studio has, but the truth is that I never come close to overloading that. Ditto for a mere 1TB hard drive. And, this frees the Studio to migrate into being a home server, which I haven’t had since my last Mac mini’s hard drive kicked the bucket. (Which also means the laptop can have even less on it!) I know the argument that the base MBP is “Pro in name only,” that you might as well get an Air, but if you really look at the specs, that’s not true. Besides, I love working at cafés and breweries where a nanotexture screen might really come in handy. But one aspect in my internal arguments, my point/counterpoints, was new this time around. For the first time in a quarter-century, I wondered whether my next computer should be a Mac. All that I wrote about last month in “If not the Mac, what?</a>” remains weighing on my mind. The sloppy direction of the Mac’s user interface, which used to be its crown jewel. The drive for ever more services revenue leading to ever more dubious in-system promotions from Apple itself. The obsession with control in the name of security leading to Vista-like permission prompts and, in edge cases, absurd hula dances to install supposedly “unsafe” software. The UI flubs can, and hopefully will, be turned around, though, and the other issues are more policy decisions than technical ones, and policies may change with new leadership. But they might not, and I have to reckon with that. Worse, sucking up to our current fascist administration is also a policy decision. I can’t disagree with John Gruber’s argument when he writes</a>, Faulting Cook for playing ball and kissing Trump’s ring (among other things) is like faulting a local business for paying into a protection racket—one in which the cops are complicit, and the whole scam exists at the behest of a crooked mayor. Moral rectitude can feel good, but not so much when the cops are burning your store to the ground. There is no authority to appeal to for help when the highest office in the country is running the protection racket. </blockquote> And yet, I can’t help believe that there’s a qualitative difference between reluctantly paying for “protection” and proactively funding the new ballroom. It’s the difference between wheeling to stay on the Don’s good side and wheeling to be the Don’s new best bud. Like Gruber, I don’t believe that Tim Cook is a MAGA believer; it stands against everything that Cook appears to espouse. But that’s the whole damn point. Apple has spent decades consciously chasing a demographic which leans left of center—positioning itself as the tech of choice for artists, creators, and even environmentalists. For them to now cozy up with the far right in the name of realpolitik is perilous. Just ask Tesla, which has been going down in flames like a Cybertruck hitting a tiny pothole. Now, again, I don’t think Apple is in danger of having most of their loyal customers stop being loyal, whether it’s over this or over Liquid Glass or over the way they treat developers or over their increasingly self-aggrandizing long-form commercials pretending to be keynotes. But they’re in danger of having their loyal customers (hello, hi) stop and think about new purchases in a way they wouldn’t have just a few years ago. That’s not a good sign. With me, they’ve won this round. For the design elements I care about—which include that annoyingly nebulous term, “fit and finish”—Apple’s hardware is still second to none. It’s hard to even get close in the PC world, and when you do, the PCs aren’t that much cheaper. And despite all the, uh, Tahoeness of macOS Tahoe, it remains more comfortable for me than Windows 11 or Linux with current versions of KDE and Gnome. It’s a mix of the solid UX foundations that have so far survived Alan Dye’s depredations, the apps that don’t have satisfying replacements on Windows or Linux, and the relative seamlessness of the Apple ecosystem. But the trend lines—both how surprisingly nice KDE is getting, and well, how everything Apple is getting—make me wonder if I’ll choose the same way in 2030. To support my writing, consider a tip on Ko-fi.com. The Emacs for everything obsession 2025-11-04T00:00:00+00:00 If you poke around the Emacs community, you’ll quickly find some—all right, loads of—users who want to do everything in Emacs. Not just writing, not just programming, not just publishing their blog. No. They want to read and write their email in Emacs. Read their RSS in Emacs. View their PDFs in Emacs. Manage their tasks and their calendars in Emacs. Browse the web in Emacs, even though as far as I can tell the options for doing so range from bad to bad. Before I got into Emacs myself, this struck me as profoundly weird. As both an old Unix user and an old Mac user, “the best tool for the job” has become deeply ingrained. Emacs is excellent at text manipulation and at leveraging its Lisp machine</a> heart for a remarkable level of extensibility, but surely that doesn’t make it the best tool for everything you can possibly wedge into it. Does it? As I started writing this, friend and former coworker JCS posted Living in the Browser vs. Living in Emacs</a> in his very Emacs-focused blog; he is one of those “Emacs for as much as possible” types, and explains why (and why it’s better than being so focused on the browser): When I’m in Safari there’s not much I can do to adapt it to my way of working. Sure, there are some UI adjustments and other settings to tweak […] but basically I have to work the way the browser authors thought I should. Even if I were using, say, Firefox, which is open source and theoretically open to user change, browsers are sufficiently complex that user modifications are not a realistic option. Emacs is different. Even putting aside its extensive user level configurability, it’s easy to modify the way any particular function works. You can make your desired changes, install the source in your init.el and those changes will be reflected whenever you load Emacs. Similarly, you can write entirely new functions in the same way. You don’t need to understand everything about Emacs to do this, only enough Elisp to express your desired behavior and a few simple rules to install it. </blockquote> I’ve also found this to be true—yet it also explains why there are tasks I’m not interested in stuffing into Emacs. As I’ve written before</a>, Emacs is manifestly not a Mac-assed Mac app</a>. It’s a native app, and it adopts more Mac conventions than your average Electron app does (it even works with the Services menu, mostly), but that’s not its forte. This isn’t a knock against Emacs; when it comes to text manipulation, it kicks all kinds of ass, and I do find myself wanting to move all those tasks inside it. I don’t want to just replace Panic Nova and BBEdit, I want to replace Ulysses</a> and Scrivener</a> and Plottr</a>. (Okay, technically I replaced Scrivener with Ulysses, and Plottr is a weird visual timeline app, but the point stands.) This isn’t trivial. I’ve long asserted that editing prose is fundamentally different than editing code. Ulysses understands the difference intrinsically; Emacs needs to be taught. But other code editors can’t be taught. For instance, I regularly use one of Ulysses’ most wonderful features, the ability to copy Markdown to the clipboard as rich text; I spent about an hour hacking that together in Emacs. I have that saved as a GitHub gist here</a>. It works with Org, too, and has functions for copying Org/Markdown as HTML and Org as Markdown. I’m still working on using Pandoc</a> to replicate Ulysses’s exporting features, but I’m sure it’s not only possible, but will probably end up being a little more featureful and flexible than I could do before. For tasks that aren’t editing, per se, but still text manipulation, Emacs also kicks ass. Sometimes its solutions aren’t pretty, but they’re always extremely functional. Magit</a> is as good a Git client as you’re going to find anywhere else. The built-in Ediff</a> is, as traditional for Emacs, saddled with a somewhat inscrutable UI, but it’s an amazing three-way merge tool (and Casual Ediff</a> gives it a more scrutable UI if you want it). Tower</a> and Kaleidoscope</a> are terrific Mac-assed apps for those purposes, but are they $195 per year better? Cue someone explaining how $195/year is trivial if you’re a Professional. Let me offer a well-considered rebuttal: oh, shut up. If you’re already using Emacs, the answer is no—and I know of people who use Emacs only for Magit/Ediff, just like I know of people who use it only for Org mode (the plain text outliner and task/agenda manager). But while reading mail, RSS reading, and PDF viewing are all about text, they’re not text editing, and to me that makes them fundamentally different. For one thing, I’m often doing them on the iPad or iPhone, where Emacs isn’t available; more importantly, though, I don’t want to read long articles sitting at my computer desk. I want to read them on a couch, on a tablet or e-reader, using something that’s typographically pleasant. To me, that’s Unread</a> or Instapaper</a>. As for email, obviously Emacs is a great mail composer, but it’s not a great mail viewer, and I do more email reading than writing. If you’re the sort who uses Org mode as not just your task manager and outliner but your DEVONthink</a>, it may well still be worth it for that integration. I’m not, in part because I need my task manager to be well-integrated with the whole Apple ecosystem rather than the Emacs ecosystem. And that’s worth keeping in mind: a lot of the arguments for “Emacs everywhere” I see are arguments for how well different Emacs packages tie together. It’s absolutely true, but I’m on a Mac, where the combination of Shortcuts, Automator, Services, and third-party utilities like Alfred and Keyboard Maestro can tie everything together. Org-capture</a> sounds great, but I have “Things</a> capture” everywhere. It’s trivial to turn things into Apple Notes or Tot</a> scratchpads. I use Org where it makes sense to me—:keeping track of tasks at work, where I am in front of my Windows PC 100% of the time. I’m experimenting with using it for story notes and plotting as an alternative to Plottr</a>, which I love in spirit, but find unbearably ugly design-wise. I suspect I’ve learned enough Org at this point, though, to confirm that for my personal task management, it won’t beat the Mac-assed system I already have in place. I’m wise enough to never say never about moving more things into Emacs; it’s possible I’ll eventually jump through the hoops necessary to set up Emacs as a mail client and see if I like it better than MailMate</a>. But I might try MailMaven</a>, too; one of the things I don’t like about MailMate is that even though it is a native app, it doesn’t feel very “Mac-like,” and I’m going to go out on a limb and guess Emacs isn’t going to improve on that. A lot of the other things Emacs can do are things that I don’t see much advantage for me except in specific contexts. “But you never have to use the mouse” isn’t a big draw for me. For one, I have Alfred</a>, so I can do a lot without using the mouse</s> trackpad if I want, and for another, sorry: I like using the trackpad. I’ve been using pointing devices for decades, I’m pretty good with them, and I think keyboard-only jockeys frequently undersell their virtues. Now, though? I’m going to run a spelling and grammar check on this blog post, preview it, publish it, and commit it to my Git repository, all here in Emacs. I may not do everything in this program—but I don’t find myself closing it very often, either. To support my writing, consider a tip on Ko-fi.com. Collaboration or privacy: pick one 2025-10-23T00:00:00+00:00 Matt Novak, writing at Gizmodo on October 22</a>: President Donald Trump is currently demolishing the East Wing of the White House in one of the more shocking acts of his second term. Trump plans to build an enormous $250 million ballroom in its place, a structure that will dwarf the existing White House. But who’s paying for it? As it turns out, a bunch of private companies and individuals. And, presumably, they’ll now expect to get a lot more access to Trump. Trump invited tech executives to dinner at the White House in early September, and CBS News reports that the president’s meeting was “interpreted” by at least one company as an appeal for donations. That kind of language suggests that Trump is again utilizing the mob-boss style of language that he’s become known for. Trump can’t explicitly say, “give me money or I’m going to make life hard for your company.” But he’s suggested exactly that several times since taking office in January. Trump had another dinner at the White House just last week, and it featured some of the biggest names in tech and crypto, including Apple, Google, Meta, and Microsoft, as well as Coinbase, Tether, and Ripple. </blockquote> I’ve seen defenses of Apple’s past kowtowing to the Trump administration; they boil down to, what else do you expect them to do? Are we expecting Apple to lead the resistance? They have a business to run, and this administration has made it clear that they’ll make life difficult for businesses that don’t play ball. I get it, and I have as much sympathy for a giant multinational megacorporation as I can possibly muster here. For all its missteps and foibles, I love Apple’s products, and—up until the last decade or so—loved the personality they conveyed. “Think Different” wasn’t just a great slogan; it expressed a genuine philosophy. But there’s a difference between “playing ball” and proactively offering to pay for a new baseball field. Apple could be doing the least of what they need to do to get by in this current moment; they’re actively choosing to go above and beyond. Giving obsequious gifts is a choice, not a necessity. Taking down ICEBlock due to an informal demand rather than a legal injunction is a choice, not a necessity. And funding the destruction of the East Wing of the White House is a choice, not a necessity. While funding the Epstein Ballroom isn’t as egregious in a First Amendment sense as removing ICEBlock was, there’s a way in which it feels more troubling, because it can’t be taken as anything but servile submission. You can’t credibly defend that with what else do you expect them to do. So far, every time this administration has said “jump,” Apple has not only said “how high,” they’ve offered to throw a few more jumps in for good measure. Apple, in the Tim Cook era, has built much of their reputation on privacy, and from all appearances, they’ve genuinely meant it. Their platforms do offer a higher degree of privacy than you can get by default with Android or Windows. The Trump administration, by contrast, has built their reputation on going after all enemies, real or imagined, with all the tools they can dream up—chief among them being the rapid expansion of the surveillance state</a>. They’ve gone out of their way to monitor constitutionally protected speech</a>, to let ICE use facial recognition to track immigrants and tie their identities to “derogatory information”</a>, and to break down barriers between data silos so they can profile you without court oversight</a>. It doesn’t take much imagination to think of ways this administration might pressure Apple (or Google or Microsoft) to give up their users, to quietly—or even openly—build in the kind of back doors that Apple has, in the past, resisted</a>. Even a year ago, if you’d have asked me whether Apple would keep resisting that, I’d have answered “yes” without hesitation. Now, though? Even before this ballroom fiasco, we recognized that the question was no longer “will Apple cooperate,” that it had become “is there a line Apple won’t cross.” But every time you say “yes sir, yes sir, three bags full, sir” to someone like Trump, the more he’s going to ask for next time—and the less power you’re going to have to say no. Maybe there still is a line Apple won’t cross, but each new collaboration pushes that line farther into the distance. To an authoritarian regime, there can be no such thing as a right to privacy against the government. They require, they demand, the right to surveil their citizens in any and every way possible. And a company whose reputation is built on a promise of privacy cannot collaborate with such a government and keep that promise. I’m no longer confident that, at the end of the day, Apple has the courage to not cross that line. To support my writing, consider a tip on Ko-fi.com. If not the Mac, what? 2025-10-15T10:56:16+00:00 Back in the dark ages (the 1990s), I was an inveterate operating system switcher. At various points, I ran DR-DOS with DESQView/X (what Windows 3.1 should have been), FreeBSD, several distributions of Linux, Windows 2000, QNX Photon, and BeOS. With the exception of QNX, I didn’t just run them, I used them. Used BeOS? Full-time? For more than a year, very productively—but that’s another article. (Just say I wish some of the folks who’ve written or podcasted about BeOS in the last few years had mayyyyybe spoken with some of us who were there at the time.) BeOS is, ironically, what brought me to the Mac: I’d used Macs at college and loved them, but they were expensive and, by the mid-90s, falling behind the times technology-wise. And I liked Unix. BeOS was a kinda-Unix with a kinda-Mac UI, much better than what Linux and BSD had at the time, but arguably torpedoed itself through impatience. So when Mac OS X burst onto the scene, I was blown away. Full Unix in all its nerdery, combined with an Apple-designed desktop environment and the full complement of Mac software! For desktop use, I’ve been here since the days of the tangerine iBook. </a> I did not have the matching mouse. </figcaption> </figure> Even so, I kept running Linux on servers for years, and I dabbled with Linux desktops in VMs. As for Windows—well. I didn’t spend a lot of time with Windows, serious productive time, until my current job shipped me a laptop with Windows 11 Pro. It’s, you know, fine. There are a lot of UX decisions Microsoft makes that I still don’t like (the ribbon interface has made Office near-incomprehensible). And if you think Apple’s attempts to get you to try “Apple Intelligence” are annoying, boy howdy you have not seen annoying until you’ve seen Microsoft Copilot. And there’s all manner of little things that the Mac still does better, tiny nicks and cuts that add up. Trackpad acceleration curves. The relative fiddliness of grabbing window borders on Win11 compared to macOS. Accessing alternate characters the Mac way versus the Windows way. Okay. I’ve done my best not to be that guy who complains about new things—I really do like the look of High Sierra, and don’t pine for the days of Snow Leopard, let alone the pinstripe plastic aesthetic of early OS X. But macOS Tahoe marks the first time I’ve ever sat down in front of an Apple-designed UI and thought: this just looks bad. It’s not “Liquid Glass” that’s the issue, per se: it’s how half-assed the Mac implementation of it is. I already disliked the design direction they’ve been going in the last few years, hiding the incredibly useful proxy icon from title bars, making all the keyboard shortcuts in menus grayed out so they look disabled, and the like. Tahoe, though, makes the Mac feel like—well, like some of those early Linux desktop environments. You know what I mean even if you hate that I said it, don’t you? UI widgets just look too big, with weirdly low information density. There’s drop shadows under buttons for that First Day With Photoshop feeling. Sometimes icons or text labels don’t appear to be quite lined up correctly. Nothing quite goes together. It lacks fit and finish. Combine that with other directions Apple’s been taking in the latter part of the Tim Cook era, and I confess to a growing uneasiness. The focus on service revenue has brought us promotions in System Settings, which feels junky. Apple’s control-freak nature encroaches on the Mac a little more with each successive OS version; I can still do everything I want, yes, but more and more “edge cases” are becoming wrestling matches. They’ve never been great at fixing long-standing bugs, but it feels like it’s been getting worse with each passing year. SwiftUI seems woefully behind where it should be after six years, given that it’s supposed to be the future of UI development on all Apple platforms. Tahoe’s design choices have led to another round of “see, iOS and macOS are merging!” from various corners; while I remain extremely skeptical of that, it’s undeniable that their UX design is drawing closer together, largely to the Mac’s detriment. I do think it’s possible that Apple is working on an entirely new operating system to potentially replace both macOS and iOS, but that’s another article.</aside> And, of course, we haven’t gotten into Tim Cook’s decision to kowtow to the authoritarian-in-chief. Let’s just say that I think that it’s not just a bad decision from an ethical/moral standpoint, it’s a bad decision from a tactical standpoint. So I won’t lie: I’ve found myself thinking, over the past year and change, if I ever did leave the Mac, where would I go? Windows? Aesthetically, Windows 11 may beat macOS Tahoe, which shocks me. In the majority of other respects, though, it’s at best a wash—and for all of modern Apple’s missteps, I’d hardly give Microsoft higher marks in the ethics department. And, on a practical level, I want my damn Unix. Enter Asahi Linux</h2> This is the Linux distribution for Apple Silicon. So far it only supports the M1 and M2 processors, and the completeness of what it supports varies from model to model, but the M1 Air is pretty well supported. While the text-based installer isn’t the friendliest-looking by 2025 standards, it’s straightforward, running with minimal user intervention. After a much shorter time than I expected, my Mac rebooted into Linux. And…everything just worked. I didn’t expect it to be as fiddly as it was a quarter century ago, but it really is at a Mac/PC level of “just get going now.” (Like both macOS and Windows, there are bits and bobs where that fails, but not at an unusually high level.) But: I think it actually looks good? </a> </figure> It took me a bit to figure out how to make the file icons in the details view reasonably-sized, but hey, unlike Windows, at least it has that view. (Windows has a list view, but still no disclosure triangles to look in subfolders, because who would want that, apparently.) Since I still haven’t migrated away from 1Password, all my passwords are right there; since I recently switched from Panic Nova to Emacs, my blog editing/publishing system worked with almost no fiddling. And KDE Plasma’s UI chrome is…pleasantly subdued. Linux still has its own set of little nicks and cuts when you compare it to the Mac, but it’s nothing like it once was, and it’s definitely more in tune with my way of doing things than Windows is. So it’s hard not to start thinking, at this point, “if I really wanted to try ‘living’ in Linux for a while, what would I need to do?” I’m back on my Mac as I’m writing this section, looking at what I use day-to-day. The basics would be easy enough to fill in: a mail client, a Discord client, a Telegram client, a graphics editor, a music player. The big problem would be all the things that I do that are tied to the whole Apple ecosystem, applications that seamlessly sync between all my devices. Things</a>, my task manager, is everywhere, as is my weird math notebook program Soulver</a>, as is my creative writing program Ulysses</a>, as are Apple Notes and Tot</a>. The list goes on. While Apple critics see this as nefarious lock-in, it’s arguably Apple at its best: it really is great that all this stuff works together pretty seamlessly. As long as I actually have a Mac running macOS, this isn’t an insurmountable problem. If I really wanted to move completely off the Mac, it becomes a pretty huge one. Lots of people who don’t use Macs have iPhones, sure, but they work together tremendously well. I’d need another cloud storage provider other than iCloud. I could technically keep Apple Music, but I’d probably better off switching to Tidal or Spotify. (There are some arguments for switching to one of those even while staying on the Mac, because they work better with other non-Apple gear I’m already using, but so far I’ve remained stubborn.) I’d lose Vellum</a>, the amazing ebook (and even print book) creation program. I’d lose Acorn</a>, a photo editor which—in my personal opinion—blows the doors off anything I could get under Linux. (I’m willing to be convinced otherwise, but Acorn does a lot with a brilliant user experience, something GIMP</a> has never been accused of. We won’t even get into the cringe name they refuse to change.) Having enumerated all that, though: this thought exercise has revealed that the software isn’t the real blocker for me. Linux is good, but PCs still (mostly) suck</h2> In the past couple of years I’ve had two Windows laptops in the house. One was what I guess you’d call a low-end gaming laptop from Dell, borrowed from a friend when I was playing around with VR. The other is the work one, a fairly high-end Lenovo 15″ ThinkPad from 2023—the kind of model people often point to as the PC equivalent of a MacBook Pro. Dear reader, it is not. It has, to be clear, a much better GPU. It has two USB-A ports to go along with its two Thunderbolt USB-C ports, while the MBP has no USB-A ports and three Thunderbolt USB-C ports. They both have an HDMI port; the ThinkPad does not have the SDXC card slot the MBP does, but it does have a TrackPoint mouse, if that’s your thing. But in nearly all other respects, an MBP from 2023 kicks the ThinkPad’s ass twelve ways to Tuesday. For a start, the ThinkPad is plastic. Just like macOS Tahoe, ironically enough, the fit and finish is questionable. When the lid’s open, the gap between the lid’s bottom and the body of the laptop—a gap that doesn’t exist on an Apple laptop at all—isn’t perfectly even. The trackpad is, as always, not quite as nice. And it’s hardly fair to talk about the difference between the screens, given that a 14″ MBP’s screen is a 3024×1964 micro-LED display with a 120Hz refresh rate. Also, the Lenovo’s fan runs audibly even when it’s not using its separate, fancier Nvidia GPU; when it really gets going, it sounds like a hovercraft. You kind of expect it to lift off the desk. I could say a lot of the same things about my M1 MacBook Air in comparison. The M1, introduced in 2019, is only about 10% slower in single core performance than the 13th generation Core i7 from 2023. (It’s about 50% slower in multi-core, but it’s eight cores versus 14.) The Air just has a standard LED screen, but it’s brighter, crisper, and much sharper, with 2560×1600 resolution compared to the ThinkPad’s 1920×1200—almost twice the number of pixels in a smaller display. And the Air is fanless. Yes, this means that if you push it really hard it’s going to throttle down faster, but you don’t buy an Air because you want to constantly push it to its limit. I’m sure that there are Windows PCs that are closer to a MacBook in meaningful ways; the Microsoft Surface line is solid, and even Lenovo makes snazzier machines than this ThinkPad. But I wouldn’t want to run Windows, I’d want to run Linux, and I’d want to run Linux well. None of this “it’s pretty good except you only get half the battery life, the webcam doesn’t work, and if you have this particular WiFi chip getting online requires entering these 28 commands at the bootloader and sacrificing a goat” stuff. And, I’m also sure some readers are frothing at the mouth at my assertion that Apple makes good hardware at all, let alone some of the best hardware. What about the butterfly keyboard? What about heat dissipation? What about the poor DIY repairability? The butterfly keyboard is long dead, along with a few other bad design decisions dating back to the late Ive era; the heat issues are negligible for people who aren’t animating feature films on their laptops. Wouldn’t I prefer a cheaper and more repairable laptop? Sure, but not at the cost of living with an ugly, loud, dim plastic black slab from which no spark of joy can escape. So…?</h2> Unsurprisingly, I’m staying with macOS. I have a quarter-century of inertia behind me at this point, and I remain unconvinced that anyone else makes hardware this damn good. But I’m considering disentangling what I can of my computing life from dependence on Apple’s ecosystem. This will be a pain in the ass in all sorts of ways, and I don’t know how complete it will be. Do I want to move off Apple Music? Off iCloud Drive? Right now all of those feel like horrible ideas. But they could work. I could adjust to Tidal. I could move back to iA Writer, and be able to use Emacs for heavy lifting as a bonus. (iA Writer just operates on folders of Markdown files, rather than Ulysses’ iCloud-specific system.) I don’t want to move back to Dropbox, but it’s there, and it works well, and it’s cross-platform. Yet I want to drag my feet. Maybe next year’s macOS (I’m hoping for “macOS Yuba City”) will be substantially refined. Maybe there won’t be any more roadblocks thrown up to dissuade users from doing radical things like installing “unapproved” applications. Maybe Tim Cook will pull his head out of Donald Trump’s ass. These are all big maybes, though, and we’re living through an epoch where “free software, free as in freedom” may stop sounding like a pithy, slightly annoying slogan and sound like an essential, slightly annoying battle cry. So. I still expect, at this point, to buy an M5 MacBook Pro when the higher-end processors come out. I still expect, at this point, that if the weird writers’ brainstorming app I’m working on ever comes together, it’ll be a SwiftUI app. But I’m keeping my M1 MacBook Air booting into Asahi Linux by default. Just for practice. To support my writing, consider a tip on Ko-fi.com. Seriously, Markdown 2025-08-20T10:51:05+00:00 Every so often, there’s a surge of “here’s why no one can do serious writing in Markdown” articles. They don’t come on any obvious schedule, or with any obvious trigger; usually, somebody complains about something Markdown can’t do that {reStructured Text, AsciiDoc, Org Mode, DocBook, DITA, LaTeX, Pollen, Typst, a Selectric Composer you found at a garage sale} can, it gets picked up on Hacker News, and a few hundred nerds chime in to explain it’s because Markdown {has no formal specification, isn’t extensible, was just designed for blog posts, ran over their childhood dog}. When this happens, I look over from the technical writing I’m doing in Markdown, or possibly the next novel I’m writing with it, and nod thoughtfully. I should, I presume, prepare my resignation letter now, apologizing for not being a serious technical writer, while gently explaining to my current and past clients and employers the error of their ways. Yes, the same people on Hacker News a decade ago were holding up RethinkDB’s documentation as terrific, but if they only knew I’d been writing in Markdown, it would suddenly become unserious. The entire Microsoft Learning Portal? Away with it! And folks I haven’t worked with better be informed, too. Yes, Stripe’s docs are considered a gold standard, but they are written in Markdown! Get the Shame Nun in here, stat! All right, let’s be slightly more serious. Yes, there are countless different Markdown processors out there, and yes, all of them handle edge cases differently. But where they “fail” 99% of the time are edge cases, and if you follow reasonably good practices—the sort of nitpicky things that MarkdownLint</a> whines about—you’re pretty unlikely to run into trouble. I’ve been doing this for a long time, and I’ve never had to learn a “new” style of Markdown or unlearn bad habits from previous work. I know I can’t speak for everyone’s experience. But if you write Markdown like this on the regular: # My Great Document Who needs blank lines after headings; this works _most_ of the time! This next line will surely become the horizontal rule tag I intend. ----- - Bullet 1 - Bullet 2 - Bullet 3, which is not connected to the link below despite the lack of spacing [A [cool website](http://bar.com)](http://foo.com)! I'd really like to **emphasize *this and also** this* right now. </code></pre> And somehow you get away with it until you switch Markdown processors, I’m sorry, but the processor is not the issue. If you write crappy Org Mode markup, you’re going to get crappy Org Mode output. “Yes, but it’ll be the same crappy output no matter the processor! That’s the problem!” Yeah, but is it? Most other lightweight markup systems only have one or two implementations. If you’re using Org Mode, you are almost certainly using the actual Org Mode in Emacs; if you’re using AsciiDoc, you’re probably using AsciiDoctor to process it. AsciiDoc’s formal specification is still a work in progress, and there’s no formal spec for Org Mode’s markup, as far as I can tell. If there were only one or two Markdown processors out there, this probably wouldn’t be an ongoing debate. Of course, CommonMark</a> does give Markdown a formal spec, and in practice most Markdown processors are compatible with it in all but the edgiest of edge cases.</aside> Here’s my theory: what’s at the root of these complaints is that people want Markdown to be something it isn’t. It’s a markup syntax that represents a subset of commonly-used HTML. You can plug it into a content management system or a static site generator, but it isn’t a CMS or an SSG any more than HTML itself is. AsciiDoc and reStructured Text, on the other hand, are markup languages closely tied to specific documentation generators, AsciiDoctor and Sphinx. Org Mode is a plain-text structured editing system for outlining, task and agenda management, literate programming, and publishing to multiple output formats. They’re complete solutions for specific problems. If you have that specific problem and you want a complete solution, you don’t want Markdown. So: why do all these companies use it anyway? Why is your favorite documentation so likely to be written with Markdown behind the scenes? Well, I’m going to go out on a limb here: it’s because it’s a markup syntax that represents a subset of commonly-used HTML, rather than a complete solution. Markdown doesn’t care what content management system you’re using, or if you’re using one at all. How your site is constructed, whether you’re using a template system or a CSS preprocessor or crazy-ass client-side rendering, is not in Markdown’s domain. Use the CMS you’re already using—there’s probably a plug-in. Need extensibility? That’s your site generator’s issue. Write your own damn shortcodes. “Wait,” somebody’s surely saying. “How is that an advantage?” It’s an advantage because you have all of HTML and CSS at your disposal. If you know how to use them, you get exactly what you want, full stop. That’s not a goal of AsciiDoc or reStructured Text or Org Mode, or much of anything else I’ve seen proposed as better alternatives. (It is a goal of Dean Allen’s nearly forgotten Markdown predecessor, Textile, but like Markdown, Textile isn’t in and of itself a site generator.) I’m not positive that, say, Stripe’s documentation site couldn’t be reproduced in rST, but I think it’d be really damn hard, and a lot of that has nothing to do with the capability of the markup system. It has to do with the flexibility of the output. To be sure, you can integrate non-Markdown markup into another system. GitHub renders README files with a variety of lightweight markup styles; some SSGs and CMSs have plugins that let them process rST or Org markup or what have you; you can always set up a workflow that lets you export from your favorite format to HTML or (ahem) Markdown transparently. That’s great if you’re more familiar with another markup style and want to keep using it, but that’s not an argument for superiority, it’s an argument for familiarity. When you focus purely on the markup semantics, it’s hard to argue that any other system is “better” than Markdown. (I find text that uses Markdown markup to generally be more readable in its unprocessed form than most other lightweight systems, and that’s important to me, but it may not be important to everyone.) I’m not saying “give up and just use Markdown.” Being familiar with another markup system is a powerful argument for staying with it, for a start, and there are other reasons to use something else that have nothing to do with markup. There are specific kinds of technical documentation—usually in contexts where a lot of sections are shared between manuals, revised independently, and assembled programmatically—where AsciiDoc is your only practical alternative to full-bore SGML/XML. And as a newly anointed Emacs nerd, I find myself using Org Mode daily at work, and that includes writing extensive notes using its markup system, not Markdown. I’m using it primarily for its task management capabilities, though, because that’s the kind of stuff Org was designed for. But you know what else Emacs is excellent at editing? The lightweight markup I’m actually using when I write developer documentation for one of the most popular games in the world, published by one of the biggest software companies in the world? Here’s a hint: not reStructured Text. To support my writing, consider a tip on Ko-fi.com. It's the heat and the humidity; also, the fascism 2025-08-05T00:00:00+00:00 “It’s not the heat, it’s the humidity” isn’t as much an aphorism as a punchline, as people point out—often with an unwarranted hint of smugness, as if nobody else had made this observation before—that 95° F is too damn hot no matter how humid it is. And, I mean, sure, but imagine these two scenarios: Tampa: The high today will be 95° F. The low will be 75° F and the relative humidity will be 70%, for a peak dew point of 74° F and a peak heat index of 123° F.</li> Sacramento: The high today will be 95° F. The low will be 59° F and the relative humidity will be 40%, for a peak dew point of 67° F and a peak heat index of 99° F.</li> </ul> For those of you who do Celsius, Tampa is 24° C–35° C, dew point 23° C, heat index 50.5° C; Sacramento is 15° C–35° C, dew point 19.5° C, heat index 37° C.</aside> Is the hottest point too damn high in both cities? Obviously. But in Sacramento—my favorite underrated California metro area—it only sucks for the afternoon hours; the morning is pleasant, the evening is warm, and the humidity stays basically comfortable. Here in Tampa Bay, you wake up to warm, muggy air that turns into a sauna before noon. A sauna with a blistering UV index. In theory, I knew all this when I moved back to Florida. I grew up here! I understood what I was signing up for, right? Again, I mean, sure, in theory. But it feels worse this time around. In part, that’s simple reality: it really is hotter now than it was a quarter-century ago. Summer gets here earlier and leaves later. The storm season lasts a little longer and gets a little more violent. And, some of it’s subjective. I didn’t travel at all, not even a staycation, from July 2023 through June 2024, the longest stretch I’ve gone without a single night somewhere other than home in decades. I’m not actually in Tampa proper—I’m in the rural exurbs, so I can’t even go out for lunch breaks except to a handful of places. The net result has been a level of stir-craziness I haven’t felt in years. (It hasn’t helped that I gave up my “remote work Thursdays” since I moved back, where I picked up my work laptop and relocated to some entirely new place for the day.) Even so, my first two summers back in Florida—2023 and 2024—didn’t engender the kind of restlessness I’ve been wrestling with this summer, the lack of focus, the sense of frustrated wheel-spinning. Last summer was a summer of burnout</a> for me; this is a summer of…well, whatever the nervous energy version of ennui is. But, I know this year it’s not just the heat or the humidity. It’s the sense of helplessness. Fascism is here, a third of the country is greeting it with thunderous applause, and neither the legislative nor judicial branches seem motivated to provide either check or balance. The industry I fell into thirty years ago is in the midst of another bubble, and its leaders are more insistent than ever that it is not a bubble, it is in fact either the greatest invention since fire or an existential threat, while they’re also either providing direct support for American fascism or just sucking up to it. The Venn diagram of “techbros who have gone MAGA” and “techbros preaching the gospel of AI” is more or less a perfect circle, and this is something we, as a society, need to face in a way we’ve been doing our damnedest not to. I am much closer to traditional retirement age now than not; when I moved from Florida to California in 2002, I figured I would be out there for most or all of my working-for-someone-else career. Even as I fell in love with northern California and the Pacific Northwest at the same time Florida became just a little more hot, a little more humid, and a lot more politically antagonistic, I carried that around in the back of my mind. By 2020, though, I couldn’t. There are still places in Florida that, taken in isolation, I’d be happy living in—some neighborhoods around Orlando, some towns in Pinellas County—but the politics aren’t getting better any time soon. Also, despite what our mini-Trump governor claims, Florida is no longer cheap: housing costs here have climbed so fast in the last five years that Sacramento, Tacoma (a suburb of Seattle), and even Portland, Oregon are now cheaper to rent in than either Orlando or Tampa. That’s based on a spot check of rents in April 2025. Just before I actually did move back to Florida, I wrote, “it’s hard not to daydream of moving back even though I haven’t left yet”; I’m still daydreaming of it. I can’t—won’t—leave my mother behind, but even setting aside the Herculean logistics of moving an 80-year-old with chronic health problems (and a cat) cross-country, the places I want to live aren’t places she’s attracted to. On top of that, the house here is completely paid for, on an acre and a half of wooded land that has, so far, proven resistant to flooding. For all the mounting problems with Florida, those are compelling reasons to stay. For now. So that leaves me in a holding pattern, the one I’ve been in since July 2022. I don’t know what breaks it. The uncomfortable spectre with me for all that time has been that I am, in a real sense, here to take care of my mother until she dies, that considerations of where to move to next—where to retire to—are, for the indefinite future, moot. Increasingly, though, the heat and humidity are getting to her, too, and the still-rural community we’re in is getting a lot of house farm development happening. In a year or two, we’ll have the traffic of suburbia while still having essentially nothing urban—or even suburban—around us, the worst of both worlds. Add in two bad hurricane seasons in a row with a third (and fourth and…) likely, and we may hit a practical limit together. I don’t know what happens then, where we’ll end up, what it will mean for me when she does pass. Will I be somewhere I want to stay, in a home I want to keep? But for now, we persist here, together, and watch to see what the next year brings. To support my writing, consider a tip on Ko-fi.com. Dr. Neckbeard, or how I learned to stop worrying and love Emacs 2025-06-04T09:44:20+00:00 So, you might know that I’m a text editor junkie. At a job in the 1990s, I used a great Windows-only editor called HomeSite</a> for web development. When I moved to the Mac in 2001, I wrote to Allaire, the developers, and asked if there was a Mac version in the works. They wrote back and explained they’d written it in Delphi, a Windows-only development environment, so probably not. But, they added, had I tried BBEdit</a>? So I tried it, and loved it—and got the “always something more” itch. I moved from BBEdit to TextMate</a> when the latter hit version 1.5 and stopped sucking. After giving up waiting for the TextMate version 2</a> unicorn, I tried Sublime Text</a>, which was fine, but never really grabbed me. From there, I’ve been all over the map. Back to BBEdit</a>. In an on-again, off-again affair with Panic Nova</a>. Trying and failing to love Visual Studio Code, despite a certain level of grudging admiration. On the creative writing side, I’ve been through Markdown-specific editors from iA Writer</a> to MultiMarkdown Composer</a>, finally settling on Ulysses</a> for fiction where I’ve mostly stayed despite continuing to play around with Obsidian</a>. And then, there’s the editor I kept trying every so often and giving up on: Emacs. If you don’t know Emacs, it’s a legendary text editor, dating back to the 1970s</a>. What makes it legendary, other than being five decades old and still in active development? A lot of editors have some level of extensibility: BBEdit can be scripted in AppleScript, Code and Nova in JavaScript, and all of them can run shell scripts. But Emacs isn’t just scriptable—it’s an application platform, one specifically built around textual data. It’s not a pretty program, but install Markdown Mode</a> and Emacs becomes a better Markdown editor than whatever you’re using now. (Unless you’re using Emacs.) Magit</a> makes it as good a Git client as the $99 Sublime Merge</a>. And there’s whatever the hell Org Mode</a> is, sort of an outliner that’s also a GTD-like system and a literate programming</a> environment and possibly a floor wax and dessert topping. The problem, though, isn’t just that Emacs is spiritually opposed to being a Mac-assed Mac app (it’s actively insufferable about it</a>), it’s that it has a terrible “out of the box” experience. It’s not just the alien keyboard shortcuts, or the eclectic terminology. (A window is not a window, it is a frame, while a split pane is a window; you do not cut, copy and paste, you kill, save, and yank; etc.) The real issue is that it’s a monumental exercise to merely match, let alone surpass, what you get after ten minutes of extension browsing in VS Code or Nova. Even so, I tried to make the switch, every so often. I wanted to like Emacs. But it always defeated me. My experience always ended with some variant of “it’s been three days and I still can’t figure out basic code folding or word wrap the way every other editor does it and don’t tell me I need to learn Lisp to solve this, motherfucker, I have WORK to do.” A few years ago, “starter kits” became more popular, configurations that you install along with Emacs to have it set up right out of the box. They didn’t help. They might have made things worse. I’ll come back to that. So why keep trying? I have tools that work, right? Multiple things, no less. Why try to get into an infamously difficult editor whose past is inextricably tied to an infamous creep</a>? There are a few reasons. The first is the most straightforward: functionality. As much as I love BBEdit, there are things I do fairly routinely—like edit templates in my static website—it’s not as good for as other editors are, and its extension system isn’t powerful enough to let third parties add new ones. Worse, even out of the box, Emacs clobbers BBEdit in terms of its package management UX. That’s worth repeating: Emacs has a better user experience than BBEdit, a native Mac app, when it comes to package management. Nova has great package management UX, and that’s a major reason its extension ecosystem is far bigger than BBEdit’s even though it’s only been around five years. Yet Nova feels like a bit of an also-ran at this point. Visible development has slowed; many of its third-party extensions, including critical ones like JavaScript/TypeScript, have been effectively abandoned. What about Visual Studio Code? At first, it felt like what Emacs might have been like if it had started thirty years later. But the more you dig into it, the more it’s clear it’s open source the way Google products are: more, you know, open-ish. If you’re not using the official build you don’t get access to everything, critical extensions are closed source, and there’s an increasing sense its real purpose is to lock you into an ecosystem. And that leads into a second reason. I don’t think of myself as particularly anti-commercial—hell, I sell books, I have an idea for software that I would, in fact, like to sell—but there is a definite advantage when something you like, even rely on, is developed by a worldwide community of nerds rather than a corporation. Big companies enshittify; small companies go out of business, or close because the owner wants to retire (see Script Debugger</a>, and even my favorite line of shampoo and shaving products, Nancy Boy), or (ahem) become much more successful as a game publisher than a Mac tool developer. The third reason, though, is more abstract. By approaching it with the right mindset, building up Emacs is—mostly—fun. And Emacs itself is “always something more.” The nerdy itch stops being “do I like this new program that does something cool I haven’t seen before” and starts being “can I do this cool thing I haven’t seen before in Emacs.” The truly annoying thing about Emacs, you see, that it really is as powerful as its proselytizers tell you it is. I couldn’t help but notice that Emacs got a fire lit under its neckbeard a few years back. It has modern features like language server and Tree-sitter support, and—sorry to harp on this, BBEdit—actual package management. Now it includes some popular third-party packages like Org Mode, too. So a couple of months ago, I decided to give Emacs one more try. First I tried a very minimal starter kit, Crafted Emacs</a>. For reasons I don’t remember now, this ran aground yet again. But rather than give up, I restarted from a blank slate, working through the book Mastering Emacs</a> (a purchase I made a few years back in a previous failed attempt to migrate) and building a configuration file up one step at a time, not adding any package or setting without understanding what it was there for. I borrowed some bits from Crafted Emacs, tweaking when I wanted to. Improve the completions and search. Turn on the language server. Add a few languages I regularly use. Bring in TextMate-style snippets. And…this time, it’s kind of working. Almost every time I’d reach for Nova—which mostly relates to blog stuff these days, but also some PHP coding—I reach for Emacs instead. There are some things it already does better, and a few things that are a little clunkier (if there’s an Emacs equivalent to Nova’s beautiful task system</a>, I haven’t found it yet), but not much I miss. For BBEdit, it’s harder to say; there are some ways in which it’s the farthest behind the other editors, but other ways in which it’s neck-and-neck with Emacs as a Swiss army knife for text with a better UX. I have years of little tweaks built up, too, that I use on a regular basis. They’re duplicable in Emacs, but it’d take a while. YASnippet</a> is probably best in breed, though, and Emacs’s keyboard macros might let me build up equivalents to BBEdit’s text factories. I’m using Emacs for technical writing now, too, because my current contract job requires me to use a Windows laptop. In a way, I’ve come full circle. I doubt I’ll ever be all Emacs, all the time, even if this sticks (we’ll check back in a year). Ulysses has lovely ways for organizing and exporting long-form text that I’d be hard-pressed to hack up equivalents to, and of course it works on the iPad—a major feature when I go out on day trips with just the iPad in tow. And while I’m starting to use Org Mode for my work tasks, Things</a> will stay with me as my personal task manager for the same go-everywhere reason. I have a few unorganized thoughts to wrap up. First and foremost, I’ve come to believe that starter kits are a bad idea for new users, which is a problem to the extent they’re presented as just what somebody needs when they’re migrating from a more turnkey editor. As a newbie, I have no idea what the hundred or more packages being pulled onto my system do, I don’t know whether a given error being thrown up is important or not, and when something goes wrong—which becomes ever more likely the more dependencies there are—I don’t know how to fix it. And it’s quite possible that even if I find an apparent solution online, it’s a solution that won’t work with that starter kit. I suspect starter kits are more useful for people who’ve gotten their Emacs sea legs under them already, but I’m already not sure whether I’d want to bother switching. Second, while there’s a lot of great Emacs tutorials out there, there’s surprisingly little that steps you through building a modern setup from scratch, particularly if you’re looking for explanations of why you want to embark your corfu with vertico. I’m emphasizing “modern” because the closest I’ve found, the aptly named “Emacs from Scratch</a>,” hasn’t been fully updated since 2020, and a fair number of convenience packages—things that make Emacs feel more modern—now can be replaced by built-in functionality, or newer, slimmer packages that take advantage of new features. Also, Emacs from Scratch is a video tutorial, which is fine, but rarely my thing. I don’t know if I’m the guy to write an “Emacs Configuration for the Moderately Patient” tutorial, but if I actually stick with it, we’ll see. Lastly, I don’t think Emacs will ever be particularly approachable to folks who aren’t willing to hack around with their editors in a more involved way. And that’s okay. I admire the work being put into modernizing it, but the very idea of an extensible, scriptable text editor is, well, pretty nerdy. Either you’re going to get really into the notion, or you’re just, well, not. And even in that realm, Emacs is unabashedly uber-nerdy. MacStories’ Federico Viticci, famed for championing scriptable editors like Obsidian and Editorial, would be all over Emacs if it wasn’t, well, Emacs. I get it. But the thing is, I think I also get Emacs. And once you get Emacs, there’s probably no going back. To support my writing, consider a tip on Ko-fi.com. Maybe the waiting is actually the easiest part 2025-05-12T15:45:00+00:00 I’ve written before how there are two opposing views of large language models and generative AI, with little to no space between them. In one view, they’re already an essential tool and anyone who isn’t wholeheartedly embracing their use in day-to-day work might as well be insisting on keeping their job as a telegraph operator. In the other, they’re a maddening case of the emperor’s new clothes—costs are only going up, the frequency of hallucinations is increasing rather than decreasing</a>, and the sooner we admit that OpenAI is the next Enron, the sooner we can all get on with real work. Opposing viewpoints about new technologies is hardly unheard of, and it’s more difficult than we like to believe to not only spot winners and losers, but even to make generalizations. Most new technologies fail and world-changing technologies are hard to spot aren’t mutually exclusive. Are they both true? Kinda? Maybe? Just because boosters say a technology is world-changing doesn’t mean that it is, but that doesn’t make nay-sayers automatically correct. What’s weird about generative AI, though, is the seeming rise of a third position—one that accepts most of the criticisms of AI while nonetheless taking it as a given that we should be incorporating it into our workflows wherever possible. Call it, perhaps, the “AI triumphalist” position: it doesn’t matter whether the ethical concerns, or even the practical business considerations, are ever addressed—it’s the future, like it or not, baby. A surprising number of people whose opinions I (usually) enjoy reading/listening to seem to have adopted this: they acknowledge all the problems, then go on to use ChatGPT for coding, incorporate it into their shortcuts, and so on. So far none of them have used the image generation tools, as far as I know, but it’s quite possible that’s only because these folks are mostly in the Apple-centric space, and Apple’s “Image Playground” is a dumpster fire even by generative AI standards. I can’t help but wonder if the proof by assertion fallacy</a> is in part at work here. When we keep hearing anything repeated over and over, eventually we can’t help but wonder if we’re missing something vital. Every tech company is racing to AI all the things, from our code editors to our image editors, our search engines, even our keyboards and mice. Surely, surely, there must be something to this all. When we try it ourselves, it’s promising, isn’t it? The responses to our natural language search queries are just what we want, confidently presented, and correct like nine out of ten times. The message summaries only occasionally screw up in laughable and/or horrific ways. The header images we’re creating for our blog articles might not be what was in our head, but they’re interesting and colorful and you have to look closely to see the mistakes, and hey, we wouldn’t have paid a real artist to do this in the first place. That medical diagnosis definitely sounds authoritative, and look, what are the chances those wild mushrooms are that poisonous, really. So when I hear pundits and analysts look at what Apple is (and isn’t) doing in the AI space, I often see it just taken as a given that Apple has to do this. They have to shove AI desperately in all the things, because they’re way behind the curve. The delays in bringing the magical Star Trek-level Siri they pretended they were just about to ship might be existential. The quality isn’t the issue—this is the future, goddammit, and they have to be there yesterday. The AI triumphalist position is the dominant one. And yet, there’s still no indication that hallucinations are a solvable problem. There’s no obvious path to profitability for OpenAI (even their paid plans cost them more money to run than they make), and there’s virtually no “AI industry” without OpenAI. What if the triumphalists are wrong? I’m not (necessarily) suggesting this is a binary, that there will be no value in generative AI, let alone value in other machine learning that, for right now, is unfortunately lumped in with it. Even in the anti-triumphalist case, there will be use cases for large language models. But in this “worst” case, people realize that more often than not, the time they have to spend checking their amazing robot assistant’s work reduces or even eliminates the subjective benefits. And companies quietly start backing away. Google rolls out an “improved” summary for results that applies machine learning in different, non-bullshitting ways. The Copilot brand sticks around, but it gets quietly de-emphasized. Companies whose business model is “repackage somebody else’s LLM” disappear, unless they’ve found a genuine niche. Companies whose business model is monetizing LLMs, like OpenAI and Anthropic, get bought or, in some cases, just fold. (Since LLMs as they exist today are largely fungible, building a moat around such a business is incredibly difficult.) In this scenario, Apple’s mistake won’t prove to be ignoring AI for as long as they did, it will prove to be not continuing to ignore it. Adding a few carefully focused features here and there, like ML-assisted image cleanup and better text prediction, sure. But there are just so many other more pressing issues—from declining software reliability and concerningly poor UX decisions of late to what on God’s green earth they’re doing with the Vision Pro—they should be prioritizing. So could Apple’s foot-dragging end up being accidentally genius? Maybe. They’ve always pushed “machine learning” as their brand, and if it’s in their best interest to stop talking about LLMs, they’ll do it in a heartbeat. While Apple rarely kills popular or even cult hit software and services with the wild abandon that Google does, stuff that never took off quietly gets murdered with no fanfare—Image Playground could easily go the way of Ping and Music Memos. I suspect WWDC is going to be a big tell this year. The rumors that have people buzzing with excitement, dread, or skepticism aren’t rumors about new “Apple Intelligence” features—not even rumors like “they’ll finally ship what they announced at the last WWDC.” I don’t subscribe to the idea that the rumored new huge UI changes across the line are an evil plot to distract us from Apple Intelligence’s failures, but it’s surely not lost on anyone in Cupertino that a year spent talking about the bold decision to change the UI font to Zapfino or make application icons non-Euclidean or whatever is a year not spent talking about Siri. And that might just mean that if an LLM-backed Siri doesn’t ship until mid-2026, it’ll be entering a market which no longer breathlessly expects generative AI to cure cancer, replace all software developers, and animate Oscar-winning movies over a lunch break. That might just be a better market for all concerned—including Apple. To support my writing, consider a tip on Ko-fi.com. The robots have come for your em dashes 2025-04-14T11:11:00+00:00 Did you know em dashes are a sign of AI writing? Since when? Well, that’s what all the LinkedIn influencers say! Humans don’t use them in writing, right? They use hyphens. Of course we use em dashes—they’re basic punctuation marks! We use em dashes as a pause for emphasis—like I did in the first sentence in this paragraph—or to set off a quasi-parenthetical phrase, like I just did in this sentence. A hyphen is used to form compound words, or to break words by syllable at the end of lines. They’re not the same. Nobody better tell the influencers about en dashes; they’ll spontaneously combust. So where’d this idea come from? You got me, but I’d guess it stems from the nature of AI-generated text. It always shows a certain level of polish: no spelling or grammatical mistakes, complete sentences, a clear organization and structure. LLMs get this right in their “first drafts” because the majority of text they were trained on is itself polished—and it turns out that this level of polish covers the correct use of em dashes. Is there a better way to detect AI-generated text than looking for em dashes, then, smart guy? All generative AI is, by literal definition, statistically median, and this isn’t something that can be overcome by creative prompt engineering. Flair and personality in human writing comes from individualistic, quirky choices born of a writer’s likes, dislikes, and life experiences. LLMs don’t have any of those. So LLM-generated text tends to read like pleasantly anodyne business memos, or book reports by a precocious ninth-grader. So if the text you’re playing “robot or not” with is a business memo, a high school paper, or a blog post by the kind of LinkedIn influencer who thinks em dashes are a sign of AI writing, it might be a challenge. In other circumstances, it shouldn’t be. But it apparently is! The problem—and, dear reader, there is no gentle way to put this—a lot of ostensibly smart people neither write particularly well, nor read enough to get a sense of what good writing is like. Most of the writing they deal with—memos, emails, text messages, Amazon and Yelp reviews—is slapdash, unpolished. So if they come across anything in those contexts that looks refined yet relatively generic, they suspect ChatGPT. Apparently, we’ve so come to expect a lack of writing style from humans—and I mean style in the sense of Strunk & White, not in the sense of actual style—that we’re suspicious of it when we see it. If you don’t want your writing to be mistaken for AI, what should you do? Use “fuck” more often. To support my writing, consider a tip on Ko-fi.com. What makes an app feel “right” on the Mac? 2025-03-31T10:45:00+00:00 On Mastodon recently, I wrote: As I play around with more code editors: Zed is pretty neat, although it pushes “AI code assist” to a mildly annoying degree—and despite technically being a native app, it’s no more Mac-like than Neovim is. </blockquote> I got an interesting response from Fred McCann: What do we even consider to be a native Mac app? Cocoa/AppKit? SwiftUI? Catalyst/UIKit? Apple ships apps that don’t feel like Mac apps. I guess I’m already accustomed to dev tools being weird and cross-platform. </blockquote> So I’ve been thinking about that. The UI toolkits are obvious answers, but I’m not sure that they’re completely satisfying. For instance, I love MarsEdit</a> in theory—enough that I bought version 4 and the upgrade to version 5—but in practice, it’s never really stuck for me. Granted, in no small part that’s because I keep doing things like redesign my website using Zola, a static site generator that MarsEdit doesn’t work with. But even back when I was using Tumblr and, later, Micro.blog, it turned out that being a Mac-assed Mac app wasn’t enough for me—the aesthetics mattered. I could never make MarsEdit look as good to my eyes, be as pleasing a long-form writing environment, as I could iA Writer, Ulysses, or my old stalwart, BBEdit. What’s startling to me, though, is that another writing environment I’ve come to really like is the one I’m using now: Obsidian. Which is not, in any sense, native. So it’s possible that the right question—at least for me—isn’t “is this app using a native UI toolkit,” it’s “is this app a good Mac citizen.” In other words, does it embrace long-standing Mac conventions? The standard menus (Application Name, File, Edit, View, Window, Help) should be present, in that order, with the expected submenu items.</li> Settings should come up when you select Settings… or hit ⌘,. Settings should appear in a window, not a tab (and definitely not a text file that opens in a tab).</li> Everything in the Services menu should work as expected, including when it’s invoked from the context menu.</li> Any text field should work the way it does everywhere else in the system. It should respect the subset of Emacs key bindings that all Cocoa text fields do, it should respect new key bindings you’ve put in the system key bindings folder, text replacements you’ve defined system-wide should work, smart quotes should be made smart if you have that turned on globally (and left stupid if you do not), and so on. Text editors that have their own key binding system get a partial pass here.</li> The general “fit and finish” needs to be Mac-like. Icons and symbols, even if they don’t come from the SF Symbols</a> library, shouldn’t look like they came from Windows, GNOME, Mac OS 8, or Mars; the same should be true of UI widgets and form elements. The UI font should be San Francisco, or at the very least something similar.</li> </ul> If we hold things to this list, programs like Nova, MarsEdit, and Apple Pages—canonical Mac-assed Mac apps—all do unsurprisingly smashingly. But Obsidian, the Electron-based program I’m writing in right now, does shockingly well, too. Microsoft’s Visual Studio Code doesn’t do quite as well (most notably, it opens its settings “window” as an editor tab), but it does better than Sublime Text (which opens a text file for settings), and much better than the banana crazypants menu and icon design of the cross-platform e-book management program Calibre. On the flip side, being 100% native is no guarantee. The long-running story plotting app Dramatica Story Expert</a> was native, yet such a poor Mac citizen that it not only violated several of those bullet points for decades, its fonts somehow, incredibly, never rendered at retina resolution. And modern apps written using Apple’s own UIKit-based Catalyst</a> frequently break the Services menu or text fields or both. Also, there’s the insanity of the new Settings app on the Mac, which replaces checkboxes with toggle switches, hides functionality behind the circled-“i” icon that traditionally means “information”, and groups preferences less by well-defined category than by, I don’t know, astrological signs or something. Don’t worry, though! The search function is (checks notes) absolute bollocks. What about web apps, you might ask? I’ve been thinking about Federico Viticci’s recent post about the iPad’s “Sweet Solution”</a> with respect to “feel”: In working with my iPad Pro over the past few months, I’ve realized something that might have seemed absurd just a few years ago: some of the best apps I’m using – the ones with truly desktop-class layouts and experiences – aren’t native iPad apps. They’re web apps. </blockquote> This frustrates me because he makes a good case in that article—and for me, at least, that’s a problem. While it’s possible to make apps that are relatively good Mac and iOS/iPadOS citizens using web technologies, that’s not the same thing as making a web app, an application that runs entirely in the context of a web page. I’m not against web apps—Google Maps and Google Mail are both canonical examples of good ones (setting aside for the moment one’s feelings about Google as a political/corporate entity). But at least for me, document-based web apps are usually a bridge too far. Unless multi-user collaboration is essential, give me Apple Pages and Numbers over Google Docs and Sheets any day and twice on Tuesday. So, I’m not sure where this ultimately leaves me. I’m likely to keep using Obsidian and keep failing to use MarsEdit. I’m likely to keep rooting for Panic Nova over Visual Studio Code (I’m using Nova to update the website, and run preview and deployment tasks). Yet I’d take Code over Sublime Text</a>—although I might just take a well-configured NeoVim</a> over either. It’s not like it looks any more out of place on the Mac, and hey, you can’t kvetch about it having incomplete Vim emulation. To support my writing, consider a tip on Ko-fi.com. Having to think about essential computerness 2025-03-24T11:35:00+00:00 Years ago, a friend of mine who also uses Macs told me a story about why another mutual Mac-using artist friend chose the Mac over the PC some three decades ago. His technophile friends—of which he had many; this was Silicon Valley in the 1990s—often showed him what they were doing with their computers. The Mac users said, See what I did with my computer? But the PC users said, See what I figured out how to get my computer to do? And so he got a Mac. This has always been crucial subtext not for Mac lovers as much as for Mac haters. The specifics change over the years, but even the nerdiest Mac fans tend to appreciate that, by and large, they don’t have to think about the essential “computerness” of their Macs unless they choose to, while even the most casual PC fans—not users, but power users—tend to be inherently suspicious of any computer that doesn’t make its users think about that. See, every PC fan knows at least one less tech-savvy PC user in their family or their workplace whose system is regularly FUBARed because they don’t understand things that are, to the power user, very basic shit. And then there comes the goddamn Mac, saying, “Hey, relax, you shouldn’t need to understand that. Look, I didn’t even have a command line for the first fifteen years of my existence! Have some kombucha. Kick back.” And it drives them absolutely bananas. Okay. I suspect the power users I know (and don’t know) are already seething at me, but look. I ran Windows 2000 on my home PC and used Windows in various flavors at work. I installed Linux back when getting a GUI running on it required you to find the dot clock values your video card supported to enter them by hand in the XF86Config</code> file. I’ve edited binary files by hand in hexadecimal on a TRS-80. What I’m saying is, I have walked the walk. And, man, after OS X came on the scene, the most amazing thing happened: now Macs were Unix workstations—and still Macs. To this day, I still don’t think I can get PC users to really understand how flexible and extensible the Mac became under the hood once it became the successor to NextStep. (I have found that PC power users tend to judge customization on “can I reskin the windows” and Mac power users tend to judge it on “can I automate it”; I’m pretty firmly in the latter camp.) I can do everything on the Mac that I could back when I was running Linux and FreeBSD. But it’s still a Mac. This isn’t about being a walled garden—the Mac isn’t one, and I’m tired of uninformed or cynical arguments to the contrary—it’s about not finding myself trying to debug mysterious problems after the most recent major system update, or just because it’s been a year or two since I did a complete reinstall of the system. “Things will always eventually get weird and slow, and you should do a clean reinstall” has been advice for going on forty years in the Windows world. But a disquieting thing has been happening over the last few years—some might say as long as the last decade, but I think I’d put it right around macOS “Big Sur” in 2020. That’s when the OS version number finally moved from 10.x to 11, the UI was redesigned in small, subtle ways (not all for the better), and Mac hardware started transitioning from Intel CPUs to Apple Silicon ARM-based CPUs. More and more often, I have to think about my Mac’s essential computerness now. Because sometimes things that were working just stop. Or don’t work right at all. For the first year and a half, the Studio Display itself occasionally had to be rebooted—the monitor had to be rebooted, because it runs goddamn iOS. For the last few months, I’ve been getting weird glitches with USB audio on occasion, including my external USB speakers just not working until they’re disconnected and reconnected. Lately, I’m frequently chided that I should have remembered to eject the backup drive from my laptop before disconnecting it, but there is no backup drive: it’s backing up to another Mac over the LAN. Most recently, in Big Sur, some of Apple’s new fancy screensavers and dynamic wallpapers just stop being dynamic until you kill their process to restart them. And, like I said, I can walk the walk. I created a little Shortcut called “Kick wallpaper” that lets me restart that process easy-peasy. I was proud of that for the thirty seconds it took me to remember that I wouldn’t have to have written that little Shortcut if it wasn’t for Apple’s bug, one that they haven’t fixed in any Big Sur update. What’s worse, sometimes the system just gets weird and slow. An application doesn’t launch at all, or some background task that should be running clearly isn’t. Desktop widgets stop updating, and since widgets have been the first thing in decades that’s got me finally embracing a task manager—that’s a future post—that widget better be current, or my task system falls apart. Hazel starts reporting weird errors, and after debugging, I realize Spotlight indexing has stopped. Why? Who knows. Will rebuilding the index fix the problem? Maybe. Would anyone but a power user even be able to debug this? Doubtful. Maybe they could just log out and log back in again, but I’m not sure most Windows or Mac users would think of that. They’d just reboot and hope. Or do a clean reinstall. I’m not going to pretend I’m going to switch to Linux any time soon, let alone Windows. There’s too much I’d miss. While I could find rough equivalents to Mac-only applications like BBEdit, Nova, Acorn, Ulysses, and Marked, there’s a coherence that I just don’t see anywhere else, still, after all these years. The Mac is, for me, still a better platform. But I can’t help but wonder if modern Apple’s obvious inattention to software quality—there’s no kinder way to put it—is letting an important differentiator, potentially the most important one, slowly vanish. Now, once in a while, I sit down in front of my Mac and realize I’m not thinking what can I do with my computer. I’m thinking what can I figure out how to get my computer to do. To support my writing, consider a tip on Ko-fi.com. Just slap an LLM on the front end—how hard can it be? 2025-03-17T11:31:00+00:00 My last post</a> mentioned one of my favorite new-to-me bloggers of late: Ed Zitron</a>, who’s staked out a position for himself as the AI skeptic’s AI skeptic. His recent hits include “The Generative AI Con</a>”, “Open AI is a Bad Business</a>”, and “Sam Altman is Full of Shit</a>”. Even without clicking any of those links, you’ve probably picked up what Ed is laying down. Zitron certainly isn’t alone in his pessimism, but he’s an outlier among most of the tech pundits and podcasters I follow and listen to. Even as everyone seems to recognize that what Microsoft and Apple and Google are furiously cramming into every available inch of UI real estate isn’t particularly good, there’s an omnipresent vibe of “yeah, but it will get better” and “they’re still working out the kinks”. And, sure, in the general case, that’s the story of all successful technologies. Lots of them start out kind of sucky and get better over time—sometimes so much better, you couldn’t have reasonably predicted how good they’d get from where they began. This is Amara’s Law</a>: We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run. </blockquote> Now let me introduce you to Watts’s Corollary to Amara’s Law: If the technology is fundamentally incapable of reliably doing what you’re promising, it doesn’t matter how long the damn run is. </blockquote> Oh, God, is he talking about “hallucination” again? Sorry, yes. Yes, I am. To quote from one of Zitron’s earlier pieces, “Godot Isn’t Making It</a>”: Separate to any lack of a core value proposition, training data drought, or unsustainable economics, generative AI is a dead end due to the limitations of probabilistic models that hallucinate, where they authoritatively state things that aren’t true. The hallucination problem is one that is nowhere closer to being solved—and, at least with the current technology—may never go away. </blockquote> As I explained last year</a>, LLMs “reason” by performing lexical analysis and interpolation between words represented as points in multidimensional vector space, and deriving how words commonly relate to one another is not the same as deriving what words mean. LLMs are amazing at creating technically original, syntactically correct sentences, but a good text adventure game from the 1980s understands locks, containers, light sources, being open or closed, and so on at a symbolic level that the most powerful OpenAI model does not. The current trend of “reasoning models” that essentially ask LLMs to check their own work doesn’t address this concern, because symbolic logic is fundamentally different. You can’t get there just by LLM-ing harder and faster. But here’s an idea: use LLMs not for generating final output, but as front ends to other systems. LLMs are arguably the best technology we have now for natural language processing (NLP), and it’s not just an incremental improvement over what came before, it’s a quantum leap. I can easily imagine things I’d put this to use for right now. Natural language queries that searched my own projects (“what color are Autumn’s eyes”) are obvious, but it wouldn’t be a stretch to imagine this being a front-end for automating actions that touch existing pieces in Apple’s ecosystem. And, behold! The most interesting parts of “Apple Intelligence” promised last year are exactly that, on steroids. Imagine the ability to say “give me directions to the restaurant I’m meeting Agatha at for lunch and text her I’m on my way” and have Siri understand that it needs to find Agatha’s contact information, search recent text messages and email messages from her until it finds one that mentions a restaurant and a lunch meeting, feed that restaurant into Apple Maps for driving directions, and send Agatha a text message. Amazing! Well, keep imagining it, bucko, because you can’t do it yet, and you may not be able to do it for a long, long time. Apple recently announced that “personalized Siri” won’t be ready “until the coming year,” i.e., 2026. John Gruber of Daring Fireball is genuinely angry about missing the likelihood of this</a>: What Apple showed regarding the upcoming “personalized Siri” at WWDC was not a demo. It was a concept video. Concept videos are bullshit, and a sign of a company in disarray, if not crisis. </blockquote> He’s right, and I had to think for a moment about why I wasn’t particularly angry: because I wasn’t particularly surprised. (By the delay. Apple being in disarray—which I also agree with—is another post.) “LLM as front-end to other systems”, it turns out, is way more work than “LLM does everything”. In the front-end model, the work gets done by the non-LLM parts of the system. So far, so good, at least in theory. But not only are hallucinations still a problem in this scenario, they’re now a compounding problem. When the LLM takes user input and spits out, say, JSON API calls on the back end, nothing stops it from hallucinating incorrect parameter values, or even imaginary parameters. In certain circumstances, this may be fine. But in the examples Apple promised—ones that take the user’s input, break it down into multiple steps, possibly generate hidden intermediary prompts, and end in actions like “give me driving directions to where you’ve deduced I mean”—all of the steps have to be correctly inferred. Also, any auto-generated internal prompt offers another chance for a hallucination. If any one of those steps produces a bad answer, the whole chain will go off the rails dramatically—or worse, go off the rails subtly. Being less wrong is not good enough; the end result has to be correct, which means every step does, too. So, we don’t have that yet. The parts of Apple Intelligence we have are the parts that create anodyne, occasionally misleading summaries of an email thread, rewrite a business letter so it has all the personality of cafeteria mashed potatoes, and synthesize images of eldritch horror in the style of a Memoji. You remember the old Steve Jobs mantra of over-promise and under-deliver, don’t you? Wait, is that how it goes? Let’s ask Siri. “But Amazon’s recently announced Alexa Plus</a> can do all this!” No, it can’t. They were explicit that it won’t be able to do everything it promised when it starts shipping, either. Amazon’s Head of Devices and Services says, “The engineering team had to hook [Alexa’s] API to the LLM and make sure it can do it all without hallucinating”, and I’m going to make two predictions here: one, they haven’t actually gotten that to work yet; two, they’re going to discover that, no matter how hard they think it’s going to be to make it absolutely completely without a doubt work, they are still underestimating the problem. And while Google’s “Gemini with Personalization” is being touted by some as Google beating Apple to the punch, it can’t do what Personalized Siri promises, either (it can utilize your Google search history “to provide relevant and personalized responses”, but right now, that’s it). Here’s the question we’re dancing around: what if it’s impossible to get LLMs to do this reliably? So far, Ed “Pissing on Sam Altman’s Parade” Zitron’s critiques have mostly been right. There are specific, constrained use cases that I know people insist generative AI helps them with, and even when I have my doubts, I’m not going to insist they’re wrong. But all uses of LLMs are so piggishly resource-intensive that OpenAI, Anthropic, and friends are poster children for the “we lose money on everything we sell, but we’ll make it up in volume” joke. All major LLM training corpuses incorporate copyrighted material that isn’t licensed for free commercial reuse, and everyone’s just gambling that courts won’t tell them they have to pay up. Yet those issues, we can at least pay lip service to addressing. The issue of what if you just can’t get there from here, the possibility that the entire computing industry is frantically shoveling money into a technology that might be less “next Internet” than “next Minidisc”, is just too much to even contemplate. So we’re not. I know the common wisdom is that the functionality Apple described with Super Siri will inevitably be on not just Apple’s platforms but Android and Windows, too; I’m not so sanguine. Maybe there’s a workable solution involving giving the LLMs as little to do as possible—i.e., letting them handle the NLP and only the NLP, passing the rest off to a fully deterministic “AI” like Bixby’s planner</a>. (You laugh, but parts of Bixby—mostly, the parts that were Viv</a>—were way ahead of their time.) But I’m not positive even that will do it. Hallucinations will still happen. Errors will still accumulate in multi-step actions. And when Super Siri sends you to meet Agatha at the Red Robin in San Bruno instead of Birdsong in San Francisco, you will have a lot of time to contemplate the difference between “less wrong” and “correct” as you sit, sad and alone, with your barbecue bacon cheeseburger. To support my writing, consider a tip on Ko-fi.com. The Reluctant Technologist 2025-03-10T11:15:00+00:00 Back in the early, early days of the web, Slate ran a column called “The Reluctant Capitalist.” The columnist, Heather Chaplin, grew up in a leftist, deeply anti-capitalist household; she absorbed those values, but was nonetheless fascinated by the machinations of capitalism. The column resonated with me—while I grew up in a more conventionally liberal family and didn’t think, at the time, I was particularly anti-capitalist, I found myself agreeing with both her criticisms and her fascinations. This has been on my mind while I’ve been thinking of “soft rebooting” my blog lately, by which I mean getting back to writing for it regularly. That inevitably raises the question: why haven’t I been writing for it regularly? While it’s not obvious if you look at the archive page on Coyote Tracks Dot Org here, I wrote several things a week for several years on the original Tumblr-hosted version of the blog, and at one point was a featured technology blog on Tumblr, had a hundred thousand followers, got cited occasionally on Techmeme, the whole shebang. There is another universe in which I figured out how to actually make Coyote Tracks into a living. What happened in this universe, though, is that Coyote Tracks led to my career shift from web development to technical writing, right at a moment that coincided with me re-focusing on fiction writing by attending a two-week residential workshop at the University of Kansas’s Center for the Study of Science Fiction. (Due to university politics, the CSSF is a shell of its former self, but the Ad Astra Institute</a>, led by the Center’s original directors, carries on the tradition.) So, it would be easy to say that I stopped blogging regularly because I was writing so much else, in no small part thanks to the blog. Easy to say, but not entirely accurate. In recent years, I’ve come to realize that what I saw as the “good bits” of capitalism are real, but they’re being drowned out by the bad bits. They’re properties of well-regulated, properly functioning markets, and market socialism</a> is a thing. Capitalism doesn’t, at the end of the day, want properly functioning markets, because corporations don’t want prospective customers to be able to make informed choices or, ideally, to have choices at all. So the bad bits of capitalism are (checks notes) capitalism. In parallel, I’ve been finding that a lot of what I saw as the good bits of technology are real, but they’re being drowned out by the bad bits. The bad bits of technology are (checks notes again) also capitalism. In July 2024, I wrote about feeling burnout</a> on my job hunt. Part of that comes from how bad job hunting has become in the tech market the last few years, but it’s also tech itself. It’s the sense of desperation in pushing whatever the next big thing is—currently, LLMs—as the most revolutionary invention since the steam engine. As Ed Zitron memorably puts it, it’s the Rot Economy</a>. If the sole meaningful measure we apply to companies is growth, all companies, no matter now interesting or artsy or humane or quirky they may start out as, are inevitably driven to enshittify</a> their products. Maintaining your existing market is not enough, because that means you’ve stopped growing. You must, must, extract more revenue from your existing customers and find new customers, and if you can’t find new customers, you have to get into new markets. And, you know, I love Apple TV+ as a service, even if the naming conventions lead to Who’s-on-first comedy routines. (“How do I watch Severance?” “It’s an Apple TV+ show, so open the Apple TV app on your Apple TV. Now, you can watch other shows in the Apple TV app on your Apple TV, but those aren’t Apple TV+ shows, so don’t get confused. Also, make sure your Apple TV is set on the correct Apple ID</del> Account or you won’t see the correct Apple TV+ shows in the Apple TV app.” “Maybe I’ll just see what’s on Netflix.”) New products and services in unexpected areas aren’t intrinsically bad. Hell, nobody expected Apple to make a great phone, and it’s no exaggeration to say their designs changed the entire smartphone field. We can debate whether Apple also makes the best tablet computers and the best smartwatches, but we can’t debate they’re sure as hell in the running for both. But we don’t need “Apple Intelligence” everywhere, especially in its current, degraded form. (The flaws people assume will get fixed later are very likely intrinsic to large language models, but that’s a different post.) We especially don’t need the System Settings panel to show us an advertisement for Image Playground that we literally cannot dismiss without opening Image Playground. That is not a dark pattern, it is a Vantablack pattern, and it neatly encapsulates everything that I hate about what technology has become. That it’s coming from Apple, a company whose products I’ve mostly loved for four decades, is fart icing on a shit sundae. Sorry, I’m getting off track. The point here is, technology in its current state pisses me off. A lot. Web development—software development in general—has become absurdly complicated. Selling software is more difficult, with more gatekeepers, more corporations to placate. Our hardware and operating systems are ever more difficult to customize. Websites are harder to read, with eye-searingly intrusive ads and whiny requests to sign up for mailing lists. Half the time they’re full of reposted, AI-filtered slop when you can read them, because you can no longer rely on search engines. Everything from your phone to your television wants to track you. It’s enough to make you want to sell everything electronic and move to a cabin in the mountains (assuming you can find a place that won’t be routinely subject to floods and/or wildfires). And yet, in spite of it all, I still kind of love technology. I still love finding new—or merely new-to-me—apps. (I’m trying to use Obsidian instead of Ulysses for the blog, starting with this post!) I love it when a program or feature I didn’t see the use for suddenly “clicks.” (I could never make task management apps work for me, until I used Things 3 with macOS’s desktop widgets.) And the Mac, for all the steadily increasing number of things there are to complain about, remains a joy to use in a way that no other system I’ve routinely used—and I’ve used a lot—ever matched. (The year I used BeOS full-time came close, but again, that’s a different post.) So, I’m a reluctant technologist. I don’t like the industry much at all, and the things I dislike about it are the things that make its products worse. Yet I’m still fascinated by gadgets, by software, by big swings and big misses, by hidden gems. And, yes, by the so often self-destructive business practices. I know I’m not alone in that love/hate feeling, and I’d like to explore the why of that more, dig into it. While there’s too many pundits on both the pro- and anti-AI beats, as someone who worked in the “intelligent assistant” industry for a while, I have love/hate feelings on that front, too. And, of course, one of the big weights lately on the “hate” side of the equation is the politics—the way Silicon Valley’s ill-considered technolibertarianism has curdled into an altogether darker technofascism. While I never moved in those circles, I got close enough to peek in on occasion. As much as I don’t want to dwell on topics that make me want to walk into the Gulf of Still Fucking Mexico, they’re important to face. What I don’t think I’m going to try to do is keep up with current tech news the way I did fifteen years ago. That doesn’t mean I won’t talk about it, but I think I’d rather embrace Last Week Tonight energy rather than Daily Show energy. It’s more my speed these days. To support my writing, consider a tip on Ko-fi.com. The pressure to stay genteel 2025-01-24T12:30:00+00:00 For years, one of the highlight shows of Anthrocon, a furry convention in Pittsburgh, was a comedy act put on by an aspiring professional comedian who was part of that community. He was good friends with the con’s chairman—a guy who’s a pretty good comedic storyteller in his own right, and who did his own show at the con. They often shared the stage together. Now, that aspiring comedian always played with a kind of cynical “it’s funny because I’m a bastard” sort of persona, and over the years—this was the 2010s—this metastasized into a genuine meanness. He was still often funny, but he was ever more frequently punching down. Jokes about transgender people. Jokes about depressed and suicidal people. Attacking gays who didn’t fit his conception of how queers should behave in public. (The comedian was gay, but didn’t like pride parades due to the “out there” behavior of some participants: e.g., don’t make the straights uncomfortable.) A growing chorus of regular con attendees wanted them to stop inviting this guy to put on shows, arguing that supporting him in that way—in this case, literally giving him a stage—was tantamount to supporting his noxious views. And the con chairman really, really resisted this. Wouldn’t giving into the mob be even more divisive? Wasn’t there a better way to handle this? He took a lot of heat for it, faced accusations that defending the comedian made him just as bad. But the chairman knew that guy. He was friends with that guy. And that guy just couldn’t be that bad! The picture of the comedian painted by his critics had to be, just had to be, an exaggeration, because dammit, he wouldn’t be able to be friends with someone like that. So I understood where the chairman was coming from: our instinct is to defend our friends. There was a little explosion, of sorts, in another community I’m in over the last week. Adam Newbold, the developer of the delightfully weird omg.lol</a> service, called out developer Vincent Ritter, a contractor who works with Micro.blog. Micro.blog is “a unique platform that combines blog hosting with a social media-style timeline and community,” as their about page puts it—a spiritual competitor to X/Twitter started by Manton Reece, a developer who left Twitter years ago due to his dissatisfaction with the site’s direction. I’d argue Micro.blog is a cornerstone of the IndieWeb</a>, an approach to decentralizing internet publishing using existing tools like RSS and webmentions</a>. Newbold deleted both his Mastodon callout post and his 2024 blog post after receiving a legal threat</a>, so I have removed those links, and am instead linking to a random Wikipedia article</a>.</aside> What set Newbold off was discovering that Ritter is not only still an active X/Twitter user, not only pays for a blue checkmark, but posted cheerful congratulations to Musk for SpaceX’s recent successes, adding that he sure hopes he gets to buy a Tesla someday in the future and that he’s “happy to be living through this age.” Newbold continued by linking to a post he wrote in mid-2024 about another one of Ritter’s social media posts, where Ritter wrote, When a game asks for “pronouns” and doesn’t let you skip that part. Uninstall. Sigh. </blockquote> As Newbold wrote in that mid-2024 post, There’s only one reason to write that, and let’s be real: it’s not about the pronouns themselves. When someone puts the word pronouns in disgust quotes, and when they complain about being asked to provide theirs, it’s never about the pronouns. </blockquote> And, boy, did this set off a firestorm. Reece effectively leapt to Ritter’s defense, Newbold fired back with, in so many words, “how can you keep defending this guy,” and it got worse from there. People accused Reece of gaslighting. People accused Newbold of being super-aggressive and confrontational. Daniel Jalkut, Reece’s friend (and co-host with Reece of the Core Intuition podcast), accused everyone of failing to see the nuance in it all, and people accused him of trolling. I want to make my position clear here: while I think there’s some value to all those takes, there’s not equal weight to them. Newbold’s take on Ritter is fundamentally correct, for the simple reason that everything he said is true. Saying that a game merely asking you to select pronouns for your avatar is reason to uninstall it is transphobic. Full stop. If someone wrote “When a game asks for ‘race’ and doesn’t let you skip that part. Uninstall. Sigh,” we would not be having a discussion about whether that was racist. Unless, perhaps, we were having that discussion on X—which is why calling out Ritter for paying for his blue checkmark is also appropriate. X has become undeniably right-wing in the past two years, to the point where it has helped Elon Musk drive our country toward actual, non-metaphorical fascism. And again, that point’s not seriously debatable. Aspiring to buy a Tesla now, in the year of our lord 2025, is aspiring to drive around in a rolling MAGA hat. Now, it is arguably true that Newbold was aggressive and confrontational about this all, and no, it wasn’t a conversation with a great deal of nuance. But I don’t know that being confrontational was that inappropriate, or that there really is a lot of nuance to the situation. However, I don’t have the same feeling of ill will toward Reece as it appears Newbold does, for the same reason that I didn’t have ill will toward the chairman of that convention. The picture of Ritter painted by his critics had to be, just had to be, an exaggeration, because dammit, he wouldn’t be able to be friends with someone like that. And our instinct is to defend our friends. I love many things about the Micro.blog community (I backed them on Kickstarter and had a Micro.blog account for years; I don’t anymore, but that’s more for financial reasons than anything else). It’s hard not to notice, though, that there’s a distinct vibe of “no politics</a>” there. While I have no statistics to back this up, my subjective observations were that the user base is proportionately whiter, older, and more well off than old Twitter was and that Bluesky and Mastodon are now. And it’s decidedly more genteel. Shitposting, even meme-sharing, is all but absent; so are strident political expressions of virtually any stripe. It doesn’t have Twitter-style reply guys as much as tut-tutters. When participants in the spiraling “is Micro.blog secretly horrible” threads express shock and dismay at one another, that centrist “can’t we all get along” culture may be responsible for a chunk of it: one side can’t believe the other side is saying such impolitic things, while the other side grows ever more frustrated that important questions about inclusivity and safety are being shunted aside to argue about phrasing. I’ve read some discussion on Micro.blog itself about this, and it’s all pretty civil—and all mostly about, as one person put it, “distaste for the way this was executed.” Tut-tut! Having a genteel social media site is, in many ways, nice. But it can also provide cover for people who are, if not openly advocating for those harms, approving of them. Ritter seems genuinely apologetic</a> about the firestorm he precipitated, but as (mid-2025 update: former) Micro.blog user Hollie notes</a>, there’s an air of I’m sorry that I made you all so upset to his response that lacks both a clear explanation (for instance, “I see now how that earlier post was transphobic, and I understand why you are all upset that I am paying eight bucks a month to a Nazi so I have a better experience on a platform that the service I develop for was explicitly founded to counter”) and amends (for instance, “I will stop sending eight bucks a month to the Nazi and drop the Transgender Law Center fifty bucks instead”). So what’s the solution to this? I don’t know, man. Like most social media explosions, it’ll die down, and leave everyone with vaguely bad feelings. Some people will come away thinking that Adam Newbold is taking a principled stand (yes), and/or that he’s too confrontational about it (eh?); that Manton Reece is a transphobe (doubtful), and/or that he veers centrist in ways which undercut his allyship (hmm</a>); that Vincent Ritter is MAGA-aligned (signs point to yes). And, I get that this particular instance has a tempest in a teapot aspect to it, that—like the comedian at the convention—it’s become what it has due to existing relationships rather than the world being on fire. The thing is, the world is on fire now, and I can’t help but worry how much of this is a harbinger of the next four years—or, God help us, decade or longer. There’s going to be pressure to not take strong stands. To shift uncomfortably and stay silent when someone says, “I support the transgenders, you know, but don’t you think they’re asking too much?” instead of saying, “No, I don’t, and ‘transgender’ is not a noun, you absolute fucking tomato.” To defend friends and coworkers when they’re the ones Just Asking Questions™. There’s going to be, in other words, pressure to be genteel. That means not calling people out when they deserve to be called out. It means looking the other way. Maybe there are times in history where that’s fine, or at least not harmful, but we’re not in such a time. We’re in a time where staying above the fray, where tut-tutting people sounding the alarm for being so uncouth about it, is choosing a side. You’d better be damn sure it’s the side you want to be on. To support my writing, consider a tip on Ko-fi.com. Platforms are not neutral 2024-11-26T12:30:00+00:00 I’ve been thinking a bit about two pieces that talk about Substack, the massively popular and occasionally controversial newsletter platform: Anil Dash’s “Don’t Call it a Substack</a>”</li> John Gruber’s “Regarding—and, Well, Against—Substack</a>”</li> </ul> They both come to the same conclusion, essentially: that you shouldn’t put your writing on Substack. As Dash writes, The reason Substack wants you to call your creative work by their brand name is because they control your audience and distribution, and they want to own your content and voice, too. </blockquote> Or, as Gruber puts it more simply: “The company’s long-term goal is clearly platform lock-in”. Yet, they don’t make the same arguments. Dash’s is more about taking a philosophical and political stand: Substack is, just as a reminder, a political project made by extremists with a goal of normalizing a radical, hateful agenda by co-opting well-intentioned creators’ work in service of cross-promoting attacks on the vulnerable. […] The counter-argument people generally have is convenience…and the theoretical benefits of network effect from being on Substack. Which is largely a myth (most referrals are thanks to other writers, not the platform) and means you have to be open to the platform using your writing to introduce people to the most insidious anti-trans and white supremacist rhetoric on the internet. </blockquote> Gruber largely dismisses that argument, but dislikes Substack’s homogenizing aesthetics: What I object to isn’t their laissez-faire approach to who they allow to publish on their platform, but rather how they present all publications. […] Substack publications do all look the same, most of them right down to that telltale serif typeface, Spectral, which is kerned so loosely it looks like teeth in need of orthodontia. </blockquote> I fully agree with that (down to the kerning complaint), but I’m not so quick to elide the political argument. Gruber sees it this way: I think Substack sees itself as a publishing tool and platform. They’re not here to promote any particular side. It makes no more sense for them to refuse to publish someone for being too right-wing than it would for WordPress or Medium or, say, GitHub or YouTube. Substack, I think, sees itself like that. </blockquote> Substack certainly claims to see itself like that, and they may be absolutely sincere about it. Yet virtually all platforms make that claim. Elon Musk and his supporters maintain that Twitter’s transmogrification into X has made it far more like that than it had been. And while that’s transparently mendacious bullshit, I think it raises an important point: publishing platforms can have editorial viewpoints, not just publications. It’s still possible to find a wide range of political and cultural viewpoints being expressed on X/Twitter, but you can’t make a credible argument that all viewpoints are treated neutrally on the system. Editorial decisions are made across nearly all platforms that allow user-created content, all the time. You see it in what language and which subjects are more frequently moderated and the severity of the moderation; you see it in what’s being promoted by the platform itself, whether algorithmically or human-curated. Bluesky’s moderation team exhibits little sympathy for blatant transphobia, while X’s moderation team is on board with the idea that “cis” is a slur. Creators booted off Patreon for content policy violations often find SubscribeStar welcoming, and darned if a disproportionate number of the most successful creators at SS aren’t visibly far right. I don’t know if it’s fair to call Substack an extremist political project, but Jude Ellison Sady Doyle brought receipts</a> about far-right writers who aren’t just hosted on Substack but were actively recruited (and given large advances, to boot) by the company. And as Annalee Newitz wrote</a>, while Substack claims that their paid writers came from across the political spectrum, they don’t offer any proof of that beyond “trust us,” and as far as I can tell the only people who’ve spoken up about being paid are precisely the contrarian conservatives Doyle was writing about. On a technical level, one can argue with Doyle’s assertion that Substack isn’t a self-publishing platform: anybody can sign up and start publishing, after all. But that doesn’t mean Substack isn’t doing curation. They choose creators to actively recruit to their platform and to give advances to. They decide who is—and isn’t—promoted on their front page. And, of course, they determine who is—and isn’t—in violation of their terms of service. Is there a bias in who Substack actively promotes? You couldn’t make a slam-dunk argument for that based on their front page as of today (November 26), where out of eight “featured” publications, four are arguably political, with two (Kareem Abdul-Jabbar and Mehdi Hasan’s Zeteo) on the left and two (Bari Weiss’s Free Press and Andrew Sullivan’s Weekly Dish) on the right. Of course, we know that Sullivan and Weiss were actively courted, and as far as I can tell, Abdul-Jabbar and Hasan were not. Yet, the terms of service violations—and lack thereof—are revealing. Despite what right-wingers like to believe, nobody gets banned from major platforms for being “too right-wing”. They get banned for hateful rhetoric, harassment, and threats of violence, often after repeated warnings. From all appearances, Substack lets a lot more of that slide than other platforms do. Maybe there’s a whole coterie of active, popular Tankie</a> Substacks being given the same pass that literal Nazi Substacks</a> have been, but if so, nobody’s reported on them. An objection that could be raised: if a platform successfully insulates you from things you don’t want to see, does it matter? Gruber again: For the last few weeks I’ve tried using [Substack’s] iOS app, and I’ve never once seen a whiff of anything even vaguely right-wing, let alone hateful. Not a whiff. If it’s there, I never see it. If I never see it, I don’t care. </blockquote> I don’t have the iOS app, and when I visited the Substack website, I wasn’t logged in. It’s quite possible that Substack is good at keeping the Free Press out of your recommendations if you’re subscribed to Heather Cox Richardson. Maybe, if you consider yourself on the left but more left of center than activist progressive, Substack is smart enough to show you content that’s just slightly right of center. It doesn’t show you the Free Press headline about the Kindergarten Intifada I swear I am not making up, but it shows you the interview with Seth Moulton about how the Democrats are “too focused on trans issues.” You like that? That seems reasonable? How about how a few of RFK Jr.’s policies are similar to a few policies you can find around Europe? Huh! That’s not right-wing, is it? We’re just asking questions, right? That’s where Dash’s argument rings more true to me. It’s extremely unlikely you’re going to see Substack push a link to a white supremacist article under Matt Yglesias’s newsletter. But how about a “gender critical” one? I don’t think I’m being unfair when I describe Yglesias as an increasingly centrist pundit, and YouTube amply demonstrates how easy it is for an algorithm to hopscotch you from centrist material to right-of-center to far right to holy shit. And YouTube, for all its many problems, doesn’t have an ideological thumb on the scale. Maybe Substack doesn’t, either. But in our present age, it’s hard not to notice that the platforms and publications that trumpet their support for free speech the most fervently all too often have very specific kinds of speech in mind that they want to “protect”—and to promote. At the end of the day, I think Substack is a platform, not a publication. But being a platform doesn’t make it neutral. To support my writing, consider a tip on Ko-fi.com. On evacuations and hurricanes 2024-10-08T12:30:00+00:00 It’s so easy, watching on the news, to scream at people who don’t evacuate ahead of oncoming storms. I get it. I scream at them, too. It’s often a fitting reaction. And yet. Evacuation is a process. It’s not “throw a handful of things in an overnight bag and floor it.” You still need to put in the time, money, and effort of preparing your house (or condo, or apartment) for the storm, to maximize the chance of having anything to come back to. You need to have somewhere to go: if you don’t have a friend somewhere out of harm’s way you can stay with, the farther you’re going to have to drive to find a hotel room or an open shelter, and the worse the traffic jam will be getting there. You can solve those problems if you evacuate earlier—before the mandatory orders come down—but only if you have the resources to do so. As I’m typing this, on a beautiful-looking Tuesday morning with a partly cloudy blue sky outside my office window, I’m more or less in the path of Hurricane Milton. If it sticks with projections, it’ll pass within 60 miles of here. Over the past day or so, I’ve watched many people who need to panic do so—and many people who maybe don’t need to panic do so, too, often because people are shouting at them on social media to panic. I don’t doubt this is well-intentioned, but I doubt it’s helpful. There’s a straightforward metric to use to decide if you should evacuate: If you are in an evacuation zone that’s been told to evacuate, evacuate.</li> If you aren’t, don’t.</li> </ul> Sure, there are exceptions—for instance, if you need power for medical devices, get to a special needs shelter. But evacuating when you don’t need to is at best unnecessary and at worst makes things fractionally worse for people who do have to evacuate by putting one more vehicle on the road and occupying one more hotel room. I’ll be surprised if we don’t lose power and wired internet by Thursday morning, but despite being in the “Greater Tampa Bay Area,” I’m over 30 miles inland on relatively high ground. (“High” is always relative in Florida, of course, but we’re 70 feet over sea level.) There are flooding concerns if the conditions are right, because they’re inescapable here—the state is full of lakes and rivers—but for now they’re not likely. See you on the other side. Footnote: wildfire threat in California vs. hurricane threat in Florida I feel—tentatively—like it’s easier to provision a home (or subdivision) with fire breaks than it is to protect against storm surges, hurricane force winds, and floods. Beyond that, this makes three out of three years I’ve been back in Florida with the area I lived in under the threat of a major hurricane, while there were zero out of twenty years in California with the area I lived in under the threat of burning down. Also, the last two years have taught me to stop treating “it’s okay, it’s a dry heat” as a joke. I’d take a Sacramento August over a Florida one in a heartbeat. To support my writing, consider a tip on Ko-fi.com. Generative AI comes for imaginary friends 2024-09-18T12:30:00+00:00 On LinkedIn, the social network of mandatory happiness, software engineer Michael Sayman writes about his new product</a>: Introducing SocialAI, a private social network where you receive millions of AI-generated comments offering feedback, advice & reflections on each post you make. </blockquote> I will admit my first thought about this was: this is a prank, right? This is a fake product that somebody who hates generative AI and the “techbro culture” would come up with to mock it. My second thought was: oh no, it’s real. SocialAI is more than just another project for me – it’s the culmination of everything I’ve been thinking about, obsessing over, and dreaming of for years. I’ve always wanted to create something that not only showcases what’s possible with tech but also helps people in a real, tangible way. SocialAI is designed to help people feel heard, and to give them a space for reflection, support, and feedback that acts like a close-knit community. This app is a little piece of me – my frustrations, my ambitions, my hopes, and everything I believe in. It’s a response to all those times I’ve felt isolated, or like I needed a sounding board but didn’t have one. I know this app won’t solve all of life’s problems, but I hope it can be a small tool for others to reflect, to grow, and to feel seen. </blockquote> And my next thought was—forgive me, but does Mr. Sayman know how genuinely sad an application that creates artificial parasocial relationships for you sounds? I don’t love the phrase “go touch some grass,” but Jesus Hasenpfeffer Christ, if there’s any sign you need to delete Visual Studio Code from your laptop and go take up goat yoga, isn’t this it? As far as I can tell, his intention here is sincere; his new company, “Friendly Apps,” has a mission “to foster connections that prioritize mental health and holistic well-being”. But if you’re thinking “the only kind of person who would think a flood of AI-generated ‘support and feedback’ would be therapeutic would be, I don’t know, a Gen-Z product manager at Facebook or Google,” I have unsurprising news for you. To be sure, Sayman is a prodigy</a>, joining Facebook when he was 18 after already published successful iOS apps. His priority there, though, was helping them go after teenagers, developing an ill-fated Snapchat competitor</a> that lasted a year. One could make the blunt but not unfair case that Facebook/Instagram has contributed more to the need for therapy than any other company, and that perhaps one should have thought of that before, you know, becoming a product manager for them. Maybe I’m being cynical. Maybe SocialAI will be fruitful in combating loneliness and isolation. But maybe you should hang out with a non-robotic friend or two, whether online or off, instead. To support my writing, consider a tip on Ko-fi.com. How much of yourself do you put online? 2024-08-26T12:30:00+00:00 It’s as much of an age-old question as anything of the internet age is, but I think about it every few years. Slapping up a web page full of personal thoughts in 1999 didn’t carry the same weight that it does in 2024; employers, even ones literally in the internet space, didn’t necessarily do a web search on your name to see what came up. Now, though, I don’t know. Now everyone does that kind of search. And, as someone who’s been conducting an unsuccessful, albeit woefully low-key, job hunt in 2024, it’s hard not to ask if what I have online at this point is working against me with more skittish employers. Even before I wrote about conservatives being snowflakes</a>, it wasn’t exactly hard to suss out my left-wing politics, and I’ve been vocally</a> skeptical</a> of generative AI. In a moment where tech companies madly chase AI in either a sincere belief it’s going to change everything or mere FOMO that it might, vocal AI skepticism might be a career-limiting move—and in a moment where alt-right techbros cow companies into walking back DEI efforts, merely noting my pronouns at all, let alone noting them as “he/they”, might also be one. (“What kind of woke beta cuck uses pronouns?” Anyone who speaks English, you stupid motherfucker.) (Also, cursing in my blog might be a career-limiting move.) And yet, I don’t know. Do I want to frantically dash around the internet, tearing down anything that might give a prospective employer the heebie-jeebies about me? Is that even possible? While I’ve never reached “internet famous” status, I’ve posted lots of stuff in lots of places. When I was more actively doing tech blogging at Coyote Tracks, it got linked to by other blogs often enough that I used to joke that while you’re probably not reading me, somebody you’re reading probably is. There is an alternate universe in which I figured out how to monetize that. And, well, while I don’t advertise being a furry, I’ve been writing furry fiction for decades and have been guest of honor at more than one furry con. One does not have to possess mad internet sleuthing skills to put two and two together. Frankly, I’m always amazed when I find people I grew up with, especially members of the Original Internet Nerd cohort, have next to no footprint online. The one whose LinkedIn includes “Futurist at the Center for 21st Century Teaching Excellence” you’d kind of expect to have, I don’t know, a low-effort Substack, if not a YouTube channel, right? If you can tell someone your job title is “futurist” and keep a straight face, I’m pretty sure they’re legally required to give you a TED Talk. But apart from that LinkedIn, there’s basically nothing out there about them, and that’s true for the majority of other folks I knew in that place at that time. Does that help with job hunts? Maybe. On the other hand, maybe it hurts. It’s my understanding the former futurist’s LinkedIn is long out of date, and they struggled to find work for years before moving into non-tech fields. I can’t say that actually being, you know, present on the internet might have helped if they’d wanted to stay in tech, but I wonder. I made the shift to tech writing largely because of my loudmouth tech blog, when Joey Zwicker at RethinkDB came across it and thought, “we should talk to them about our new tech writing position.” Of course, I don’t know that they did want to stay in tech. If they didn’t, I couldn’t blame them. In any case, I don’t think erasing my digital footprint is either possible or worth it. Anyone who reads what I write and decides they can’t work with someone like me is probably right. To support my writing, consider a tip on Ko-fi.com. Conservatives are the real snowflakes 2024-08-02T12:30:00+00:00 There’s been plenty of words written already explaining why calling Trump, J. D. Vance, and the rest of the MAGA</abbr> branch of the Republican Party—which is to say, the Republican Party—weirdos seems so effective. These conservatives have spent years, so the thought goes, asserting their worldview is the standard, the default, the normal worldview, and everything that deviates from it is wrong, deviant, and yes, weird. Calling them weird flips that on its head. While that’s undoubtedly true, I think the real reason it’s effective is that conservatives are—and have been for years—snowflakes. This isn’t new to the MAGA</abbr> wing; it was the real animating factor of the Tea Party, not economics. (It’s never really been about economics; if it was, we’d be having some real engagement over why, if Republicans are better at running the economy, the economy has, for the last seventy years or so, consistently done better under Democratic administrations</a>.) “But,” someone might say, “surely liberals are the real snowflakes, the ones who want their feelings to be protected from facts?” To which I say: no, imagined interlocutor, that’s transparent bullshit. I mean, come on. Is it liberals and leftists who’ve been screaming for years that merely seeing queer people in television shows and movies more often now is literal warfare? No, it is not. Have you met anyone truly offended by seeing Christmas decorations or having a cashier say “Merry Christmas?” No, but you’ve met dozens who claim to get the vapors when a cashier, in a store full of Christmas decor, says “Happy Holidays”. I had a conservative (ex-)friend strenuously argue that yes, in fact, a state giving gay people the right to marry somehow devalued his good conservative Christian marriage. His arguments were terrible, because all the arguments for that position are terrible. No matter what kind of intellectual varnish one tries to give it, the heart is the same as every other culture war argument: No one should be allowed to make my family and I aware of things I don’t personally approve of. Books show queer people in a positive light? Get ’em out of the libraries, because kids might read ’em, and the mere knowledge that gay, lesbian, and transgender people exist is corrupting. (Everyone knows there were no gay or trans people before Bill Clinton was elected.) Boycott stores that put up rainbow flags, or hire women and blacks. (Sure, you’re not against hiring minorities, you’re just against visibly making an effort to do so. Gotcha.) Don’t teach our children history that could conceivably show their ancestors in a bad light—that might make them slightly uncomfortable. What kind of monster makes children engage with difficult topics? The anti-MAGA</abbr> narrative has always had an undercurrent of these people would be laughable if they weren’t so dangerous. But the “weird” argument sticks because it acknowledges one can be both dangerous and laughable. Any guy irresponsibly blasting away with an assault rifle is dangerous. One irresponsibly blasting away with an assault rifle at cases of Bud Light because he was made aware that the beer company sells to trans people, too? I mean, come on. How do you get more snowflake than that? Actually, forget I asked that. Based on MAGA</abbr> reactions to the Harris campaign so far, we’re gonna have some real blizzards. Buckle up. To support my writing, consider a tip on Ko-fi.com. It could be burnout 2024-07-29T11:00:00+00:00 Two years ago, I moved back to Florida for family reasons; a year in, I mused on the changes, both good and bad, that the move brought. Now two years in, the most obvious life change is, (probably) unfortunately, employment. The executives at the company I’d been working at for over five years, remote full-time for over three, issued an edict declaring that remote work was no longer acceptable. Having moved three thousand miles away—with permission, to be clear—was no reason to grant an exception. Moving back to Silicon Valley wasn’t a realistic option, so I “quit” in December. (That absolutely requires air quotes. Yes, it was technically my “choice” not to scramble to find a way to move myself and my mother back to the most expensive metro area in the country, but come on.) I’ve applied to a few companies since then. The closest I’ve gotten was with a company which I think I interviewed very well at. They assured me that my lack of real Kubernetes</a> experience wouldn’t be an issue, since they had other technical writers who didn’t have any when they started. But, at the last minute I was told, no, actually they wanted a candidate with more Kubernetes experience. Well, then. None of the other companies moved me past the recruiter screening stage. I won’t lie; this has been a mild shock. Not that long ago, companies you have assuredly heard of pinged me about job opportunities fairly regularly. I didn’t follow up on them because my job paid stupidly well. Now that I actually want to hear from them, the pings have largely stopped. Is it just the soft tech job market? That can’t be ruled out. Tech writers don’t face quite the same bias that engineers do, but it can’t be completely ruled out, either. Is it the fact that recruiters have a documented bias for currently employed candidates, because obviously, anyone who’s been out of work for six months must have something wrong with them? That possibility seems disturbingly high. Are companies that were hiring for tech writers now assuming they can create docs with Generative AI and not hiring? No direct evidence of that, but put a pin in “AI”; I’ll circle back shortly. Also, though: it’s possible that I’m not trying very hard. I have, on the surface, good reasons. First, I wanted a break for a month or so. Then, I knew I was going off on an already-planned vacation to Texas in March. Later, I had an opportunity to return to a two-week residential writing workshop</a> in Lawrence, Kansas, at the end of June. So I told myself to be selective, to only apply for companies I thought I’d really want to work at. This approach lines up with what the best-selling job hunting book of the last several decades, What Color is Your Parachute</a>, prescribes: figure out what fields and positions truly call to you and then figure out how to put your best foot forward, even if you’re not obviously qualified. Of course, I haven’t really worked through all of that, because the truth is that it’s genuinely damn hard to follow that advice. I tried a quarter-century ago and got the assessment that I should do what I’m good at—writing, (basic) web development, desktop publishing—for wildlife or environmental nonprofits. You know what? That’s hard to break into, and I didn’t have any real connections there. So I ended up staying in tech, doing tech things. This is not the approach the rest of the job-hunting industry has. The most common advice you’ll find on LinkedIn can be paraphrased as: Apply to everything all the time if you’re not applying to ten positions a week you’re a laggard better yet ten positions a day but still customize your resume and your cover letter for each one and figure out how to write each one so they get past the AI scanners also have you considered standing outside hiring managers’ homes playing “In Your Eyes” on a boom box for Christ’s sake GO GO GO GO GO </blockquote> I’ve always found job hunting tiring, and I suspect I’ve grown too complacent from the perception that jobs would mostly come to me now. Also, I have a strong suspicion that the hiring system, such as it is, has gotten worse over the last few years, as both jobseekers and job posters try to game things out in ways that don’t truly reward either side. The skill of optimizing résumés to get them past automated screeners is, in nearly all cases, unrelated to the skills employers are actually looking for. And if you are one of the 7 in 10 hiring managers who thinks posting fake jobs is morally acceptable</a>, I mean this in the nicest possible way: go fuck yourself with a rusty fondue fork, you soulless bottom-feeding pond mold. So as I was saying, I wonder if I’m just a little burned out. Some of it—maybe most of it—is shifts in the tech industry itself. I won’t pretend that there haven’t always been flimflam artists and fads, but the desperation has gotten palpably worse in the last few years. We’re no longer claiming the Next Big Thing will merely be a Big Thing; we’re claiming it will revolutionize everything, that whatever it is will surely be bigger than the iPhone, or the internet itself, or the entire microprocessor revolution. What is this incredible thing, the biggest thing since sliced bread and/or fire? Why, it’s virtual reality, of course! No, wait! It’s cryptocurrency! No, it’s NFTs! No, it’s AI! It’s not just the fads, though. It’s the politics. Again, this isn’t new; there have always been companies with toxic cultures (Uber) and actively malicious raisons d’être (Palantir). But the gravity has drifted from a sort of left-leaning libertarianism—anti-regulation but socially liberal—toward a weird neoreactionary</a> techno-authoritarianism coupled with passive hostility to anything that smacks of “wokeness”. In 2014, startups got money by aspiring to be Facebook for dogs; in 2024, startups get money by aspiring to be Chick-Fil-A with LLMs. I miss the naïve complacency of not having to give a damn about Marc Andreessen’s politics, of not fretting that startup engine Y Combinator</a> is now run by a technofascist troll</a>. So, all this is to say that I feel like I might be at a crossroads, a turning point, a belabored metaphor about personal change. I’m not actively looking to leave technical writing, but I think I’d like to find other venues for it. Indie software developers looking for documentation polish. Smaller companies that want manuals or internal documentation. Heck, wildlife or environmental nonprofits needing writers for…something? I know how to research and review and be moderately entertaining. None of that is to say that I wouldn’t take a full-time job with a tech startup, assuming there are any left that won’t slam the door in my face after reading this article. But I’m not sure how much energy I’m willing to invest in pursuing full-time jobs with tech startups at this point. Unless they’re very interesting. (And don’t slam the door in my face after reading this article.) Beyond work, life is…good? The question mark comes, of course, from politics again. Florida’s political climate hasn’t improved since last year. Neither has Florida’s climate, for that matter. As I write this, it’s before nine in the morning and already 78 °F—with 96% relative humidity, for a 77 °F dew point. 25.5 °C and 25 °C respectively, for those of you who don’t want to do the math. And, both political climate and climate climate contribute to Florida being in the middle of the largest Covid-19 surge in the country, something it feels like we’re studiously pretending isn’t happening. (Granted, most of the country</s> world is pretending along with us.) Despite all that, both greater Tampa Bay and Orlando remain appealing, vibrant places. I’ve mostly made my peace with what is (and isn’t) available within a 20-minute drive, accepting that I’m going to put a couple of hundred miles on my car each Saturday on excursions. A revival of one of the original tiki bars, Don the Beachcomber</a>, opened its first location in Madeira Beach and a smaller bar in downtown Tampa. I’m getting a lot of creative writing done on various projects. None of this is likely to become serious income (as I joke, I’m making tens and tens of dollars on fiction), but who knows. I was able to make it to a two-week residential writing workshop</a> to get serious about rewriting a messy fantasy novel. (The first time I attended that workshop resulted in my first completed novel—and, generally, getting me to figure out how to write novels well, period.) I have a software project in mind that I may eventually manage to get to, which also ties into creative writing, even though part of me wonders whether I should be trying to just write it as a plugin for Obsidian</a>. And the truth is that financially, I’m doing okay. The decade of higher-paying tech writing that I got in hasn’t set me up for early retirement, but it’s set me up with a pretty big safety net separate from the IRA and 401(k). I do hope to get more written for this weblog, although the vague sense of burnout makes focusing on tech articles more difficult than it used to be. Ed Zitron has the market for profane tech rants largely cornered. While I’m still a mostly happy Mac user, it’s increasingly despite Apple as a corporation rather than because of them, but I haven’t felt like I had much to contribute to that conversation, either. (Sorry, anti-Apple folks, I still find Windows and Linux worse for my personal tastes, and still find a lot of anti-Apple rants to have a distinctly performative look at how I am not a sheeple subtext to them.) In the meantime, I’ll keep too-selectively applying at jobs, try to figure out if there’s a way to reach those ones and ones of indie software developers looking for documentation help, plug away at my creative writing projects and possible software project, and keep on keeping on. To support my writing, consider a tip on Ko-fi.com. Creative writing and AI's failure modes 2024-05-23T15:30:00+00:00 In which I fall down a rabbit hole with a purported “AI novel writer,” and wonder what the result tells us about generative AI’s future. I spent nearly six years as a technical writer at an AI company, and it was one of the best jobs I’ve ever had. I worked with some brilliant people, several of whom have gone on to found other AI-focused companies. The AI that’s taken off in the last few years is not what we were working on there, though. It’s specifically generative AI, a variant based on machine learning and neural networks. And “taken off” is an understatement; I haven’t seen anything like this since the original dotcom boom. You may recall that the original dotcom boom ended in a spectacular dotcom crash. While I am not saying that this is foreshadowing, if you can read about Logitech adding an AI prompt builder to their mice</a> and not think of a certain sock puppet</a>, you are more starry-eyed than I am. As someone who’s been a professional technical writer and at least nominally a professional fiction writer, I’m caught between two opposing viewpoints. I watch folks in tech circles—including many who are indie creators, of writing, podcasts, software, whatever—using generative AI without any seeming concern over the ethical issues. And, almost every tech blogger/podcaster I follow seems to take it as a given that GenAI is here to stay and is obviously the future of everything. I understand the enthusiasm, but I also find the handwaving past significant problems with GenAI output, ranging from copyright violations to unreliable outputs (“hallucinations”) to be frustrating. On the flip side, many creators I know get instantly prickly if I admit even slight enthusiasm for any of this stuff, let alone suggest that there could ever possibly be value in generative ML. They don’t want to hear about how cool it is even in the most limited contexts. As far as they’re concerned, this is the second coming of crypto, a solution in search of a problem, a scam pushed by techbros who neither understand nor like creative work. I’m sympathetic to this take, but it’s also frustrating. It sweeps everything “AI” up into what should be a condemnation specifically of generative AI, and it dismisses any possibility that GenAI has value now—or could ever have value if we can work out those concerns. (That’s a big “if,” which I’ll come back to at the end.) Authors might love a system that could keep a local index of the three volumes of the fantasy epic they’re on volume three of. When they try to remember what town they said Brenda the Elf Maiden was from, instead of searching for the word “Brenda” through four hundred thousand words of text, they could literally type “What town is Brenda from” and get the answer. And as anodyne as ChatGPT’s outputs tend to be, I can imagine using it as a brainstorming partner, prompting it with things like “list a few themes in a myth about a woman given the power of a weather god,” or using it to analyze your story through different narrative theories, suggesting improvements. But AI boosters—developers, pundits, genuine enthusiasts, and techbros looking for the Next Big Thing—don’t want to stop there. They want it to actually write stories, to take the mere kernel of an idea and spit out a whole screenplay or novel, maybe even the cover art. If they get their way, AI will soon free us from the drudgery of (checks notes) creative pursuits that give our lives meaning! Cool. Cool cool cool. Seriously, though, looking at how well LLMs do at such a task tells us more than whether AI is creative. “Good” fiction is subjective, but competent fiction involves defining a structure, formulating a plan, and executing that plan to produce long-form output. This describes more than writing novels: most of the more fantastic predictions about what AI can and will do for us (or to us) requires similar skills. My thesis has been that LLMs will suck at such complex tasks because of they actually are at their heart: predictive lexical analysis engines. Yet, even the “old” ChatGPT 3.5 can confound my expectations. Recently, I prompted it to describe the seven-point story structure</a> to me, which it did, and then in the same session prompted, Using the seven-point story structure, write a short synopsis of a story about a 1920s private detective hired by a wife to find evidence her husband is cheating on her. The detective discovers the husband is being blackmailed, and this leads the detective to uncover a smuggling operation run by gangsters. </blockquote> It responded: Hook: In the bustling streets of 1920s Chicago, private detective Jack Malone is hired by a distraught wife to uncover the truth about her husband’s infidelity. Plot Point 1: Jack begins his investigation and quickly discovers that the husband, Thomas, is not cheating on his wife, but is instead being blackmailed by an unknown individual. Pinch Point 1: Jack’s efforts to uncover the blackmailer lead him to a seedy underground gambling den, where he narrowly escapes a dangerous confrontation with a group of thugs. Midpoint: Jack stumbles upon evidence that the blackmail scheme is connected to a larger smuggling operation run by notorious gangsters in the city. Pinch Point 2: As Jack delves deeper into the criminal underworld, he becomes a target himself, with the gangsters determined to silence him before he can expose their illegal activities. Plot Point 2: In a thrilling showdown at a hidden warehouse on the outskirts of town, Jack confronts the gangsters and uncovers the full extent of their smuggling operation, ultimately bringing them to justice. Resolution: With the truth revealed and the criminals behind bars, Jack delivers the evidence to Thomas’s wife, who is grateful for his help in not only saving her marriage but also bringing down a dangerous criminal enterprise. Jack walks away from the case with a renewed sense of purpose and a reputation as one of the city’s most skilled and fearless detectives. </blockquote> Okay. “Jack Malone” is the detective Anthony LaPaglia played in the long-running crime drama Without a Trace, but setting that aside, this is impressive work at first glance, isn’t it? It knew what the seven-point structure was and applied it to my prompt to come up with a perfectly credible outline. Isn’t this showing both reasoning ability and a touch of creativity? Wellllll… Again, LLMs respond to prompts based on statistical probabilities. There’s a sense in which they’re reasoning about the text, but that’s not the same as understanding the text. This is where AI boosters leap into smugly say, “Maybe that’s the way human brains function, too, did you ever consider that?” (And yet, you participate in a society! Curious.) But the unique failure modes of machine learning tells us that’s not, in fact, the way human brains function. An image classifier may correctly identify a picture of a dog 99.99% of the time, but once in a blue moon it’ll tell you a Mondrian painting is definitely a labradoodle. However they’ve taught themselves to identify dogs, they haven’t done it the way humans do. They don’t teach themselves to generate text or images the same way, either. Even setting aside the fingers thing, floor patterns change, people subtly merge into chairs or keyboards, straight lines in the background don’t quite match on different sides of a foreground object. Stable Diffusion, Midjourney, and friends “know” how to correlate the text in your prompt with images in your database and blend them together in an often magical-seeming way. That’s not the same as knowing what floors and chairs and people are, though, and how backgrounds and lighting and perspective works. The same is true for text, sometimes in obvious ways (factual errors, failing at elementary school level logic puzzles), and sometimes in subtle ones. In our outline’s case, the LLM has countless examples of both detective stories and seven-point story analyses in its corpus, and computed the lexical equivalent of averaging the prompt and the structures together. You’d expect what you get out should definitionally be average, and that’s precisely what we got: every time ChatGPT expanded on the prompt, it did so in the most generic way possible. The only details it gave were (plagiarized) names. Pinch Point 2 isn’t the “all is lost” moment it should be—it’s a description of how literally every detective-versus-the-mob story works. How does Jack get from confronting the gangsters to bringing them to justice? What evidence did he find? What’s the connection between the gambling den and the smugglers? And what’s with the implication in that pinnacle-of-absolute-median resolution that Jack had to renew his sense of purpose? Come on, he’s goddamn Anthony LaPaglia! Another run-through with the same prompt might produce a better outline, one that hints at higher stakes and specific dangers and less hand-wavy resolutions—but it might produce a worse one, too. This is why some detractors dismiss this as “spicy autocomplete”; I think a better description is “galaxy-brain level Mad Libs</a>.” Tweaking the prompt might help, but you’d be better off using the generic outline from the first run as a jumping-off point, filling in your own details, raising the stakes, eliminating copyright violations, and so on. Now, that right there—the generated output as a jumping-off point—is arguably a solid case for using LLMs in creative writing: they can get you started, get you past writer’s block, spit out ideas that might prove helpful even when the ideas themselves aren’t that great. An LLM trained, perhaps with retrieval-augmented generation</a>, to more directly apply narrative theory could be even better. Subtxt with Muse</a> does this with the Dramatica Theory of Story</a>, and while I haven’t used it myself, the actual-use examples on its website are very impressive. Of course, you could also take your prompt through brainstorming exercises manually, or with a non-ML program like the original Dramatica Story Expert</a>. Story Expert will probably not run on your Mac if it was bought in the last decade. The developers spent two decades before that doing the bare minimum to keep their codebase running on modern hardware (seriously, the thing somehow never had retina fonts!), and are now buried under a mountain of technical debt. But I digress. Meet Claude</h2> But, GPT-3.5/4 isn’t the only LLM in town. Many say Anthropic’s Claude is more “creative” (as squishily defined as that may be), and AI enthusiast/CEO Matt Shumer has created a script that uses Claude to write an entire novel. As he put it on Twitter/X</a>, Claude-Author is the first AI system that actually produces readable books. Still not perfect, but it’s a leaps and bounds improvement over previous approaches. </blockquote> When someone asked, “Are the books any good, though? Thoughtful twists or exciting plots,” he responded, “Shockingly good.” Is it true? Can Claude show adequate reasoning and creativity—or convincingly simulate both of them—well enough to write an imperfect but still shockingly good book? A disclaimer</h2> While I used generative machine learning for testing purposes here—to put the claim that there’s a there there under a microscope—I want to recognize up front that’s not a “Get Out of Ethical Concerns Free” card. I’ll get into those in the last part of this essay. If you think that means I shouldn’t have even run this experiment, I get it. But there’s no way to test the increasingly extravagant claims for AI besides, well, testing those claims. Critics of AI sometimes implicitly take the claims of the boostiest of AI boosters at face value, proceeding under the assumption that, yes, AI is on the verge of replacing novelists and screenwriters and artists and filmmakers. If those assumptions are wrong, that has big implications for both boosters and critics alike. So what is this?</h2> Claude-Author</a> is an open-source project that “utilizes a chain of Anthropic and Stable Diffusion API calls to generate an original novel.” (The original project used GPT-4 and specified “fantasy novel,” but the most recent version uses Anthropic’s Claude 3</a> and drops internal prompts that limit it to fantasy.) According to its README, The AI is asked to generate a list of potential plots based on a given prompt. It then selects the most engaging plot, improves upon it, and extracts a title. After that, it generates a detailed storyline with a specified number of chapters, and then tries to improve upon that storyline. Each chapter is then individually written by the AI, following the plot and taking into account the content of previous chapters. Finally, a prompt to design the cover art is generated, and the cover is created. Finally, it’s all pulled together, and the novel is compiled into an EPUB file. </blockquote> This is the sort of promise that gives AI buffs thrills and novelists hives. Scammers have filled Amazon’s ebook store with nonsense for years; that’s not a subjective dig at the quality of self-published books, but a statement of fact</a>. So it’s not surprising that scammers have embraced AI</a>, and that’s before AI promised to actually be good at it. So how good is it? Getting Claude-Author running</h2> As with many science projects of late, Claude-Author runs in JupyterLab</a> (née Jupyter Notebook), an interactive Python environment. As written, the code lacks any error handling; I added some to help debug repeated crashes. (It kept exceeding token limits in the Anthropic API.) Prompts get converted to probabilistic tokens</a>, numeric representations of words or “subwords.” The LLM also uses tokens for generating its output. I also separated out the steps so I could view the generated outline and cover art prompt. Claude charges by the token, both in and out. Claude 3 “Haiku,” the model Claude-Author uses, has a rate limit of 50,000 tokens per minute at the cheapest paid tier. The prompt for each chapter consists of the user prompts, the generated outline, and every previously-written chapter. With about 4,000 tokens used per chapter and a few thousand more for the outline and cover art, you’re going to run out of tokens by 12 or 13 chapters on the cheapest paid tier. This is not a documented limitation. Being a simple wrapper script doesn’t fully excuse its bugs. Book 1: The Elven Crown</h2> So, my first prompt: Plot summary: An elf princess finds herself made the queen after her mother (the current queen) and her sisters are killed in a suspicious accident. She must learn to become a good ruler while navigating court political intrigue, dealing with rivals to the throne who would like to see her removed, and solving the mystery of the accident that killed her family.</li> Writing style: literate but still friendly and conversational.</li> Chapters: 12.</li> </ul> You may recognize this as a gender-swapped version of The Goblin Emperor</a>, an acclaimed fantasy of manners</a> novel—basically, court intrigue in a fantastic setting—from a decade back. The outline Claude wrote for itself, which it called The Elven Crown: Ascent of Queen Elara, was a list of bullet points, optimistically specifying how many pages each chapter should be. Here’s the first two chapters: Chapter 1 (10 pages) Introduce Elara, the young elf princess, as the story’s protagonist. Describe her personality - intelligent, caring, a bit shy and unsure of herself. She lives a comfortable life in the royal palace, but longs for more adventure and responsibility.</li> Set the scene of the royal palace and the kingdom Elara lives in. Describe the luxurious surroundings and the daily life of the royal family.</li> Foreshadow some tension and political intrigue brewing beneath the surface, with mentions of Elara’s mother (the Queen) being stressed and preoccupied lately.</li> End the chapter with the shocking news that the Queen and Elara’s sisters have been killed in a mysterious carriage accident. Elara is now the sole heir to the throne.</li> </ul> Chapter 2 (10 pages) Elara is reeling from the sudden loss of her family. Describe her grief and the funeral proceedings. She is overwhelmed at the thought of becoming Queen.</li> Introduce some of the key members of the royal court - the High Council, the Captain of the Guard, the Royal Advisor, etc. They offer their condolences but also start subtly vying for Elara’s favor and trust.</li> Elara meets with the High Council, who inform her that she must be crowned Queen immediately, per the laws of succession. She is unsure and scared, but accepts her new role.</li> </ul> </blockquote> In the following chapters (according to the outline), the new queen “has a strange dream/vision that seems to be a warning about a dark force threatening the kingdom.” Chapter 4 introduces a “potential romantic interest, perhaps a handsome young member of the royal guard or a visiting diplomat.” After “strange occurrences and accidents start happening around the palace,” a mysterious cloaked figure who’s been watching her warns her about an impending threat and urges her to flee the palace for her own safety. She agrees (because why not), ducking the royal guards and traveling “incognito through the kingdom,” staying one step ahead of pursuers. A band of rebel fighters agree to help her sneak back into the palace, but their leader warns her about “powerful forces at work, with deep roots in the kingdom’s history.” Eventually, she returns, confronting “the true mastermind behind the plot against Elara’s family, perhaps a disgruntled member of the High Council or a rival noble family.” A battle ensues. “Just when all seems lost,” her cloaked ally reappears, “joining the fight and turning the tide.” The last chapters find her “rooting out corruption and restoring trust in the kingdom’s institutions,” and the cloaked figure is “revealed to be a powerful sorcerer who has been guiding and protecting Elara all along.” The romantic subplot comes to a resolution. So, it’s like that Disney film where the princess has to leave home to Learn a Lesson before returning triumphantly (you know the one), with a generous dusting of Discount Tolkien. The biggest problem with the outline as it stands, besides the remarkable lack of originality, is that Elara has no character arc beyond moving from unsure she’s ready to be a queen to being confident she can be. What does she want, besides not screwing things up? What does she need to learn? What does it take to succeed? Even so, a good writer could take this outline and come up with, well, something from it. Let’s read the actual opening! Elara gazed out the ornate window of her bedchamber, her emerald eyes scanning the lush gardens and towering spires of the royal palace. Though she had lived here all her life, the grand, gilded halls and meticulously manicured grounds had never felt like a true home. There was an emptiness, a disconnect, that nagged at her constantly. At 18 years of age, Elara was the youngest of three princesses - a position that afforded her every comfort and luxury imaginable, yet left her feeling isolated and unfulfilled. While her elder sisters reveled in the pomp and privilege of court life, Elara longed for adventure and the freedom to explore the world beyond the palace walls. She sighed, her slender fingers tracing the intricate patterns etched into the windowsill. Elara knew her role was to one day succeed her mother, Queen Alaria, as ruler of the kingdom of Valeria. But the weight of that responsibility filled her with a sense of dread rather than pride. Was she truly ready to bear the burden of an entire realm on her shoulders? A soft knock at the door interrupted Elara’s pensive musings. “Princess? Your mother requests your presence in the throne room.” It was Mara, her most trusted handmaiden. Elara straightened her posture, smoothing the delicate folds of her ivory gown. “Thank you, Mara. I shall be there momentarily.” </blockquote> “Show, don’t tell” may be over-prescribed as basic writing advice, but this opening is the platonic archetype of why that advice exists. We’re told everything and shown nothing. Beyond that, note the utter lack of concrete imagery: we’re told the royal palace has gilded halls, manicured grounds, and towering spires, but what do they look like? “Ah, Elara. Thank you for joining us,” Queen Alaria said, her voice betraying a hint of strain. “Please, come and stand with your sisters.” Elara obediently took her place between Aeyla and Celeste, acutely aware of the numerous pairs of eyes watching her every move. The members of the High Council, the kingdom’s most powerful advisors, were gathered in a semicircle around the throne, their expressions unreadable. “My daughters,” the Queen began, her gaze sweeping across the three princesses. “I have summoned you all here today because… because I have grave news.” She paused, her lips trembling slightly. “Your father, the King, has been killed in a terrible accident.” </blockquote> Oh, no! The king! *record-scratch* Wait. My prompt said the queen and the princess’s sisters had been killed. What happened to the king might need to become backstory, but if the queen is still alive, she takes over, not Elara. (Unless the country doesn’t let women be rulers, but that ends the story right here too, doesn’t it.) As the Queen dismissed the Council and the sisters were ushered out of the throne room, Elara couldn’t help but feel a growing sense of foreboding. Something was not right, she could feel it in the air. The kingdom she had known all her life was on the precipice of a great upheaval, and she was powerless to stop it. That night, as Elara lay awake in her bed, haunted by the day’s events, she couldn’t shake the feeling that her life was about to change forever. The weight of the crown loomed over her, a burden she was not sure she was ready to bear. But with her father gone and her mother’s health visibly deteriorating, Elara knew she had no choice. She would have to find the strength within herself to become the Queen Valeria needed, even if it meant sacrificing her own dreams and desires. For the sake of her family, her people, and the kingdom she had sworn to protect, Elara would have to rise to the occasion. Little did she know, the challenges that lay ahead would test her resolve in ways she could scarcely imagine. The future of Valeria hung in the balance, and Elara’s very survival would soon be at stake. </blockquote> Okay, Claude simply isn’t following its own outline, is it? It never described the daily life of the royal family, or mentioned the Queen being stressed and preoccupied. The chapter didn’t end with the news of the Queen</s> King’s death. And again, those last few paragraphs are cliché after cliché with no specificity. Chapter 2 The morning sun peeked through the ornate windows of the royal palace, casting a warm glow over the solemn proceedings taking place within. Elara stood beside the grand, gilded caskets that held the lifeless forms of her mother and sisters, her emerald eyes glistening with unshed tears. She had spent the past two days in a fog of disbelief, her mind struggling to comprehend the devastating turn of events that had robbed her of her family. One moment, they had been alive and well, and the next… gone, taken from her in the blink of an eye. </blockquote> Surprise! Back to the original outline: the King goes unmentioned, and the Queen and Alara’s sisters are dead. A soft knock at the door interrupted her pensive musings. “Your Majesty?” It was Mara, her most trusted handmaiden. “Forgive the intrusion, but there is a matter that requires your attention.” </blockquote> Wasn’t Mara introduced in Chapter 1 with the same phrasing? Why, yes, she was. Claude will introduce her again in Chapter 3 with exactly the same paragraph as above. And again in Chapter 4. I could continue fisking The Elven Crown, showing how it repeats not just phrases and actions but whole entire scenes, sometimes three or four times, but you get the idea. We’re told Elara is “determined to emerge stronger, wiser, and more resolute” and has “a renewed sense of focus and determination” and is “determined to be the beacon of hope and stability that her people so desperately needed” within four paragraphs, because (within the same short passage) “she had been forged in the crucible of her family’s tragedy.” No, she hasn’t, because it’s only chapter four. The novel is about that crucible, how it forges her. The phrase “the weight of the crown” occurs eight times in the text, usually as “the weight of the crown never felt heavier.” Did you know the crown is heavy? It’s a metaphor! Get it? A big stinkin’ hoary cliché of a metaphor that Claude’s infallible calculations predict you want to read at least once a chapter! Like I said, this isn’t spicy autocomplete as much as galaxy-brain Mad Libs. Claude generates phrases, paragraphs, and even whole scenes paraphrased from a vast library of public domain fantasy stories and far vaster library of copyrighted fiction scraped off the web. Anyway, that’s why the plot has this shape (the most common kind of princess story is, obviously, a Disney Princess® story), why there is a mysterious cloaked figure, why we have sorcerers, why the outline ends with a “Scouring of the Shire</a>” vibe. Claude-Author is generating text that matches a statistically median elf princess novel. There is no romantic interest introduced, ever. The cloaked figure introduced in chapter 5 is her most trusted advisor, Lord Everett, but later on the figure is magically the rebel leader. The text refers to the “accidents” the outline says were happening around the palace, but we never know what they are. (We never know what the “unrest” among the people is, either.) Chapter 5 ends in the middle of a sentence, a bug that continues from here on out. Chapter 6 ends in the middle of the same sentence, in fact. The real villains are sorcerers in an evil cabal, rather than a rival family (dispensing with the one part of the outline that actually followed the prompt: we asked for a novel of political intrigue, remember?). They’re revealed by Everett first, then by the rebel leader as if the first revelation never happened. (Twice.) Elara goes to meet the rebels, who are “working to undermine the forces that seek to destroy [Elara’s] kingdom,” because Claude does not know what rebels do. She returns to the castle with the rebels. Everett is revealed to be conspiring with the sorcerers! Elara is, with no foreshadowing, a master swordswoman. (Claude does not do foreshadowing, at all.) We seesaw back and forth around the outline. Scenes repeat themselves near-verbatim with reckless abandon. Everett switches sides. Another cloaked figure who is not Everett or the rebel leader appears, giving a variant on a speech we’ve read at least three times by now about dark conspiracies and sorcerers and blah blah blah. The novel ends not with Elara but the new mysterious cloaked figure saving the day: Seizing the moment, the cloaked figure launched a barrage of arcane energies, the raw power of their sorcery slamming into the sorcerer’s weakened frame. The towering figure let out a bone-chilling scream as they were engulfed in a blinding flash of light, the very air crackling with the intensity of the magical onslaught. When the light finally faded, the sorcerer’s lifeless form crumpled to the ground, their dark power dissipating like a wisp of smoke in the wind. Elara felt a surge of triumph and relief wash over her, her heart pounding in her chest as she surveyed the </blockquote> Yes, it ends mid-sentence. Don’t worry, though—I’m sure everyone lived Happily Ever Stable Diffusion created this cover prompt: The cover features a regal portrait of Elara, adorned in her royal attire and standing tall with a determined expression. In the background, the silhouette of the royal palace looms, with shadowy figures lurking in the foreground, hinting at the dark forces conspiring against her. </blockquote> As you can see, it decided to just make Elara the shadowy threatening figure. There are also a few typical GenAI bugs (e.g., the moon isn’t actually round). </a> Cover image for The Elven Crown </figcaption> </figure> Book 2: The Elf Queen’s Ascension</h2> I lowered the number of tokens down to 3200 from 4000 to try and create a longer book. Bzzt. Instead, it lowered the word count of each chapter. (Chapter lengths tend to grow as the book gets longer. That might account for the broken sentences, but Claude 3 has a 200,000-token context window</a>—that is, the size of text that it should be able to fully consider in prompts—so it shouldn’t be.) An old truism in writing classes is that if you give twenty students the same prompt, you’ll get twenty different stories. I gave Claude-Author the same prompt the next day, assuming this would be true for it, too. Surprise! The details weren’t identical, but they were damn close, down to naming the main character “Alara” instead of “Elara.” It’s still full of clichés. The phrase “the weight of the crown” doesn’t appear eight times in the text, it appears 45 times, usually as “the weight of the crown pressed down upon her.” The plot still involves trusted advisors, sorcerers, leaving the palace, making a triumphant return, and mysterious cloaked figures who undermine the protagonist’s agency, so I won’t spend much time on it. The Elven Crown never actually uses the word “elf” or, but for the title, “elven”; The Elf Queen’s Ascension at least pays lip service to the notion that these are, you know, elves. This time Claude generated chapter titles, and boy, they are not good! The Burden of the Crown</li> The Burden of Sovereignty</li> Whispers of the Past</li> The Burden of Destiny</li> Whispers in the Shadows</li> Whispers from the Shadows</li> The Gathering Storm</li> “The Unveiling of a Dark Conspiracy”</li> The Awakening</li> Forging Alliances</li> The Crucible of Battle</li> The Awakening of Ancient Power</li> The Guardians of the Darkness</li> The Reckoning</li> “Rebuilding the Elven Realm”</li> “A New Dawn for the Elven Realm”</li> </ol> (I am pretty sure I played several of these as D&D modules.) Why are some in quotes and some not? I don’t know, but they appear that way in the generated ePub file. So do different chapter titles, because Claude generated them both for an internal “make chapter titles” prompt (which didn’t work on the first novel) and for each chapter anyway (which it didn’t bother with before). So, Chapter 2 is both “The Burden of Sovereignty” and “The Weight of the Crown” (drink!), while Chapter 13 is both “The Guardians of the Darkness” and “The Descent into Darkness.” The sole exception is Chapter 17. Wait, what? The outline was for 16 chapters, but it wrote a Chapter 17 anyway, even though Chapter 16 is the actual ending. The chapter titles for Chapter 17 are, for the record, “Echoes of the Past” and—I swear to God—“Title for Last Chapter.” By the way, the prompt specified 18 chapters. Book 3: Not Kismet</h2> For the last experiment, I went bonkers: I asked for a version of my science fiction novel, Kismet</a>. Here’s the prompt: A salvager who lives aboard her spaceship is wrongly accused of stealing valuable data from a wreck. As she tries to clear her name by recovering the data from whoever really stole it, she learns that a terrorist group that hates totemics, humans like herself modified to have animal characteristics, is behind it—the group that killed her mother years before.</li> Writing style: third-person and present tense, using modern language with a literate, stream-of-consciousness vibe</li> Chapters: 15</li> </ul> The real Kismet has more to it—Gail, the protagonist, has to grapple with what “home” and “family” mean to her in a way that she’s been avoiding for decades, and possibly ironically, questions about artificial intelligence surface—but it hits the main plot points and offers enough to work out a thematic hook. Claude wrote this outline with paragraphs rather than bullet points. Here’s the first few: Chapter 1 (10 pages) The story opens with Kira, a talented salvager who lives aboard her spaceship, the Scavenger’s Delight. She is haunted by the memory of her mother’s death at the hands of an anti-totemic terrorist group years ago. As she scavenges through the wreckage of an abandoned ship, Kira discovers a cache of valuable data. But before she can claim it, security forces arrive and accuse her of theft, confiscating the data. Kira protests her innocence, but the authorities refuse to listen. Chapter 2 (10 pages) Dejected, Kira returns to her ship, pondering her next move. She suspects the data was stolen by someone else and planted on her ship to frame her. Determined to clear her name, Kira begins investigating the incident, using her extensive contacts in the salvage community to gather information. She learns that the data contained sensitive information about a classified military project. Chapter 3 (10 pages) Kira’s inquiries lead her to a shady information broker named Xander, who claims to have information about the real thieves. He agrees to share what he knows, but only if Kira helps him with a dangerous job. Reluctantly, she agrees, and the two set out on a mission to infiltrate a heavily guarded research facility. </blockquote> This is already a little weird—she’s accused of theft after finding the data in the wreck, then in the next chapter she thinks it was planted on her ship to frame her. Huh? The outline goes on to have Kira and Xander discover “the Purity Crusaders, a fanatical anti-totemic organization,” run the facility; they plan to use the stolen data to develop “a weapon that can target and disable totemics.” This is not far off from my real story, even getting close to the name I used for the bad guys! That’s about it for similarities, though. They find more conspiracies, Kira “becomes increasingly convinced that the group is responsible for her mother’s death” (which my protagonist knew from the start, because duh), and it gets more and more action-y and Star Wars-y as it goes on. They introduce a hacker and a former member of the Purity Crusaders who helps them. It also keeps going past the final confrontation to have another Scouring of the Shire style denouement. Claude-Author doesn’t do anything with the idea that totemics have animal characteristics, and indeed seems to have forgotten that part of the prompt. Here’s the opening of (sigh) Scavenger’s Crusade: Kira’s fingers glide across the control panel of her ship, the Scavenger’s Delight, as she navigates the treacherous debris field. Her eyes scan the wreckage, searching for any salvageable parts or valuable data that could fetch a decent price on the black market. This is her life – a constant dance with the unknown, a high-stakes game of finding and claiming what others have discarded. As she maneuvers the Scavenger’s Delight through the maze of twisted metal and shattered hulls, Kira can’t help but be reminded of the day her mother died. The memory haunts her, a scar that refuses to heal. She remembers the terror in her mother’s eyes, the desperate pleas for help that went unanswered. Kira had been powerless to save her, and that failure has weighed heavily on her ever since. Shaking off the dark thoughts, Kira focuses on the task at hand. She spots a partially intact data core and guides her ship closer, maneuvering with the precision of a seasoned pilot. Her hands move with practiced ease as she deploys the salvage arm, carefully extracting the valuable component. “Jackpot,” Kira murmurs, a small smile tugging at the corners of her lips. This data core could be worth a small fortune, enough to keep her ship fueled and her supplies stocked for the next few months. She begins the process of securing the salvage, her mind already racing with the possibilities. But just as Kira is about to disengage the salvage arm, a sudden flare of activity on her sensors catches her attention. She tightens her grip on the controls, her heart pounding as she recognizes the telltale signature of security forces. Cursing under her breath, Kira knows she needs to act fast. </blockquote> This is how Kismet opens: The call that shatters Gail’s life comes disguised as a gift from her past. It starts the way all of them have for the last fourteen years: a beedle boop and her ship’s voice, a pleasant female contralto, sounding directly in her left ear. “You have an incoming call.” She’s walking outside, or at least what she thinks of as “outside” here, heading for inside. Kingston’s temperature stays at thirty-one degrees during its day cycle to make it more like its equatorial Earth namesake, but couldn’t they have improved on the climate instead of slavishly emulating it? If she were cisform it might be tolerable, but as a totemic, it’s crazy-making hot. Kis should have said who it was, but it’s got to be Dan again, with something else she can do while she’s still on Kingston. Hey, it’ll be quick and you’ll only need to buy a couple parts and you know I’ll pay you back as soon as I get back on my feet. Right. Dan owes her twenty-four thousand three hundred and counting, and she doesn’t have to pull that number up on her HUD to verify it. It’s burned into her memory right now. So’s another number: the payment she got just this morning from Smith and Sons Salvage, undervaluing her last haul by ninety percent. They’ll drag her appeal out for months, and her budget is already on fumes. She sighs. “Kis, tell Dan I can’t—no. Tell him he can go fuck himself with—” </blockquote> Claude-Author speed runs through its setup. Five paragraphs in, Kira has the data core in her possession with security forces chasing her, while the junk she’s salvaging inexplicably reminds her of the day her mother died. Once more it deviates from both the prompt and its own outline: both specify a wreck, singular, which is not what Claude’s first chapter describes. The security forces scare Kira off from retrieving the “data core”; they never accuse her of theft or confiscate it. In the next chapter, she’s sure the data core is somehow a setup, but that doesn’t fit what Claude actually wrote. It’s as if it’s missing its own story beats, then insisting “no, I didn’t, they were there all along.” But what’s Kira’s personality? What’s her setting like? Her current situation? Five paragraphs in to my actual book, you’ve got hints of that with Gail. The very first sentence foreshadows both the call she’s receiving and what it leads to. You don’t know any of that with Kira. And you never will, because Claude-Author didn’t do any world-building (and if you asked, it wouldn’t follow it anyway). As elf princesses Alara and Elara before her, Kira never shows distinct personality—she’s the default setting from the scrappy action heroine factory. No other characters are totemics. We never learn anything about Kira’s mother, what she believed in, how she died. Kira never truly grapples with her past, because she has no backstory apart from “the bad guys killed her mom.” Beyond that, the generated novel has the same issues The Elven Crown does. The same cliché-fest, the same tell-don’t-show, the same generality over specificity. Phrases, paragraphs, and scenes repeat themselves. Toward the end of the novel, it piles on conspiracies in a way that becomes comical: Talia’s voice crackles over the comm, her tone urgent. “Kira, I’ve found something else. It looks like the Crusaders have been funneling resources and intelligence to a shadowy organization called the Purifiers.” Kira’s brow furrows, her mind racing with the implications. “The Purifiers? Who are they?” (…14 paragraphs later…) Talia’s voice crackles over the comm, her tone urgent. “Kira, I think I’ve found something else. It looks like the Crusaders and the Purifiers have been funneling resources and intelligence to a high-ranking military officer named General Tarkus.” Kira’s brow furrows, her mind racing with the implications. “Tarkus? Wasn’t he one of the key figures behind the development of the weapon we stopped?” (…6 paragraphs later…) Talia’s voice crackles over the comm, her tone urgent. “Kira, I’ve found something else. It looks like the Crusaders and the Purifiers have been funneling resources and intelligence to a shadowy organization called the Synthesis.” Kira’s brow furrows, her mind racing with the implications. “The Synthesis? Who are they?” </blockquote> General Tarkus is never mentioned before or again, unless you count that this scene appears verbatim twice. Was the LLM plotting his return in a scene it failed to write because it ran out of tokens—or was it merely generating words statistically similar to Star Wars? One’s mind races with the implications! Throughout Claude’s generated novel, the battles get bigger, but Kira’s personal stakes never change. She doesn’t grow, she doesn’t learn anything (the names of shadowy organizations do not count for character purposes, sorry), she never screws up in ways that have consequences, she’s never faced with heart-wrenching choices. The cover prompt generated for Scavenger’s Crusade was: The cover should feature Kira, the protagonist, amidst the wreckage of spaceships and debris, conveying the salvage and sci-fi elements of the story. The color palette should be a mix of muted tones and bold highlights, creating a sense of gritty realism and high-stakes adventure. </blockquote> Here’s the generated cover: </a> </figure> This image gets weirder the longer you look at it. I’ll give the “debris” a grudging pass, but why is it on fire? Is it actually her boot on fire? What is she holding in her hand? Why does the space between her legs not match the rest of the background? Does one of those legs disappear below the knee? Does the other leg have two knees? These are the two cover variants for the real Kismet, by the way, both by artist Teagan Gavet</a>. Gavet’s art far more clearly conveys the idea of a wrecked (singular) spaceship, and in the second one, Gail is clearly an animal-human hybrid. (Rat. Thanks for asking.) </a> </figure> </a> </figure> </div> So how good is AI at creative writing?</h2> It’s not. It’s really bad, you guys. People who think AI-generated fiction is “shockingly good,” or even readable, are not people who care about fiction. (This does not imply it’s useless as a creative writing tool, although that comes with caveats. We’ll come back to that.) A few anticipated objections</h3> You should have used GPT-4, not Claude! Claude-Author creator Matt Shumer appears to disagree with you: the scripts he wrote used GPT-4 originally. The repository includes example output from the GPT-4 version</a>, though, and it’s not better. It doesn’t have the repeated scene bug, but its plotting is even less coherent. Seemingly important characters are never returned to. Chapter and even book titles sound almost random: in The Awakening of Eldoria’s Last Legend, no legend is ever described, and the last chapter is “Ascendancy of the Last Arcanum,” a term never used until then. Beyond that, it has all of Claude-Author’s foibles: no character arcs, no themes, no specificity in its imagery, all-you-can-eat clichés. Claude “Haiku” is the smallest Claude model. You should have used a bigger one! This is the model Mr. Shumer chose for Claude-Author, so it’s the one I stuck with. “Sonnet” costs twelve times as much to use as Haiku does, and the best one, “Opus,” is sixty times the cost. I’m not dedicated enough to this experiment to pay that much. I didn’t want this to go completely untested, though, so I used Claude’s chat interface—which uses the Sonnet model—with the same prompts that Claude-Author would use to generate an outline for the elf princess story. It was comparable to Claude’s other outlines, and even reused two chapter titles, “Whispers in the Shadows” and “The Gathering Storm.” So far, not so good. I cajoled it into writing the first three chapters, and hey! Sonnet’s writing is free of the scene-repeating bug, and it’s competent enough you could almost believe it might develop a style. A LaCroix of style, if you will. But it’s not all artificial sunshine and robot roses. The princess is again named Alara, again has emerald eyes, and is again in an action story. The outline isn’t far off from the others it generated, and when it’s different it’s not better. The “accident” is a monster that kills her family on a hunting trip, which wasn’t in the outline, yet the third chapter seems to forget that. Also, the palace has an “infirmary wing,” which seems odd for a fantasy story. It’s as if Claude doesn’t actually understand genre, but knows that statistically speaking, injured people in stories often wake up in infirmary wings! So: technically more readable, but my final grade remains the same. GPT-5, Claude 4, and so on will be even better! Maybe, but we can only judge the AI we have</a>. And as I’ll elaborate on later, these problems aren’t solved by bigger, faster models. A final grade</h3> Writing mechanics: Good! Claude-Author (and even GPT-Author) is decent, and if you’re willing to spend on Claude Sonnet, it might keep you reading a while. Story structure: Bad. The outlines it produces are over-generalized and lack understanding of theme, character, and subtext. And it doesn’t follow them anyway. Even when it does, it does a poor job with rising tension beyond “wait, we’ve found another bad guy who wasn’t in the outline!”, and has no idea what foreshadowing is. Prose style: Barf. Practically every phrase is a cliché, either lifted directly or remixed from text on the web. The dialog from every character is samey-samey, and nobody has a personality defined beyond the prose repeating one or two characteristics it’s locked onto. (Elara, both Alaras, and Kira are so determined. Determined to show you their determination, determinedly. Determined!) Reasoning about machine reasoning</h2> When we ask whether computers can reason, we need to examine what we mean. We’ve been teaching computers how to make inferences and draw conclusions for decades: mortal(X) :- human(X). human(socrates). </code></pre> That’s Prolog</a>, an AI language that dates back to the early 1970s, expressing the syllogism all humans are mortal. Socrates is a human. Can Prolog reason out therefore, Socrates is mortal? If you were at a Prolog prompt, you could query it and get a response: > ?- mortal(socrates). yes </code></pre> Ta-da! LLMs certainly appear to reason the same way that Prolog can. Sure, what’s going on behind the scenes is entirely different, but when the result is the same, should you actually care about the process? Yes. You should. > There is a basket on the table in the living room, and a plum in the basket. > What room is the plum in? In the context of the classic board game 'Clue,' the character Professor Plum can be in any of the rooms on the game board. </code></pre> This silly exchange with ChatGPT (GPT-4) is real, but the catalyst was operator error: the first line didn’t get sent to the LLM. If it had, it’d have given a correct answer. Yet it’s still illustrative that it arbitrarily picked a context for the question and proceeded to answer anyway. This is what AI researchers gently term “hallucinating,” because the more honest description of “bullshitting” does not win you big VC money. Why do LLMs bullshit? Because, to beat on this drum one last time, LLMs are performing lexical analysis and interpolation between words represented as points in vector space</a>. Points that are closer together in that space map to words with greater semantic similarity. The “rules” an LLM follows are vector math operations; the rules Prolog-style systems follow are logic operations. So Prolog “knows” that it can’t answer the question “what room is the plum in” without more information. An LLM doesn’t. The vector math always arrives at an answer. Training the model to say “sorry, I don’t know that” isn’t impossible, but it’s difficult, and compared to the directness of Prolog, unreliable. Lastly, our novel-writing exercise strongly suggests that the longer the LLM input and output are, the more likely they are to break down. The prompts Claude-Author created clearly did guide the output step by step, but Claude couldn’t follow a roadmap that detailed. It became visibly buggy as the prompts hit tens of thousands of words, but it didn’t follow them even in its opening chapters. Yes, I had subjective style complaints, but here, I’m talking about the objective flaws ranging from not following the input prompt correctly to repeating output text multiple times verbatim to producing incomplete sentences. You might object that more computing power naturally solves this. But again, all the Claude 3 models, including Haiku, can handle 200,000 tokens in its context window, and GPT-4 can handle 128,000. At least in theory, these models currently have the power level where they can handle everything the Claude-Author script feeds them. In practice, they clearly aren’t handling it well. Okay, but so what?</h2> Does anyone really expect generative AI systems to write the next Great American Novel? And even if they can’t, that doesn’t mean they won’t be dazzlingly perfect in all sorts of other use cases, right? Well, first off, many AI boosters do expect generative AI systems to get that good. Many apparently believe that they’re already that good; Claude-Author’s creator, Matt Shumer, runs a company, HyperWrite</a>, whose entire premise is that AI can “deliver high-quality writing in less time, from first draft to final edits.” There’s another program called Sudowrite</a> which promises to “write 1,000s of words in your style.” Based on my experimentation, both with the opening of Kismet and with their own provided sample of The Da Vinci Code, it cannot write in either my style or Dan Brown’s. The people who were enthusiastic about Claude-Author’s announcement on X (Twitter) and GitHub were mostly people revved up to have it write novels for them. And there are certainly authors watching what’s happening to the market for commercial art and worrying that AI is coming for them, too. In a recent (January 2024) survey by the Society of Authors</a>, A quarter of illustrators and over a third of translators surveyed said that they’ve lost work to GenAI.</li> A third of illustrators and over 4 in 10 translators say GenAI has decreased the income they’re getting when they do get work.</li> Nearly two-thirds of fiction writers and over half of non-fiction writers believe GenAI is going to hit their income.</li> </ul> At this point, it strikes me as unlikely that generative AI will ever do creative writing well, but that’s slim comfort. I’ve made my living for a decade as a technical writer, and while I don’t think GenAI will be good at that, either, companies will push to treat it as good enough. And people who do not, in fact, care about reading, from solo scramtrepreneurs to bottom-feeding publishers and studios, will push hard for AI-generated shovelware. I’d bet money there are shitty broken books “written” by Claude-Author already filling Amazon’s virtual shelves and sitting in submission queues at unsuspecting major publishers. Beyond that, we have to ask whether there’s any reason to think the quirks and errors we’ve seen here are unique to long-form creative writing. If an LLM can’t follow a novel outline when producing its output, can we expect it to follow anything else comparably complex? If it can’t reliably look at the last five chapters it wrote and produce chapter number six without deviating from the series or producing gibberish, can we expect it to be more reliable with any other kind of serial content? And what happens when the content it’s generating isn’t a novel about a determined elf princess, but an operation manual for dangerous equipment? A set of hiring recommendations? A medical diagnosis? There’s got to be some positives, right?</h3> Eh? I guess? While I don’t love the outline GPT 3.5 created for the detective story in the first part or the outlines Claude-Author made, you could workshop any of them into something better. More generally, LLMs could help with brainstorming, help get you past writer’s block, and help with outlining. Earlier, I mentioned Subtxt with Muse</a>, a specialized writing app that incorporates the Dramatica narrative theory into its training. This use case renders GenAI’s particular failure modes largely inconsequential: Muse analyzes your ideas using Dramatica’s rule-based story form model, which shapes its responses into specific, meaningful guidance. If you’re down with the quirky way Dramatica looks at story structure Dramatica offers a curious mix of deep insight that other story structure approaches lack, clever but debatably actionable hierarchies of story elements, and a weirdly essentialist insistence on calling a character’s problem-solving style their “mental sex.” (and comfortable with Subtxt’s steep subscription cost) it might be for you. And I’d still love that “local author search function” idea I mentioned in the first part. As someone who’s honestly pretty bad at remembering physical details of my characters, I’d love to type what color are Gail’s eyes again? and get the correct answer of green. (This is not a new idea; academic-focused word processor Nota Bene has had Orbis</a>, a “free-form text retrieval system,” for about forty years.) But what I wouldn’t love is a search function which, if it can’t find the answer, confidently tells me Gail’s eyes are blue. Using an LLM as a natural-language front end to an Orbis-like system sounds good, but letting the LLM actually perform the search and return the result, trusting it to return “not found” rather than making an answer up? Not so good. Of course, both of those positive uses come with those other caveats that all the AI skeptics keep harping on, because: The caveats the AI skeptics keep harping on matter</h2> You know the ones: whether the training corpus of all publicly available LLMs use copyrighted work without permission</a> in ways that don’t pass a “fair use” exception; the enormous environmental costs</a> AI is starting to exact as it ramps up; how generative AI is already used to devalue creative work. As Harry “Hbomberguy” Brewis</a> wrote, If people were frank about their use of AI and also if we lived in a post-capitalist utopia, AI would be a very interesting experiment. I would love to see what people do with it. People talk about “prompt-mancy,” where they learn how to write the exact line of text to make something really interesting come out of this machine. I wish we could enjoy it on that level without in the back of our minds going, This is going to put so many people out of a job. So many other people’s work was stolen, unpaid, for it to be able to make this. </blockquote> The idea of using GenAI for brainstorming and outlining is cool! So is the idea of that local author search function. Yet in practice, anything utilizing existing, popular models inherits the ethical concerns and legal liabilities facing those models. Your AI-powered muse is well-trained enough not to spit out near-verbatim paragraphs from Harry Potter and the Billionaire Transphobe when given a sufficiently crafty prompt? Great, but if the courts rule OpenAI’s training doesn’t meet the fair use test and your system is built on top of GPT-4, you still got trouble (with a capital “T” and that rhymes with “C” and that stands for Copyright). Even if the AI companies successfully outwit the legal challenges, the ethical challenges won’t disappear</a>. We do not live in that post-capitalist utopia. As I watch Google, Microsoft, and Apple rush headlong toward AI all the things, I can’t help but notice the other elephant in the room. See it? It looks photorealistic at first glance, until you notice that the background in the space between its legs doesn’t quite match the rest of the background, and its trunk may have two and a half nostrils. Yes, that one. Here’s the thing. Is “this natural language web search makes something up three or four percent of the time” an improvement over original Google? Is asking for a list of citations that you still have to manually check a true time-saver? Is quickly generating code with a hidden bug which will bring down your production server if you don’t catch it actually super helpful? Is the environmental cost worth it? Is the cost to independent creators worth it? And do we actually want systems which, no matter how fast and advanced they get, can definitionally never produce work better than the statistical median to be writing—or even substantively contributing to—our presentations, white papers, documentation, art, music, and, yes, novels? I can only conclude with the immortal words of Claude-Author: the weight of the crown never felt heavier. To support my writing, consider a tip on Ko-fi.com. Judge the AI that exists, not the AI you imagine 2024-05-03T00:00:00+00:00 There’s a maxim I see a lot among tech reviewers, which goes something like this: Buy a product for what it does now, not for what you hope it will do later. The product may never get a major new update at all, and it certainly might not get the update that you want. The iPad, for instance, can do many things, and depending on what you need, it’s possible it could be your only computer. But there are things I do on my computer that the iPad can’t do at all, and other things it can do, but not nearly as well. And yet, many of the folks I hear frequently invoking that maxim when it comes to both hardware and software purchases just chuck it out the window when it comes to generative AI. Of course we want generative AI-powered search engines and text summaries and conversational user interfaces. But we know that right now, GenAI confidently returns bullshit answers to queries occasionally, and it’s more likely to do that the more obscure the information is. We know that right now, that tendency towards “hallucination” means that it’s often not as good at summarizing long, complex documents as an underpaid intern is. (Would you trust GPT-4 to do summaries in legal filings? Medical histories? A presentation your job depends on? Not without reviewing it yourself!) We know that right now, as good as LLMs are at parsing natural language, when they screw up, it’s very difficult to figure out how to correct it. And, of course, we know that right now, processing LLM queries requires both tremendous amounts of energy and tremendous amounts of training data whose legal status is, to be charitable, in a grey area. Yet, somehow, the assumption is that all of this is on the verge of being fixed. Of course we’re going to get all that worked out. The Star Trek computer future is here, baby! This is the biggest advance in computing history since crypto</s> the Internet</s> microprocessors</s> the Difference Engine</s> THE WHEEL! Put AI in everything, everywhere, as fast as possible! Don’t judge AI by what it does now, judge it by what we hope it will do later! Maybe. But what if nondeterministic output is intrinsic to LLMs? What if GPT-5 doesn’t hallucinate less, it just hallucinates faster? What if it turns out that while LLMs are useful for some things—and don’t over-index on my critical tone here; I think they’re already useful for some things—they really aren’t useful for all the things? What if the Rabbit R1 and the Humane AI Pin aren’t just unique flops, but warning signs that LLMs alone are not the way to make better voice assistants? What if it’s going to take new avenues of research to get this all to pan out, fusing LLMs with other branches of AI (rule-based expert systems, other techniques for natural language parsing, and so on), and it takes long years, not mere months, to get there? I’d like the Star Trek computer future, too. But judging by the AI that exists, I don’t think we should be telling everyone it’s just around the corner—or making purchase and business decisions as if it is. To support my writing, consider a tip on Ko-fi.com. A year back in Florida 2023-07-20T11:30:58+00:00 About a year ago, I moved away from the San Francisco Bay Area</a>, back to Tampa Bay, Florida, where I’d lived for (mostly) all my previous life. Florida is not the same place it was when I left. The metros feel more urban, more alive, than I remember. Some of that is undoubtedly on me, on my failure to explore them adequately back in the 1990s. But a lot of what I’ve been finding now simply wasn’t there two decades ago. St. Petersburg now has blocks of walkable downtown, starting from the waterfront museums and moving west through the Edge District, on to Kenwood and Grand Central, where they recently held one of the biggest Pride festivals in the country. Tampa’s downtown no longer feels like they roll up the sidewalks at five (a problem that San Jose struggled to solve for years as well). Just like St. Pete’s Central Avenue reminds me—a little—of K and J Streets in midtown Sacramento, smaller towns like Gulfport and Dunedin remind me—a little—of the smaller walkable towns back in California like Danville, Campbell, and Livermore. Some of the areas that were truly nothing twenty years ago have become, well, something. The town I’ve moved to, Ridge Manor, is an unincorporated area a few miles north of still-tiny Dade City, on a state road that goes straight east-west between I-75 and Orlando. The next “big small town” over, Clermont, has blossomed from a near-abandoned downtown into a genuinely interesting suburb, even if it’s hard to figure out just what it’s a suburb of. Wesley Chapel, about a half-hour south along I-75, is a surprisingly large suburb of Tampa now. A year ago, I wrote that you can find great coffee shops and craft breweries and cocktail bars in any metro area, and that’s true here, too. Dade City itself has a great craft brewery and a solid coffee shop, and there are far more throughout Tampa/St. Pete and Orlando. Great cocktail bars are the hardest to find here, I’ve found, but they are here. Florida is not the same place it was when I left. It was, back then, a relatively purple state overall. There are still Florida liberals and leftists, but the Florida of 2023 is a one-party state. And, not to put too fine a point on it, Florida Republicans lead the charge to make that party indistinguishable from the far-right fascist parties plaguing Europe and Central America. Every day brings a new attack on the rights of people DeSantis and his supporters have identified as The Enemy. Trans people. Queer people. Drag queens. Immigrants. Teachers. Librarians. Disney. A drive around rural Florida a quarter-century ago would have certainly taken you past houses and farms flying confederate battle flags; the state’s panhandle has long been an epicenter for the neo-confederate movement. On a similar drive today, though, the flags are almost exclusively for Trump. And there are many, many flags for Trump. Flags and bumper stickers and banners, and an ugliness I can’t remember seeing in America in my lifetime. When I left Florida, Jeb Bush had just won reelection; I’ve returned to a state where Republicans would consider Jeb too suspiciously liberal to elect him to a municipal utility board. I am not in the same place in Florida as I was when I left. Politically and culturally, I’m more Left Coast than I had been two decades ago, to be sure—but I spent most of my previous Florida years in Tampa or its suburbs, or the wealthy, culturally rich city of Sarasota. Sarasota is now ground zero for not just Florida’s culture wars but all of America’s, as the home of the neofascist Moms of Liberty and epicenter of QAnon conspiracy nonsense. My college, New College, is the one that DeSantis is in the process of transforming from a nationally-recognized liberal arts school into a national laughing stock. As someone who presents as a cishet male, I have little to worry about in most interactions here yet—but that yet slowly gathers weight. I’ve been open about my beliefs, moderately open about my not-so-binary, fairly asexual identity. I write queer, often political, furry fiction under my own name. So far, this has only resulted in lost friendships, but the potential for worse is real. Yet my worries don’t center on me. The majority of my friends are queer, too. Will any trans friend, including my BFF/partner, be safe here even for a visit? They’re certainly not going to move here. More and more, I’m hearing of people moving out. I am not in the same place in Florida as I was when I left. All my adult life, both in California and previously here, I could reach dozens of choices for shopping, eating and drinking in under fifteen minutes; some were just a nice walk away in good weather. But Ridge Manor’s several thousand residents spread out over rural half-acre lots. A few businesses cluster in a couple of strip malls around the I-75 interchange. There’s a grocery store, three or four decent restaurants (and three or four fast food places), so-so Chinese takeout, and a few gas stations. Anything else is twenty minutes away at a minimum. That might not sound like a big deal. It didn’t sound like one to me, either. I’d come home to this house every Christmas from California; I knew where it was. And, I’ve always enjoyed driving. For years, my BFF and I took Saturdays out, exploring towns hours away. How bad could this be? The answer, it turns out, is worse than I thought. In all my adult life, I’ve lived where I could reach dozens of choices for shopping, eating and drinking in under fifteen minutes, often in places where some were just a nice walk away in good weather. Now, hitting even most standard suburban chains is no longer a whim, it’s an excursion. Sometimes I’ve dreamt of living in a cabin in Big Sur. I don’t anymore. I want to be in walking distance of something, a short driving distance of anything. Markets, coffee shops, a neighborhood bar, an ice cream parlor. Ridge Manor is not a place where that’s possible, and despite the construction and development around the area, it never will be. Yes, it will get hundreds of new tract homes, but the people who move in there will find that they, too, are a half-hour away from everything. But do I regret moving? No. I moved to be with my mother, to help take care of her and the house. Our relationship isn’t frictionless, but it’s good, better than many such relationships that I see among my own friends and, for that matter, among hers. I know her better now than I have at any previous point in my life. It’s not just a solid, loving parent-child relationship, it’s a solid, loving friendship. That’s invaluable. I still take Saturdays out, albeit mostly by myself now, and I’ve discovered or re-discovered plenty of cool places, many of which weren’t here before and all which have changed. There are places I could truly feel at home in, if I lived closer to them, and if Florida’s politics ever become less fraught. And if I can still deal with Florida summers. The what-ifs remain, though, no matter how much I try to shunt them away. First what-if: My ability to carve out my own time has been markedly impaired over the last year, from writing to TV watching to reading. Perhaps I am not good at setting boundaries, or perhaps I am just not used to living with someone who wants a lot of attention compared to past, undemanding housemates. Would it have been better to live in the suburbs a half-hour down the road, drive up here a few times a week for dinner, spend the night every other week? I’m doubtful. The connections I’ve been making with my mom couldn’t have been made if we weren’t living together. Beyond that, I wouldn’t be here to be able to help with routine small things, and helping with large ones would be that much more challenging. She’d be markedly lonelier, and despite my penchant for solitude, I would be, too. And there’s the cost of living. Despite the isolation, there are many things to like about this house—it’s on over an acre of wooded land, for a start—but the number one thing is, simply, that it’s fully paid off. A year ago, I wrote, “I won’t miss paying as much in rent share [in California] as I would pay for an entire two-bedroom apartment in Tampa.” That turned out to be optimistic; a decent one-bedroom, not two, apartment in Wesley Chapel would be hundreds more a month than my rent share in Santa Clara was. The median rent in Sacramento is, as of this writing, lower than both Tampa and Orlando. Second what-if: my mother and I could move somewhere else, somewhere that checks off more of my boxes and, ideally, more of hers. She’d like to be closer to amenities, closer to medical care, closer to the water. We’re both concerned about the heat, too. As I write this, Florida swelters in record-breaking heat. The SF Bay Area and Sacramento are at unusual highs, too, but the old “it’s a dry heat” joke hits home. Sacramento’s projected high of 103°F tops our projected 94°, but our heat index hits 116° compared to Sac’s 104°—and our low will be 74° (with a heat index ten degrees higher), whereas Sacramento will make it down to a comparatively arctic 58°. If this is the new normal, it may be untenable for both of us. Housing prices anywhere we’d want to live are likely to be challengingly high even with our resources pooled together, though, and I don’t know what place we’d both agree on. Stay in the state, or leave it? She thinks about going back to Baltimore, where she grew up, or around Asheville, where Floridians seem to be moving to when they want to leave this state. I have no personal affinity for Maryland or North Carolina, though; the places I do have affinity for—most of California and the Pacific Northwest, parts of the Southwest—aren’t places she does. Beyond that, the thought of moving anywhere leads to uncomfortable thoughts of mortality—both my mother’s and my own. When will I find myself living alone once more? Will I want to stay where I’m living then? If it’s still here, still in this house, the answer is likely no. But if my mother and I move to a new place, she’ll push for a bigger house. I doubt I’d want a bigger house by myself, or even with a housemate. (And if it’s in Florida, the current politics all but ensure my trans BFF won’t be that housemate.) Of course, maybe a bigger house still makes financial sense; with luck, having a more expensive house means I get more money if I sell it and do move somewhere else, ultimately. The money isn’t being lost. Objectively, I know that. But I don’t feel it. So, where does this leave me? It leaves me with a loving parent and great finances; it leaves me isolated, frustrated with my inability to manage my own time, wondering why I’m even worse than I used to be at coordinating with friends. It leaves me in a good and bad place. It leaves me in limbo. I’ll check back in after another year. To support my writing, consider a tip on Ko-fi.com. You're So Vain, You Probably Think This App Is About You: On Meta and Mastodon 2023-07-10T21:00:59+00:00 Those of you not plugged into the Mastodon community may not be aware of the predominant reaction to Instagram Threads. This started when it was merely rumored, reaching a crescendo with reports that Meta had been talking to a few of the larger Mastodon instances under NDA, presumably to encourage them not to “defederate” with Threads when it came online. Let me describe that reaction for you, with only mild exaggeration: Meta is coming! If Threads is allowed to become part of the Fediverse, it will destroy it! It will steal your data! It will inject ads onto your timeline! It will corrupt Mastodon into being everything you hate about Facebook and Twitter combined! Let’s stipulate that Meta has a long history of doing demonstrably bad things, and that the argument I’m about to make—that Threads is not what people on Mastodon believe it is—should not be mistaken for an argument that Meta is just here to give everyone free cookies. Daring Fireball’s John Gruber has written extensively about how Facebook wanted NSO spyware to monitor iOS users</a>, produced their own spyware VPN and pushed it within their mobile app</a>, and how Facebook’s “unknowable megascale”</a> created “societal harm…as easy for anyone to see as the respiratory problems caused by smoking.” Threads is a product of that data-tracking, spyware-installing, society-harming Facebook, and it is not joyless unreasonable alarmism to keep that in mind when we evaluate how fun and interesting it otherwise may be. Having said that, Threads is not an attack on Mastodon to subvert it for nefarious purposes. How can I say that so confidently? Because Threads is not a Mastodon instance. It is its own self-contained, centralized social network with plans to let its users follow Mastodon accounts and vice versa. An “instance” in Mastodon parlance is one of the many distributed servers that comprises the network; Mastodon users have accounts on individual instances. Nearly all instances are “federated” with nearly all other instances, e.g., they allow their users to follow one another, but any instance can choose to “defederate” with any other instance.</aside> The difference is not mere semantics. Mastodon doesn’t care what client software you use—or even what server software you use. Threads does. Threads needs you to use their app. It’s baked into the business model. Facebook and Instagram never killed their robust third-party client ecosystem the way Twitter and Reddit recently did, because they never had one. They understood their business model from the get-go. When push comes to shove, Threads is Instagram. That’s how, as of this writing, it already has over 100M accounts created. If you have an Instagram account, you have a Threads account. If you get a Threads account, you get an Instagram account. Threads has zero-effort access to over one and a half billion users who, by definition, tolerate Meta’s privacy policies and Instagram’s monetization strategies. By contrast, Mastodon is maybe two and a half million users on a network explicitly positioned as “social networking that’s not for sale”. The users are much less receptive to monetization strategies. And as Mastodon founder Eugen “Gargron” Rothko notes</a>, the design of the network makes it effectively impossible for Threads to collect personally identifiable information on Mastodon users merely interacting with Threads users. So, on one hand: a billion users who accept Instagram showing them ads, algorithm-jamming their timelines and hoovering up as much personally identifiable information about them as they can. On the other: two or three million users on an explicitly anti-corporate platform engineered to be highly resistant to leaking private data. I dare you to make a convincing business case for Facebook spending a single cent trying to capture a fraction of the second group, when it’s less than a percent the size of the first group. Threads is not now, and never will be, about Mastodon. It’s not about embracing it, extending it, or extinguishing it. It’s not about it at all. So if Threads isn’t trying to overwhelm and destroy Mastodon, why have ActivityPub support at all? Two answers. First, “Look, see? We’re open!” is not only perceived as a great talking point these days, it’s perceived as a regulatory relief valve. Look, see? ActivityPub! We’re open! Second, remember that the business model for Threads is keeping you on Threads. If 95% of your friends are on Threads but 5% are over on that weird Mastodon thing, now you don’t have to use Mastodon to follow them! Just follow them from Threads! Woo! Will Threads be a good Mastodon client? No, but it just has to hit “good enough.” Will any Mastodon client be a good Threads client? Fuck no. They don’t want you accessing Threads from Ivory or Tusky or Elk, they want you accessing it from the Threads app, guaranteed to show you as many ads and gather as much data as possible. The argument Mastodon is collectively mustering against Threads is, at the end of the day, “but Facebook is evil!” Again, no argument. But Mark Zuckerberg is evil in the way of a greedy, privacy-flouting tech bro, not in the way of Sauron (aka Peter Thiel). Not only would the “extinguishing” part of “embracing, extending and extinguishing” Mastodon be extremely difficult at a technical level, the plausible ROI on doing so would be minimal at best—and probably even counterproductive. Yeah, but should people defederate?</h2> The aforementioned John Gruber is bullish on Threads’s chances</a>, and he wrote</a> “Threads is the most fun, most interesting new product of the year” on Mastodon (while taking a swipe with “have fun over here in the library,” as if libraries are terrible sad stern places, a weird dig for a professional writer to make, John). Seriously, while I love the estimable Mr. Gruber’s writing, when I look at Threads what I see is an influencer-infested, brand-driven, algorithmically-jammed-up crapfest. A lot like, well, modern Instagram, without the silver lining of pretty photographs. My point is that Threads and Mastodon are already really different culturally. Even when-slash-if the ActivityPub bridge exists, I don’t think many Threads fans will rush to follow us Mastodon users over here having fun in the library, nor will many Mastodon users be rushing to follow their friends on Threads through the Mastodon client of their choice. I predict the vast majority of people who want to use both networks will maintain separate accounts to do so. Instagram has thousands of content moderators, and while they’re already making decisions that will make everyone mad, they’re clearly making decisions. While I doubt Threads will officially follow the Mastodon Server Covenant</a>, in practice I suspect they’ll be more strict in some respects. Instagram has a puritan streak that Threads will carry through—there’s a non-zero chance that Threads may refuse to federate with your instance because, I don’t know, you allow titties and people who say “fuck”. The chances of Threads becoming a conduit for harassment on Mastodon are slim. Personally, I would federate with Threads in “silence” mode: my instance’s users would be able to follow Threads users and vice versa, but posts from Threads would not show up in any public timelines on my server. I think, though, this should be a choice each instance makes with input from their users, and it is a little dismaying how many instances are perfectly happy making that decision unilaterally. The truly toxic idea, though, is that Mastodon instances should not only refuse to federate with Threads, but they should refuse to federate with other servers that do federate with Threads. In other words, users should be punished for decisions they have no control over and may not even be aware of, made by the administrators of servers they don’t belong to. I am dead serious when I call this toxic. The default position must, must, be that breaking your users’ social graphs is a last resort against clear and present danger. A server explicitly welcomes Nazis, child porn, TERFs, and serial harassers? Block that fucker. But it’s absurd to insist that federating with Meta’s general-interest server presents the same threat level. Look. At the end of the day, I’m a Mastodon partisan. But I don’t love its collective tendency toward self-important dogmatism. I’ve seen more than one friend get set up only to pull back, worrying there are dozens of unwritten rules about content warnings and alt text and linking and boosting they will constantly be put on blast over. I have never seen so many self-identified queer leftists reflexively drop into well, actually mode. New users frequently get stuck on the “pick an instance” part of Mastodon’s signup, and we always say oh, it doesn’t matter that much, which is just not true. Some instances seriously up the unwritten rule count; some suck at moderation, and the admins go tinpot dictator when they’re called on it; smaller ones get their plugs pulled with some regularity. How much worse will it be when hundreds of small-to-medium servers decide they won’t federate with the largest servers—the ones new users who took our “don’t stress about picking your instance” advice ended up on—because those servers have chosen not to block Threads? That level of fracture won’t preserve the Fediverse, it will mortally wound it. The truth is, Threads is not about Mastodon. It’s about Meta and only about Meta, and Mastodon isn’t important enough to them to spend the considerable effort that would be necessary to destroy it. It’d be awfully damn ironic if the Fediverse decides it’s become necessary to destroy itself to stop them. To support my writing, consider a tip on Ko-fi.com. Twitter, failure modes, and your favorite bar 2022-11-22T17:16:46+00:00 So I’ve been seeing arguments for why, no, you should really stay on Twitter, because of the problems with anything vying to replace it. Most circle around what tech people might dub failure modes in terms of both engineering and policy. Make no mistake, many of these are solid arguments. Twitter has, as much as we like to pretend otherwise, gotten many things right. They’ve got fast onboarding. They provide a good experience on both mobile and desktop. (Please don’t @ me with your objections to ads and algorithms and whatever; I’m not saying the UX design on Twitter is perfect or free of dark patterns, I’m saying that it’s been developed by UX professionals over a 15-year period and it shows.) They understand the importance of making a service like theirs accessible. They understand the importance of well-designed terms of service that limit their legal liability without taking draconian stances toward users and their content. These are all failure modes that other, newer, smaller services have done little or nothing to address. But for many people, the real issue isn’t what’s wrong with the other places. It’s that they love this place. Twitter, for all its faults, for all the love/hate relationship you have with it—it’s your favorite bar. This is what most indie creators are feeling, I think. None of the other services have the audience reach; it’s unrealistic to expect us to be on a half-dozen new sites when we could just stay put; and, hey, the likelihood of Twitter really exploding is pretty low. All of those are true, too. The problem, though, is that just because Twitter’s failure mode isn’t likely to be “closing up shop” doesn’t mean it doesn’t have other failure modes. You might have noticed I didn’t mention harassment and toxic behavior as a failure mode—the things a Trust and Safety Team handles—but it is. As Nilay Patel observed</a>, the product of a social network is content moderation. To be clear, this is something all the Not-Twitters are going to have to come to grips with in ways they haven’t yet. Cohost, Hive, and OoobyBloobly (which I just made up, or did I, you’re not sure, are you) look good by comparison because they are a fraction of a fraction of Twitter’s scale. Your favorite Mastodon instance this week is even smaller. With Twitter’s two hundred million users, trying to regulate bad behavior is a 24/7 rearguard action. Well, guess what? Twitter’s Trust and Safety Team is now gone. By deliberate design. It’s not coming back, at least not in any recognizable form, not any time soon. You think I’m going to mention Musk restoring Trump’s Twitter account. I am. But the canary in the coal mine isn’t the who as much as it’s the how. Musk claimed in October that he’d set up a new “council” for moderation, and that “no major content decisions or account reinstatements will happen before the council convenes.” That was a blatant lie. He polled his followers—hardly a statistically unbiased group—about restoring Trump’s account, and has restored others just on his own. Tech journalist Casey Newton</a>: At the risk of stating the obvious, this sort of ad hoc approach to content moderation and community standards is completely unsustainable. It does not scale beyond a handful of the most prominent accounts on the service. And, most worryingly, it is not based on any clear principles: Musk is leading trust and safety at Twitter the same way he is leading product and hiring—by whim. </blockquote> And this is Twitter’s failure mode. All those tweets you’ve seen bitching about how a big problem with Mastodon is that you might choose an “instance” that ends up being run by an anti-woke edgelord tinpot dictator? That’s Twitter now. Oh, you say the need for advertisers will help rein in Musk’s worst impulses, because no sensible advertiser wants to have their “promoted tweets” running in line with alt-right propaganda? Good luck with that: a Twitter that’s only ten or fifteen percent of its original size requires a lot less money to run, and Musk’s been clear he aims to reduce the company’s dependency on advertising income. And those remaining thousand employees or so aren’t going to push back the way we saw happen in some tech companies a year or two ago. The shakeout isn’t just in progress, it’s almost over. The ones left either can’t afford to leave or subscribe to Musk’s worldview. Anyone who joins Twitter under his leadership will have done so knowing what that worldview is. The “liberal bias of big tech” has always been a phantasm. Silicon Valley has always had a strong libertarian bent to it, from the right-of-center Hoover think tank at Stanford University to the military/aerospace roots that long predate the 1990s dotcom boom. While many SV libertarians are socially liberal, not all are, and a few of the most prominent conservatives came out of the “PayPal Mafia”: Musk, the openly anti-democratic Peter Thiel</a>, and VC David Sacks</a>, who co-wrote a book called The Diversity Myth with Thiel a couple of decades ago. Along with professional idiot Jason Calacanis</a>, Sacks now advises Musk on how to run Twitter, and the circumstantial evidence suggests they’ve encouraged the performative cruelty Musk’s exhibited in how he’s run things so far. So here’s the thing. What conservative culture warriors always say they want is the absence of political bias, but time and time again what they mean is bias that explicitly favors them. Everything else, you see, has an innate liberal bias—it’s them against the world, fighting the good fight. They want fairness and balance the way Fox News does. They don’t want an unbiased social media site; what they want is a site with Gab and Parler’s slant, but Twitter’s reach. Now they have it. The product of a social network is content moderation, and Twitter’s new content moderators will be hand-picked by Musk. It’s going to be full of people who won’t object to racism, homophobia, and transphobia as much as object to fighting it, because “free speech”. If you do believe in the Fox News kind of balance, that I’m wrong about Silicon Valley’s political biases and especially wrong about Twitter’s, this isn’t a failure mode. It’s what you want, or at least what you think you want. It’s clearly what Elon Musk thinks he wants. But for Twitter as we knew it, this is a catastrophic failure. It’s a terminal condition, an unrecoverable crash. New Twitter will be hostile to anyone queer, or non-white, or slightly to the left of Ronald Reagan. You may be a creator who wants to stay on Twitter to reach your audience, but the audience there will inevitably tilt toward the anti-woke, All Lives Matter, gender critical, Just Asking Questions crowd. If they’re your audience, congratulations, I guess. If they’re not, you have a problem. I get that, right now, it’s still easy to rationalize staying on Twitter. The alternatives are too confusing, or have questionable terms of service, or don’t have a registered DMCA agent, or have a crappy official app, or have a crappy web interface, or just seem like they’re run out of a college dorm room. We can go down the list and acknowledge most or all of those are great points. But your favorite bar is under new management, and whether you want to admit it or not, you know damn well what kind of bar they’re making it into. You need to think long and hard about whether you’re okay with that. To support my writing, consider a tip on Ko-fi.com. Thoughts on Leaving California 2022-07-15T17:21:37+00:00 I.</h2> I wasn’t born in California. I wasn’t born in Florida, either, even though it was, until 2002, the only place I ever remembered living, the place I would say I was from. I was born in Dallas, but only lived there maybe six months. I think the next place we moved was Albany, New York; I know that’s where we were living a few years later when my parents divorced. When I was around kindergarten age, we moved to the east coast of Florida, and in little more than a year moved to the west coast, to Tampa Bay. I moved out to San Jose, to Silicon Valley, looking for computer work, just after the original dotcom crash. This might have been quixotic, but my technical background was all Unix, and Tampa’s businesses—enterprise back offices, military subcontractors hanging off MacDill AFB—were almost all Windows. I landed in wobbly fashion, doing (of all things) Excel work and then technical writing. Then finally, someone I was applying with for a different position actually read my résumé and realized I was a web developer. To call my career path spotty would be charitable at best. Some years I made less in Silicon Valley than I had in Tampa, although occasionally I’ve made considerably more. When I recently passed my four-year anniversary with my current company, it made this the position I’ve held the longest since the mid-1990s. II.</h2> Moving to the West Coast had been on my mind for years. I always had some reason not to, though. A job, whether or not I liked it. Not enough money in the bank to take that kind of risk. An impoverished roommate I would feel guilty about abandoning, even though I was hardly much less impoverished than he was. A mother who, after her mother passed away, had no local family in Florida besides me. But by the time I moved out here, I had been laid off, so had no job to leave. I had some savings after that long-lasting, high-paying job. (At least, it seemed high-paying to me at the time, which it was by some measures, and the savings seemed like a lot, which I now know it was by very few measures.) I lived alone. And my mother was in a new relationship. A friendly acquaintance had a spare room in his house in San Jose and offered to withhold rent until I could find a job, so I took the leap. When I announced I was moving to California, more than one person reacted like it was some kind of phase, something I needed to do for a few years to get out of my system. Then my wanderlust would be fulfilled, I would move back, and get on with whatever serious business I presumably couldn’t do in Cali. But I didn’t, back then, have any intention of moving back. On one of my first interviews out here, my prospective manager asked me the ultimate interview cliché: “Where do you see yourself in five years?” Somewhat foolishly, I said, “With any luck, living in a cabin closer to Big Sur, writing.” I got the job anyway. I never got the cabin. III.</h2> Every so often, I hear that California doesn’t have seasons, just climate. To someone who grew up in a place with hot summers and snowy winters, maybe. To someone who grew up in a subtropical climate—hot summers and warm winters, differentiated mostly by storm frequency—not at all. The Bay Area moves from dry, bright summers whose highs reach into the nineties and beyond to wet, grey winters whose highs rarely reach the sixties. And it gets snow almost every year, a light dusting on the Santa Cruz and Diablo Range mountains, if only for a few days. And yet, the climate changes across the Bay Area in a way it doesn’t in Florida. From Silicon Valley, it’s less than an hour inland to the Tri-Valley, where the Mediterranean climate takes on a touch of desert aridity and the highs are higher and lows are lower. And, it’s less than an hour to the coast, to Santa Cruz and Half Moon Bay, where moderate is the best description: the change between daily high and low is consistently about twenty degrees, and the change between summer and winter highs (and lows) is maybe about fifteen degrees. These climates are not what I grew up with, but I adjusted to them quickly. Every year I have flown back to Florida for Christmas, and despite it being December it always feels like a sauna when I first step outside. I don’t remember that feeling growing up. Perhaps I’ll readjust to Florida as quickly as I adjusted to California, but it’s hard to shake the feeling that adaptation isn’t reversible. IV.</h2> I don’t have a great affection for wine, but I love wine country. Some of my favorite parts of the Bay Area—the Santa Cruz mountains, Livermore, and of course Sonoma County—are all known for wine, and they’re all beautiful, each in different ways. I’ve picked up some things about wine since being out here, though. It’s hard not to. I’ve also learned about—and learned to love—cocktails, beer, and coffee since I’ve been out here. (I’ve been drinking coffee since I was a child, prescribed it for ADHD back when we called it “hyperactivity.” It took moving out to the Bay Area to appreciate coffee, though.) I was a nascent foodie my last few years in Florida, and that’s only grown in California. It’s hard to pick a single cuisine that somehow defines the Bay Area. There’s good Japanese food, and Thai, and Korean, and Filipino, and lots of good Mexican. People from the northeast seem to think there’s no good Italian, but the reason wine country is wine country is Italian settlers. There’s not just good Italian, there’s great Italian. Whenever someone asserts the Bay Area is rare in this “café culture” of terrific cocktails, beer, wine, food, coffee—cites it as a reason that they don’t want to leave, or a reason that people want to move here—it gets met with pushback, skepticism, or defensiveness. Other places have great versions of all those things, you know, they’ll say. And of course, we know. We know you’re going to be able to find a great coffee shop and a great craft brewery and a great cocktail bar and and and in nearly any metro area, often even in the smallest towns. We also know, though, that there just aren’t that many places in the world where you can find this many in this concentration. And we also know that “how many good coffee shops do you need” is a pretty lame comeback. V.</h2> As my mother got older, being closer—being able to take care of her when needed—felt more important. Her relationship, the one that gave me confidence I wouldn’t be leaving her alone if I lived on the other side of the country, had ended disastrously. In 2016, she had surgery on her carotid artery, and I didn’t go back to care for her, trusting her local friends to do so. That failure ate at me. When I look back at my journal, this is the year I find entries about me looking for my own places in Florida. What I pictured then was a situation like I’d been in before, and like she had been with her mother: living nearby but not together. My grandmother only moved in with my mother during the last year of her life. Before that, they visited every other week, then every other week. If I lived in Tampa and my mother lived in east Hernando County, about fifty miles north, we’d live even closer to one another than she had with her mother. Yet back then, this was abstract. I’d already attached to places here, the land, the cafés, the Pacific Ocean rather than the Gulf of Mexico—and I’d attached to people, too. Friends. A writing group. Even some coworkers; I had one of the best jobs that I’d ever had, not knowing the company wouldn’t survive the year. And an old friend I’d feinted at dating once, failed at, and stayed friends with anyway was becoming—I could say an aromantic partner, and maybe I could just say BFF. Either way, someone I’d rather be able to keep seeing regularly. At the time, I thought it might happen in 2018, after I passed four years with my company and my 401(k) became fully vested. Instead, by 2016 the company had collapsed, and the next job I had in 2017 was short-lived and not particularly enjoyable. I might well have moved back to Florida in 2018, if I hadn’t gotten another job, this time one I did enjoy—and that paid extremely well. So I stayed. And these last four years might well have been the best of my years here. VI.</h2> In 2020, my mother started pushing hard for us to move in together. I had been thinking about my schedule, not hers, and she was in her mid-70s marching toward the late 70s. Instead of pushing for me to move there, she wanted to move out here. This wasn’t the first time the possibility had come up. In 2016, my mother had found a potential home in Gilroy, a town about thirty-five miles south of San Jose. I didn’t like the house, and liked the location even less. In retrospect, Gilroy had more to recommend it than I gave it credit for and the location—on the edge of Silicon Valley—was its best feature, but I didn’t recognize that back then. By 2020, though, housing prices here had made homes even south of San Jose virtually unattainable. We looked at the outskirts of the Bay Area, in Vallejo and Concord, and in Sacramento, a city I’ve come to love over the last decade. Sacramento was more expensive than Tampa, but not by much. But, of course, the pandemic began in 2020, and traveling back and forth house-hunting in the midst of it became unsustainable. So this, too, was put on hold. And in the two years since, housing prices in Sacramento just kept climbing. VII.</h2> It’s tempting to try to find one specific problem to blame for California’s housing costs: it’s restrictive zoning, it’s foreign investors, it’s NIMBYs, it’s absurd tech salaries, it’s Proposition 13. It’s not any one of those things, though. It’s all of those things. And it’s also one other thing. People want to live here, despite the soaring costs, legendarily bad commutes (at least pre-pandemic, although traffic snarls are returning), rampant homelessness, a myriad of other strikes. It’s not just a matter of putting up with the area in exchange for great pay, if they’re getting one of those absurd tech salaries. I’m not going to miss paying over a dollar more per gallon for gas. I won’t miss paying as much in rent share as I would pay for an entire two-bedroom apartment in Tampa. But even twenty years on, there are moments nearly every day when I’m sitting in an outside patio, or walking in a park, or even driving down an interstate through the hills on either side of the Bay and think, I am so lucky to live here. There are other places I could have that feeling about. Other parts of California. Probably most of the Pacific Northwest. I like being near the ocean, even if I don’t visit it as much as I should. I like the moderate climate. I like hills and evergreen trees, sunny summers and grey foggy winters. I like coffee culture. I like the whole region’s mostly progressive politics. And I can’t pretend the politics, both legislative and cultural, don’t matter. My BFF-slash-aro-partner is trans. Most of my friends are LGBTQ. So am I. Words I could use to describe myself—grey-asexual, aromantic, masculine nonbinary—feel very, almost comically, Californian. I don’t think of myself as “closeted,” just quiet. But I’m going to stay a lot quieter after the move. A late friend of mine used to say Florida was paradise, and that he couldn’t imagine why anyone would want to live anywhere else. He and his partner, one of my oldest friends, weren’t particularly quiet or closeted. It was half-joking, suggesting I should move back, but for both of them, Florida really was, and is, paradise. But it’s not my paradise. And I wonder whether, as Florida’s politics shift from purple to red to proto-fascist, it’s safe for that friend there. Or my BFF, who has family in Florida. Or me. VIII.</h2> There’s a voice in my head trying to play therapist. Is your dismay at moving back partially because you feel like you’re returning to watch your mother die? Probably some. But you do want to be with her in her last years. Of course. Yet, you feel like you’re returning under duress. Why? Because it’s happened like the old line about how someone went bankrupt: first gradually, then suddenly. Back in 2020, my flatmate for the last fifteen years started considering where he would go when I moved out, seriously looking at Portland and Seattle; for a variety of reasons, he likes the Pacific Northwest, too. My former housemates, the ones I first lived with when I moved to San Jose, now live on the outskirts of Portland. When I went back to Florida for Christmas in 2021, somehow my moving back in July of this year, now, was treated as a fait accompli by my mother. I don’t remember committing to that during the year, nor do I see any journal entries about having done so—when they touch on moving back, they’re mostly concerns about personal space—yet I don’t remember not committing. Does that make sense? I left that Christmas trip with the feeling of, “So, I guess it’s happening now, then,” not the feeling of, “I am deciding that it is happening now.” At the same time, my flatmate found what seems to be his dream loft in Olympia, and put a deposit down on it before I’d gotten back to California. So then it definitely was happening. I mean, if I’d said “No, I’m not ready, and I’m not moving,” he could have gotten his deposit back, but it felt like the quantum states were collapsing. Your mother asked you to make sure that you had no reservations about moving in with her, but you clearly have them. What was I going to say? When I re-read what she wrote in March 2020, it isn’t the ultimatum it became in my head, but the subtext of “if you say no, I have to make plans to move into a group home” still comes through. And besides, at the end of the day, it’s not moving in with her that’s the issue. It’s leaving California. So why not stick with her moving out here? I don’t know. Scared of the economics. Scared of the recent wildfires and their air quality issues, which mom was extremely put off by—for an asthmatic not too far from turning 80, not without reason. Feeling like somehow this was the path of least resistance. Maybe you need a real therapist. Maybe. IX.</h2> Four days ago, junk haulers came to my apartment and took away furniture I’m not moving with; two days ago, movers came and took away nearly everything else, and I moved into a hotel in San Jose to live out the rest of the week, wrap things up, say goodbyes. My flat now has nothing in it but abandoned cleaning and packing supplies, and some flotsam and jetsam I haven’t disposed of yet. A dish rack. Three pillows. Two days from now, a Sunday, I’ll set off on a cross-country road trip. Sunday night will be spent just outside Palm Springs, on a day when the predicted high hits a balmy 115 °F (46 °C). On Monday, I leave California behind to drive to Tucson, Arizona. I’ll be following I-10, more or less; after Tucson I stop in Van Horn, Texas, then Austin, then New Orleans. From there, I could make the trip all the way to my mother’s house—my house—in a day, but I might stop somewhere around Tallahassee. I’m not sure yet. I’ve joked about retiring and opening a tiki bar somewhere around Tampa Bay, which has a surprising dearth of real ones, serving the complex rum cocktails invented by Trader Vic’s, Don the Beachcomber’s, and other mid-century bartenders. Even though I don’t think I’m serious, I’ve thought enough about it to think what I would (and wouldn’t) do: focus on drink quality, lean on beach culture and the Caribbean for theming rather than Polynesia, try to sidestep the “white dudes do the exotic Orient” issues endemic to tiki. Yet, at least now, as I sit outside a brewery in Santa Cruz, it’s hard not to daydream of moving back even though I haven’t left yet. I don’t know if I’ll be able to afford moving back to the Bay Area on my own after I leave, unless I get another flatmate. But I don’t have to, necessarily. There’s Sacramento. Or Santa Rosa, the most affordable North Bay City. I could even go up the California coast toward Eureka, which is beautiful but maybe too secluded for me. And there’s always Portland or Seattle or Eugene, the whole Pacific Northwest. I have faith I’ll find things I like in Florida: friends, old and new; rediscovering old places that survived all this time; finding interesting new places. If I’m honest, I’ve been scouting over the last few years on my return holiday trips, finding breweries and coffee shops and restaurants and, yes, even a tiki bar, although it’s seventy miles away in St. Petersburg. (St. Pete is perhaps the most interesting city in Tampa Bay these days, manifestly not the case when I lived in that Tampa suburb a quarter-century ago.) But one day, I’ll return. For all the quirks, all the expenses, all the travails I’ve been through these past twenty years, this is my paradise. To support my writing, consider a tip on Ko-fi.com. BBEdit 14, and why you should care 2021-07-19T10:32:00+00:00 When TextMate</a> burst onto the scene in the mid-2000s, it didn’t take aim at Emacs</a> and Vim</a> as much as BBEdit</a>, a Mac-only editor around more than a decade at that point. TextMate offered radically easy ways to create sophisticated new language modules and plugins compared to most editors of the day. Mostly, though, TextMate had Ruby on Rails</a>: David Heinemeier Hansson developed the framework with early versions of the editor, making it almost custom-built for Rails. That gave TextMate a boost working with other server-side frameworks. Since then, cross-platform editors and IDEs like Microsoft’s Visual Studio Code</a> and JetBrains IDEs</a> have come to dominate the coding world. This is an issue for those of us who want Mac-assed Mac apps</a>. I hung onto TextMate and then the native-but-weird Sublime Text</a>, shifting to Code somewhat reluctantly. Last year, I pounced onto Panic’s new Nova</a>, reviewing it positively</a> shortly after release. BBEdit is obviously a Mac-assed Mac app, and for reasons I’ll return to, I came back to it years ago for technical writing—but not for coding. (It sometimes seems like BBEdit’s biggest fans are writers.) Some more code-focused users, though, haven’t looked at it in years. With the release of version 14, should they reconsider? Musty, or battle-tested?</h2> Few editors have been around longer than BBEdit—and few have been as rock-solid. I don’t think I’ve ever lost work. It can even “rescue” never-saved documents you mistakenly close without saving! Yet, that venerable age has become a double-edged sword. “Looks old” is a dismissal I’ve heard a lot. I’ll be honest: to me, it looks like…a text editor. BBEdit 14, editing HTML.</figcaption> </figure> It’s information dense, but neither overly busy nor packed with unnecessary bits. (Nearly everything shown here can be turned off, too.) One thing you don’t see? Tabs. Instead, there’s an open documents list in the sidebar. If you end up with a dozen or more files open at once, this approach starts really showing its advantage. Leaping forward</h2> Here’s something cool: A language server at work.</figcaption> </figure> This editor understands PHP well enough to know that after you type $handler-></code>, it should offer methods from the Handler</code> class as autocomplete options, because it knows $handler</code> is an instance of that class. This level of introspection used to be the exclusive domain of IDEs, but a couple of years ago, Microsoft introduced the Language Server Protocol</a> for Code, so plugins could offer this functionality. Nova supports LSP natively, and with version 14, so does BBEdit: it’s what that screenshot is from. In addition to smart autocompletion, BBEdit uses language servers for: Function parameter help</li> Real-time code linting</a></li> Navigating to function definitions and symbol declarations</li> Reformatting documents</li> </ul> See the green dot in that screenshot? It’s a dropdown for showing errors and warnings in a file. Individual lines also get their line numbers highlighted and the issues shown by underlines. (The snippet above shows $response</code> with a red underline in two occurrences: the first because that line isn’t complete and so has a syntax error on it; the second because, thanks to the first error, $response</code> isn’t defined yet.) BBEdit is preconfigured to use many language servers, like Intelephense</a>, out of the box once they’re installed. This can be considerably more complicated in other editors, especially if their LSP support is itself provided by an extension. A couple of caveats: first, a lot of language servers apparently haven’t been tested with anything but Code, and can get quirky with any other editor. Sometimes BBEdit can work around quirks, but not always. Second, BBEdit doesn’t support all LSP features. For instance, documentation won’t pop up when you hover your pointer over a function or symbol. (Although some might consider that a feature.) Other new features</h2> BBEdit 14 now has a “Notebook.” This takes the already-existing “scratchpad” feature (itself unique to BBEdit) a step farther, storing multiple notes as individual sheets within an always-available Notebook window. By default, BBEdit creates notes as Markdown files, but you can change them to other languages. As with any text file, you can create new notes from the clipboard or from selected text, by dragging text, or even from the shell by piping text to bbedit --note</code>. I haven’t played with the Notebook much yet. It strikes me as the kind of feature you’re either going to rarely use or use all the time, and it’s not clear to me where I’ll fall. I use scratchpads a lot, though, and have a weakness for note-taking apps. I could imagine putting together a package that offered some basic to-do list functionality that might effectively replace TaskPaper</a>, too. It’s not Emacs’s legendary Org mode</a>, but then again, it doesn’t make you learn Emacs. Beyond that, BBEdit now supports Emmet</a> for HTML and CSS expansion if you install the Node Emmet module; if you make from-scratch HTML pages a lot, this is a big deal. If you’re a Python programmer, BBEdit is now aware of Anaconda</a>/Conda</a> virtual environments out of the box, and lets you switch between them with its shell menu. Other old features</h2> When I became a full-time technical writer in 2014, I tried a few different editors and settled on BBEdit, even though its Markdown syntax highlighting is…spartan. Why? Because BBEdit is a Swiss Army knife, a Leatherman multi-tool, for text processing. In no particular order, here’s some interesting things BBEdit does that I rarely see in other editors. What it lacks in Markdown highlighting, it makes up for in Markdown previewing. You can set custom HTML templates, CSS files, and even processing scripts. (You can do this for HTML files, too, and I suspect for other kinds of plain text markup languages.)</li> Many editors have a “fuzzy file open” feature, but BBEdit can open multiple matching files simultaneously. I use this way more often than you might think.</li> BBEdit keeps a history of find/replace searches, and lets you save complex grep patterns with names for easy recall. And you have to see its “Multi-File Find” feature to fully appreciate how great it is.</li> You can build a “text factory” of multiple actions; it’s like having a simple version of Shortcuts or Automator built right into the editor. You can save them as text filters or use them for batch processing. I have a simple-minded Markdown to BBCode conversion “script” I created this way without writing a line of actual shell script.</li> You can “process” lines in a file, searching for duplicates or lines that match specific patterns (including regular expressions), and delete those lines, copy them to the clipboard, or create a new document with them.</li> BBEdit can operate on “columns” of tab-separated values, cutting, copying, pasting, and even rearranging them.</li> While Git support is mostly (ahem) bare bones, it’s fantastic with file-specific commands like diffs and revision history.</li> The “Pattern Playground” is outstanding for constructing complicated regular expressions that work with your documents.</li> BBEdit uses Kapeli’s Dash documentation browser</a> if it’s installed.</li> </ul> The Unix Worksheet in BBEdit deserves its own paragraph. Send any line in the worksheet to your shell by tapping Control-Enter, and the output from the shell appears under that line in the worksheet. You can’t run interactive shell programs this way, but for most commands you now have an editable, modifiable history. In my technical writing, I create lots of local branches; I usually use a worksheet to clean them up by running git branch</code> to get a list of them, then adding git branch -d</code> before each one I no longer need. The worksheet is inspired by the long-defunct MPW</a>, but Emacs fans might consider it a cousin of shell mode. Lastly, another feature of BBEdit worth checking out: the 400-page user manual. Yes, that is what a technical writer would say, but it’s a remarkable boon. A few missing features</h2> BBEdit has a more “batteries included” approach than most editors, and so needs fewer third-party extensions. But, the integrated managers in Code and Nova make a strong argument for a centralized package index. Sublime Text’s Package Control</a> started as a third-party system, so it’d be possible for someone else to take up the mantle here. The closest BBEdit has is BBEdit Extras</a>, and while it has some good stuff, it’s also got outdated (and frustratingly undated!) stuff and link rot. Sublime Text popularized multiple cursors in text editors. BBEdit’s processing commands can do nearly anything these can—honest—but if you’re a multi-cursor junkie, it’ll be an adjustment. Codeless language modules only set colors for strings, comments, keywords and “predefined names,” along with delineating functions for navigation and folding. CLMs also can’t be used for templating systems like Jinja</a> or EEx</a> that exist as “embedded languages” in HTML. You can write a more powerful language definition in Objective-C—I presume including templating languages, since “PHP in HTML” and “Ruby in HTML” are standard—but that’s a hella big ask. Back to the future</h2> I’ve long believed that BBEdit’s balance of text processing power with discoverability and ease of use makes it the best tool for “documentation as code</a>”-style technical writing on the market. But at least for me, it hadn’t kept up with the state of the art for coding. With BBEdit 14, this no longer feels true. Ironic detail: Vim Awesome is built on RethinkDB, a document-store database whose documentation was primarily written (by me) using BBEdit.</aside> So, I’m moving back. I’ve created a new color scheme, SpaceBones</a>, inspired by the default color scheme of Spacemacs</a>. I’ve been working on an up-to-date package for Elixir</a> that comes with an LSP, as well as creating a new set of clippings for PHP 8. My old Editor Actions package may get a reboot. Hopefully, there will be ways to publicize these; I’d love to see a BBEdit-focused project similar to Vim Awesome</a>. If you’re already a BBEdit user, version 14 is an essential update. If you’re not—whether you used it years ago but drifted away, or never used it at all—it’s time to give it another look. BBEdit 14 is $49 new from Bare Bones Software</a>, with upgrade discounts for owners of earlier versions, or from the Mac App Store with a $39/yr annual subscription. You can use BBEdit for free with a more limited feature set. Disclosure: I beta-tested BBEdit 14, and received a free upgrade serial code. To support my writing, consider a tip on Ko-fi.com. Panic’s Nova, ten months in 2021-07-17T14:22:58+00:00 Update, April 2024: While this was all true when I wrote it in mid-2021, Nova has continued progressing, and I’ve actually ended up using it a lot more for web development than I expected to, chiefly because of its integrated workflow/task system—it turns out to be great for static site systems. Beyond that, though, a lot of these complaints remain: The autocompletion is still wonky, the LSP integration still seems undercooked compared to BBEdit’s, and Nova 10’s switch to Tree-sitter</a> for syntax highlighting hasn’t made things appreciably better from a user standpoint yet. I did in fact renew for one year of upgrades starting in 2022, but let that year expire in mid-2023. Yet, I’m using Nova at this very moment, so I’ll probably re-up at some point. TL;DR: I’m not using it much. I’m sure my review of Nova</a> made it clear that I wanted to like this editor a lot. In practice, though, it’s felt more like its predecessor Coda and a similar competitor of Coda’s era, Espresso</a>, than like Visual Studio Code or BBEdit: targeted chiefly at web developers mucking about with static websites. (Which, to be fair, is a sizable audience; my website is static, and Nova’s pretty good with it.) Nova’s built-in smart autocompletion hasn’t proved particularly smart when I’ve been using it, which in some ways makes it more frustrating than not having it at all. Backing it with a language server theoretically improves it, but Nova’s LSP support is fragile and weird. I don’t mind it being somewhat incomplete, but it’s entirely dependent on third parties writing extensions that talk to language servers—which would be fine if they worked. But at least in the languages I use, they don’t. For PHP, there are two wrappers for Intelephense</a>; one just flat out doesn’t start, and the other one makes Nova crash on startup. For Elixir, the elixir-ls</a> extension works in the sense of, you know, not crashing, but mostly what it seems to do is hover huge documentation pages over the screen if the mouse pointer rests over an Elixir keyword. (I have heard Nova’s TypeScript extension is solid, but I haven’t tried it.) Beyond that, when it comes to the basics of just editing, Nova is…fine? It does the job. But—like Coda and Espresso—it doesn’t have the selection and manipulation chops of higher-power editors. I still have no idea what commands like “Select All in Any Scope” are supposed to do because they’ve never worked. Nova’s clips are useful, but limited compared to the equivalents in BBEdit or Code (or any editor, like Code, that basically adapted TextMate’s snippets). So, am I going to renew for the $49 annual subscription? I haven’t entirely ruled it out, but, well, it’s not looking good. I’ve been spending much more time back with BBEdit recently…but that’s another article</a>. To support my writing, consider a tip on Ko-fi.com. A quick unofficial Apple Music Spatial Audio FAQ 2021-06-10T12:33:00+00:00 So, what is it? Music mixed in Dolby Atmos. So, like, surround sound? Yes, with an asterisk we’ll come back to. Most surround systems use multiple channels: the original Dolby Surround used four (left, right, center, and rear), then moved to five (splitting the rear into left rear and right rear), and a few Even More Channels variants. Dolby Atmos, though, doesn’t have channels. Instead, it assigns audio tracks to “audio objects,” which have three-axis positions in virtual space. Each object has metadata that says, “this object should sound like it’s coming from a point this far between the left and right walls, this far between the rear and front walls, and this far between the floor and ceiling.” When you play the Atmos soundtrack, the decoder—called a “renderer” by Dolby—knows how many speakers it has available in the room and their positions within 3-D space, and it maps the audio objects onto specific speakers at specific values. It’s remixing the audio to match the playback environment on the fly, with as high fidelity to the original audio object positions as possible. That’s pretty cool! What about, uh, not movie theaters? The home version of Atmos works essentially the same way, just mapping to fewer speakers—typically just five, although you can get fancy and go up to eleven. Is that a Spinal Tap reference? If you want. So how do you get all that fancy three-dimensional positioning with only two speakers, or a pair of headphones? Congratulations, you’ve found that asterisk I said I was going to come back to! Yay! Explain. For headphones or any other two-channel system, Atmos positional data has to be down-mixed into two-channel audio. If you want to hear Atmos music with the highest fidelity to the recording, you’re going to need to play it over a system with at least four physical speakers. As far as I know, the only way to do that with Apple Music is to use an Apple TV box: it outputs Dolby Atmos over HDMI. So it’s all a lie! Well, not so fast. What Dolby Atmos does for Headphones is render binaural audio. Binaural audio is, per Wikipedia, “a method of recording sound that uses two microphones arranged with the intent to create a 3-D stereo sound sensation for the listener of actually being in the room with the performers or instruments.” Basically, the theory is that we only have two ears, yet we can clearly hear when sounds are up, down, front, or back, not just left or right. In theory, two channels should be enough to capture full positional information if we record the sound properly. So: when there’s only two speakers to output to, the Atmos renderer takes each Atmos audio object and calculates volume and left/right panning settings to synthesize a binaural mix. All the positional information is still used. Is it better stereo, or just different? That’s hard to answer. First off, you may not hear much difference between binaural stereo and normal stereo. I hear what audio nerds call a “wider soundstage,” but I rarely think sounds are in front of me or behind me. Occasionally, I get that “enveloping sound” sensation they talk about, but I can also get that from a good normal stereo recording. In my experience, I’m more likely to get it from a pair of great full-range speakers than from headphones. From what I’ve read, I’m not an outlier here; the truth is that most people just don’t have great speakers. That’s very audiophile snob of you. Sorry. You said “first off.” Second? Second, the original multichannel Atmos mix just might not be as good, at least subjectively, as the stereo mix. Tracks are definitionally going to be placed differently, and you might find some are overemphasized or underemphasized to your tastes. This gets even more complicated if you’re listening to headphones, because you’re relying on Apple’s renderer to synthesize the binaural mix. (Atmos lets music producers include cues on how to make the best binaural version of an Atmos mix, but apparently Apple doesn’t use that, because as we all know, Apple Knows Best.) So is this an Apple exclusive thing? Apple talks about it like they’re doing amazing things nobody else has. Apple talks about everything that way. Atmos Music has been a thing since at least 2017, and TIDAL and Amazon Music already stream it. When Apple says they’ve been working with studios on this, I have a suspicion that what they mean is “Dolby has been working with studios on this.” I haven’t seen any indication yet that Apple’s getting exclusive tracks and deals. Huh. So what does any of this have to do with AirPods “Spatial Audio”? Apple seems to be using that to describe two unrelated things: Their own clever synthesized binaural audio, which only works between iOS and AirPods Pro and Max. This considers your physical position relative to the playback device when it’s calculating the binaural audio effects, making it seem like the “front center channel” is right where the video is playing regardless of your head’s position.</li> Apple branding for Dolby Atmos music.</li> </ul> So (Atmos) Spatial Audio isn’t (Apple) Spatial Audio. Right. Except when it is. Argh. What? If you’re using an Apple TV box, then you get the actual Atmos mix. If you want to listen to the synthesized binaural audio on non-Apple headphones, though, you’ll need to go into the Music app’s settings and set Atmos to “Always On” rather than “Automatic.” Apple Music warns that isn’t “supported” on all speakers, but I would read that as a warning that it may not sound good on all speakers—remember, the Atmos binaural rendering happens in Apple Music itself. It should be fine on any decent set of headphones, though. What about two external speakers? I wouldn’t, but if you want to, go nuts. So is this really the future of music? If anyone can finally make multichannel audio mainstream, it’s Apple, but that’s a big if—the consumer audio market has been rejecting multichannel audio for going on five decades. In its favor: multi-speaker setups have become very common for home theater systems, which they weren’t when SACD and DVD-Audio tried 25 years ago (and quadraphonic sound tried 25 years before that). Against it: a lot of these systems probably won’t actually sound that great with multichannel music. Movies and TV shows mostly just send dialogue to the center speaker and effects to surround speakers, so they don’t have to be “full-range.” But they do with multichannel music. So it’s quite possible that normal stereo tracks will sound better—not “you must be a golden-eared audiophile” better, but “obvious to even the most casual of listeners” better—on most home theater systems. And I’m dubious whether binaural downmixes for headphones will prove sufficiently compelling on their own. (Updated in April 2024) To support my writing, consider a tip on Ko-fi.com. A quick unofficial Apple Music Lossless FAQ 2021-06-08T15:15:00+00:00 So what’s the deal? Apple Music can now stream files as “lossless,” up to 24-bit resolution and a 48 KHz sampling rate (which is better than CD quality), or “hi-res lossless,” up to 24-bit resolution and a 192 KHz sampling rate. Does that really make music sound better? Depends on who you ask and what your equipment is. I feel like I can often hear a difference between CD quality and “lossy” encoding, but not reliably. I don’t think I’ve ever heard a difference between hi-res and CD quality that couldn’t be attributed to remastering, which gets into a whole different subject. You said “feel.” Isn’t that very subjective? Yes. Have you gone through scientifically sound ABX testing, ideally in a soundproofed room and conducted by qualified audio engineers? No. I sense you are not interested in this argument. You are correct. Okay, so let’s talk about Apple Music specifically. Can I get lossless quality over any Apple AirPod? No. There are no lossless Bluetooth codecs. How about the HomePod? Not yet, but it’s promised for later. The Apple TV? Yes, but not at hi-res, and it appears to be locked at a 48 KHz sample rate. (Some lossless music Apple has is likely at a 44.1 KHz sample rate, the same as CDs.) Okay, so it’s best on the Mac. Well, two caveats. One, you’ll probably need an external DAC (digital to analog converter) to get better than “lossless” quality. Two, Apple Music doesn’t do output bit rate switching. What’s that mean? The music files have sample rates they’re encoded with, and your computer sends data to either its internal or your external DAC at a specific sample rate. Ideally, those two rates should be the same, and most “audiophile” music players match them automatically. Apple Music on the Mac doesn’t. It uses whatever bit rate the output device happens to be set at when the Music app launches. If that bit rate doesn’t match the music it’s playing, the music will be resampled to match the output rate. If it’s resampled, is it still lossless? If it’s resampled down, say from 96 KHz to 44.1 KHz, then definitionally, no. If it’s resampled up, say from 44.1 KHz to 48 KHz, the answer is murkier; no data gets lost, but new data has to be synthesized. How do I fix that? A few approaches off the top of my head: Check what you want to play before playing it and set the output rate in Audio MIDI Setup. Music will only show you the sample rate for “hi-res” music, but anything that’s listed as an “Apple Digital Master” or whatever they’re calling it this week is probably 48 KHz. This is arguably bad advice.</li> Enable “lossless” but not “hi-res” in Music, and use Audio MIDI Setup to set the output at the highest bit rate and sample rate your DAC has. This is arguably less bad advice, although it’s still not ideal.</li> Subscribe to TIDAL.</li> Listen on iOS.</li> </ol> Wait, are you saying that iOS does do sample rate switching? Yes. Why doesn’t the Mac? The Music app on the Mac is really still just iTunes. Internally, iTunes is, if you will forgive the technical jargon, a trash fire. You mentioned TIDAL. Does it do sample rate switching on the Mac? Yes. As much as TIDAL gets made fun of, it’s got a decent app, and TIDAL Connect, like Spotify Connect, is arguably better than Apple’s AirPlay for getting home audio devices to stream music. But its lossless tier is $20 a month and its radio stations and curated playlists are noticeably worse than Apple Music’s. What about Spotify and Amazon HD Music? I don’t know. Like Apple Music, though, they’re more “consumer-oriented.” What about Qobuz? Gesundheit. So what are you personally doing? For my desktop, I’m following #2 above. My computer speakers (Vanatoo Transparent One Encores</a>) have a USB DAC built into them. For my living room system, I can use the Apple TV, but compared to Spotify or TIDAL Connect—both of which work natively on my A/V receiver—it’s a little fiddly. I read on a website that the best music quality possible is from MQA, “Master Quality Authenticated,” files. Does Apple Music support that? Please stop reading that website. To support my writing, consider a tip on Ko-fi.com. The Peril of “No Politics” 2021-04-27T21:35:00+00:00 Basecamp is both the name of a small tech company and their primary product, a web-based project management tool that includes forum-like message boards and a Slack-like chat component. It’s pretty good. (So I’ve heard.) In some ways, Basecamp is actually more famous for Ruby on Rails, the web framework they created for Basecamp. And, they’re famous for having capital-O Opinionated leaders, who recently banned “societal and political discussions” on the company Basecamp—essentially the equivalent of saying “no politics on the internal Slack”: Today’s social and political waters are especially choppy. Sensitivities are at 11, and every discussion remotely related to politics, advocacy, or society at large quickly spins away from pleasant. You shouldn’t have to wonder if staying out of it means you’re complicit, or wading into it means you’re a target. These are difficult enough waters to navigate in life, but significantly more so at work. It’s become too much. It’s a major distraction. It saps our energy, and redirects our dialog towards dark places. People can take the conversations with willing co-workers to Signal, Whatsapp, or even a personal Basecamp account, but it can’t happen where the work happens anymore. </blockquote> So. Basecamp’s post has provoked predictable outrage on Twitter, and, well, duh. Twitter is outrage’s natural habitat, where nothing is worth stating if it can’t be stated in the most extreme form possible. But pop quiz: what does “politics at work” mean to you? Facing fraught but important questions about company policies and culture, including pay equity, hiring practices, workplace behavior, and even the ethics of the work being done and for whom.</li> A continual verbal slugfest among coworkers who seem more interested in pwning one another for their terrible viewpoints than coming to any understanding.</li> </ol> It’s clear from the text of their post that Basecamp wants to stave off the latter. And, y’know, that’s not unreasonable. I’ve had coworkers with political views I absolutely didn’t share, and we could still, well, work together. There was no explicit ban on politics; we just understood that it’s not something one gets into with coworkers. The problem, though, is that shutting down the latter all too often means ducking the former. Suppose your company supports a politician pushing policies that would benefit the business directly; aren’t they now indirectly supporting every other policy that politician’s pushing? What if it comes out that one of your customers is a neo-Nazi network? Why does your company have only one woman and zero Blacks in its twenty-person engineering team? Why did that trans customer support engineer quit after only four months? Again, I think—at least, I’d like to—that Basecamp’s intended message was keep company chat channels civil and focused on work. But if that’s what they meant, that’s what they should have said. By saying “no politics,” what they’ve communicated is don’t ask uncomfortable questions about our workplace culture. Maybe they shouldn’t have to wonder if staying out of it makes them complicit, or if wading into it makes them a target. But they’ve tried to have it both ways. That guarantees the answer to both of those questions is yes. Postscript: Literally just after I wrote this, I came across Jane Yang’s open letter to Basecamp’s founders</a>, a brilliant—and depressing—read that makes me rather less sanguine about Basecamp’s intent. To support my writing, consider a tip on Ko-fi.com. The Mac and the iPad aren’t meeting in the middle yet 2021-04-21T12:46:00+00:00 At the end of 2010, John Gruber of Daring Fireball wrote in a Macworld column</a>, The central conceit of the iPad is that it’s a portable computer that does less—and because it does less, what it does do, it does better, more simply, and more elegantly. Apple can only begin phasing out the Mac if and when iOS expands to allow us to do everything we can do on the Mac. It’s the heaviness of the Mac that allows iOS to remain light. </blockquote> Back then he wrote that long-term (“say, ten years out”), iOS might replace macOS. But in 2020, Apple recommitted to the Mac: the Mac Pro, the return of good keyboards, and—the biggest move yet—a new CPU architecture designed in-house. Since then, I’ve seen a chorus of pundits, both professional and armchair (hi), push two theories that are either at odds or entwined, depending on how you look at them: Surely, a dystopian iOS-like future of only sanctioned App Store purchases lies ahead for the Mac. (Let’s call this the “Hacker News bait” narrative.)</li> Surely, the iPad is going to catch up or even surpass the Mac—it already does so many things so well, and it’s only held back from its potential by an OS with artificial limitations.</li> </ul> The Hacker News bait narrative is bullshit. But I’m not sure about that second, sunnier one, either. Apple has been demonstrating a consistent philosophy for over a decade: Macs are general purpose computers.</li> iPads and iPhones are application consoles, analogous to game consoles.</li> </ul> These have been true from the beginning of each platform. Macs have always been general purpose computers, and iPhones and iPads have never been such. There’s no intrinsic reason iOS devices had to be consoles; other smartphones like Windows Mobile and PalmOS phones weren’t. We all know that, but we forget that there’s also no intrinsic reason Macs had to be open. Not only was its direct antecedent the Xerox Star</a> considerably more console-like, so was Jef Raskin’s original concept for the Macintosh, which evolved into the Canon Cat</a>. Yes, if the Mac had been positioned as an appliance the way the Star and the Cat were, it would likely have joined them in obscurity—but we say that now with nearly four decades more “common wisdom” about computers. The Cat wasn’t the only early attempt at an application console; the 1990s saw the Sony eVilla</a> and other “Internet appliances.” Those products didn’t fail because the concept was bad; they failed because the technology to support the concept just wasn’t there yet. A decade later, we had small, lightweight touch screens and widespread high-speed wireless data—and internet appliances became possible. As long as this philosophy on Apple’s part holds—and there’s no evidence that it’s changing—macOS will never be locked down to the degree iOS is, i.e., unable to install non-App Store apps without jailbreaking. But the Venn diagram of “users likely to walk over such a drastic change to the Mac” and “users likely to spend boggling amounts of money on Apple hardware” is close to a perfect circle. Apple would have to not only make up the lost hardware revenue in App Store revenue but beat it. You need 30% of a hell of a lot of apps to make up for a single lost 16-inch MacBook Pro sale, let alone a Mac Pro. Even if it was just four or five percent of users—and I think that’s extremely optimistic—that’s millions of lost unit sales, and likely forgoing entire markets the Mac currently has a meaningful presence in. There’s just no business case for such a move. Beyond that, given all the radical changes Apple made to the Mac in 2020, it feels like that was the “now or never” moment. If M1 Macs and macOS Big Sur didn’t lock us into an App Store-only world, it’s pretty unlikely macOS Pismo Beach or whatever is going to. But that brings us to the second point. Is this the year when the iPad does get to do everything, not just most things, the Mac does? Will we be able to run macOS apps on M1 iPad Pros the way we can run iOS apps on M1 Macs? I do think Final Cut and Logic will come to the iPad eventually in some form. But so far, macOS has remained a general purpose OS, and iOS has remained a console OS—and I don’t think that’s changing soon. I just don’t. I’m doubtful that Apple has any interest in getting an Xcode-like iPadOS development going, and doubtful they plan to “open up” iOS any more than they must for technical, market, or regulatory reasons. Yet on an infinite timescale, this dichotomy can’t hold. It may be a minority of people who truly can’t do their work on the iPad, as opposed to just kvetching that they can’t do it the same way as they do on a Mac or a PC. But that minority is there, it’s sizable, and it matters. (As a technical writer, I’m part of it; no iPad text editor I’ve found comes close to any Mac editor for working on projects with thousands of Markdown files, and I routinely use extensibility provided by Alfred and Keyboard Maestro for my work in ways iPadOS simply doesn’t allow.) So the question is what happens to change that, and when. I’m expecting iPadOS 15 to have some major UI changes, possibly even the first tiling window manager</a> designed for humans. My pie-in-the-sky guess is that a new operating system replacing both macOS and iPadOS is already underway. Its foundations are Swift and SwiftUI, and macOS Big Sur and iPadOS 15 are early bits of scaffolding. The Mac gets lighter; iOS gets heavier. But they’re not meeting in the middle yet. To support my writing, consider a tip on Ko-fi.com. A brief chat about Chuck Wendig, the Internet Archive, and bad information spread in good faith 2020-10-01T19:36:53+00:00 Because I’ve got a bug up my butt about this again, let’s briefly dig into a social media myth that Will Not Die: “Chuck Wendig is suing the Internet Archive!” No. No, he is not. There are two important bits of background here. First, the Internet Archive. If you know them, you probably know them because of the “Wayback Machine” that archives millions of web sites. They do a lot of other archive-ish stuff, though, including collecting and scanning books. A while ago, they decided to create a digital “library” of those books: anyone could “check out” as many copies of those books at one time as the IA had physical copies of. This is more or less the way digital lending works from your local library: they pay for, say, three copies of a given ebook title, and now three library users can “check out” that book at once. Well, that’s the “more” part of “more or less”; the “less” part is that the IA was doing that with physical books and technically lending digital copies is not the same thing under copyright law. Even so, publishers mostly looked the other way. Until. At the start of the Great Pandemic, the IA decided they were now running the “National Emergency Library” and lifted the per-copy limit. If they had ten copies or a book or two or one, it didn’t matter, however many people wanted to check out a copy at once could. And the IA sent out press releases about this. They wanted everybody to know! I’m not going to argue about the ethics of modern copyright law, but as a legal matter, this is not a gray area, kids. It just isn’t. The Internet Archive was all but sending out notarized letters to publishers saying “we dare you jerky jerks to come after us with everything you’ve got,” and golly gee, they got sued by the Authors’ Guild and several publishers. Who could possibly have predicted that outcome other than, you know, fucking everyone. You will notice, perhaps, that the IA was not sued by individual authors over this. They were sued by publishers and a writing guild. Second, Chuck Wendig. Wendig is a science fiction, horror-ish author who runs a popular blog and has a freewheeling, gonzo, over-the-top style—I’d argue more in his non-fiction than his fiction—that, well, you could call polarizing. (I enjoy it, most of the time, but I could see how many might be driven far away at high speed.) He also wrote a couple Star Wars novels, famously introducing the saga’s first major gay character in Star Wars: Aftermath. And this was not popular with a predictable loud subset of reactionary fans</a>, who carried a hate-on for Wendig that culminated in the trolls getting him fired from Marvel’s “Shadow of Vader” comic book</a>, ostensibly because of his “vulgarity” in expressing what Quartz calls, with delightful understatement, “his unabashedly left-wing political views.” So if Wendig didn’t sue the IA over the Emergency Library, how did he get involved in all this? Well, he called it a “pirate site,” which he pretty quickly apologized for, but also wrote a much longer statement</a> on the subject. (He’s also deleted it since, probably due to ongoing harassment.) The problem with bypassing copyright and disrupting the chain of royalties that lead from books to authors is that it endangers our ability to continue to produce art—and though we are all in the midst of a crisis, most artists are on the razor’s edge in terms of being able to support themselves. Artists get no safety net. We don’t get unemployment and aren’t likely to be able to participate in any worker bailouts. Health insurance alone is a gutpunch cost, not to mention the healthcare costs that insurance wouldn’t even cover. I’m lucky enough (currently, at least), that I can weather a bit of that storm more easily, but most can’t, particularly young authors, debut authors, and marginalized authors who are already fighting for a seat at the table. I’m also not alone in calling this site out—others like Alexander Chee, NK Jemisin, Neil Gaiman, and Seanan McGuire have noted their concerns over this. I am all for access to information and entertainment, and remind folks that libraries here already allow you to take out e-books, even while their brick-and-mortar locations are closed. I used to work for a library system here in Pennsylvania, and libraries all around the country deserve their time to shine in this crisis, as we realize what vital institutions they are, both intellectually and as a service to the community. </blockquote> Come on, how could anyone read that and, in anything even approaching good faith, take major offense at it? This is empathetic to authors and libraries. Yes, it’s (gasp) making a claim that copyright does have value, and maybe you don’t see that. But I hope you at least see why a lot of authors feel they should be the ones to make the choice about how their books get distributed. I’m not against giving my own work away for free, but I am against you telling me that you’re going to give my work away for free and I have no choice in the matter. In fact, I don’t think the people who started this “Wendig sues the IA, film at 11” bullshit did so in good faith. I think many people spreading it are doing it in good faith, but bluntly, I think they’re being used by trolls relying on it being way easier to click “like” or “retweet” than to do fact-checking. (Frankly, I despair at how often I see left-leaning friends gleefully retweeting the most dubious shit that confirms their biases, but that’s a bridge I won’t burn today.) While this whole nonsense is months old, I’m seeing another new thread floating around today fisking an older book of writing advice from Wendig, inviting us all to mock how weird and bad his writing is and how awful his advice must be and oh yes remember he sued the Internet Archive!, and I’m out of patience nuggets for this one. If that’s your image of Chuck Wendig and what he’s like and what he writes, let me offer a different one, from “Follow the River, No Matter Its Rapids, No Matter Its Turns</a>”: It’s a lot right now. I think if we can agree on anything, anything at all between us, it’s that everything is a whole lot. It’s too much. If you’re not screaming into a couch cushion soaked with gin right now, who even are you? But here’s what I’m thinking. I’m thinking all of this is a river. It’s a dark, fast river. It crawls serpentine through the earth, through the forests. Sometimes it moves slow, other times it’s all rapids. Sometimes it is eerily serene, and sometimes it’s rough enough to knock your teeth into your knees and draw blood. It’s waterfalls and eddies, it’s deep and it’s cold. Like all rivers, it can soothe you, and it can betray you. This river, the river we’re in and on now—it’s harder, meaner, a river after a flood, a river whose waters are not sated, who will not abate. It’s mudded up and frothing like the muzzle of a rabid wolf. You can fight against that river. We often do, in writing. We often go against our own moods, against the news of the world, against bad reviews and against poisoned thinking. Our work is often an act of anchoring our boots against the soft slick weeds and the water-smoothed stones and move against the current. Upstream, stories can be born. Sometimes, though, I think you gotta do the other thing. Sometimes, you go the other way. You go with the flow. You run with the river, not against it. And what that means, practically speaking, is you let it happen. What you’re feeling, what you’re seeing, sometimes those elements demand to be seen in the work. Sometimes the river is the channel that feeds the narrative sea, and that means you need to put it in there, out there, all over it. You don’t escape. You confront. You ride the turns, you rough out the rapids, you take all your fear and your anger and your confusion and you put it on the page. And not even in a way of trying to write something that’s marketable or sellable—but just trying to speak honestly about who you are, about the world in which we’re living, and about your grappling with all of it. It’s not even about writing a cogent book or a collective piece. It can be about taking the time to punch that keyboard and scream onto the page—if only to clear the water and find time to climb back onto shore to write something else. It can be the thing you’re writing, or it can be a way to get to the thing you’re writing. I don’t mean to suggest this as good “advice”—it’s certainly no requirement. You have to do what feels best and right—and, further, what feels most productive in the direction you need to be going. I’m only saying that, if it’s that much of a slog, if the slow churning march upriver and against the current feels like you’re fighting too hard and losing to the pressure, turn around and go the other way. Sometimes we want to, even need to, write about what’s going on inside our heads and our hearts. Sometimes we can’t ignore the room on fire. Sometimes we can’t get out of the river or go against it. And in those cases, let the waters take you. Write what needs to be written. Write what the river tells you to write. Follow the water, and see where you go. </blockquote> You may still hate that writing, but if you do, who even are you? To support my writing, consider a tip on Ko-fi.com. Panic's Nova text editor (a review) 2020-09-17T13:01:20+00:00 Panic</a>, the long-established makers of Mac utility software, seems fully aware that introducing a new, commercial code editor in 2020 is a quixotic proposition. Is there enough of an advantage to a native editor over both old school cross-platform editors like Emacs and explosively popular new editors like Visual Studio Code</a> to persuade people to switch? I’m an unusual case as far as text editor users go: my primary job is technical writing, and the last three jobs that I’ve worked at have a “docs as code” approach, where we write documentation in Markdown and manage it under version control just like source code. The editor that works best for me in tech writing is the venerable BBEdit</a>. When it comes to editing code, though, BBEdit lags behind. My suspicion is that BBEdit’s lack of an integrated package manager has hurt it here. Also, BBEdit’s language modules don’t support extending one another, making it effectively impossible to do full highlighting for a templating language like JSX</a> or Jinja</a>. When I was a web programmer, I was one of many who moved to TextMate</a>, and used it for everything for a while. When the Godot-like wait for TextMate 2.0 became unbearable, I wandered the text editing wilderness, eventually splitting my loyalties between BBEdit, Sublime Text</a>, and more recently VS Code. At this point, I suspect nothing will pull me away from BBEdit for technical writing, but for programming I’m open to persuasion. So: meet Nova.</h2> I’ve been using Nova off and on in beta for months. I’ve reported some bugs, although I may mention a couple here that I didn’t catch until after 1.0’s release. And, I’m going to compare it to the GUI editors that I’ve been using recently: BBEdit, Sublime Text, and VS Code. Nova is a pretty editor, as far as such things go, and with files of relatively reasonable size it’s fast. With stupid huge files its performance drops noticeably, though. This isn’t just the ridiculous 109MB, nearly 450,000-line SQL file I threw at it once, it’s also with a merely 2MB, 50,000-line SQL file, and Nova’s offer to turn off syntax highlighting in both files didn’t help it much. This may sound like a silly test, but in my day job I’m occasionally stuck editing an 80,000-line JSON file by hand (don’t ask). This is something BBEdit and VS Code can do without complaint. Panic wrote their own text editing engine for Nova, which is brave, but it needs more tuning for pathological cases like these. They may not come up often, but almost every programmer has one stupid huge file to deal with. Nova has an integrated terminal and an integrated SSH client, and even an integrated file transfer system based on Panic’s Transmit</a>. In fact, if you have Transmit and use Panic Sync</a>, it knows all of those servers out of the box. Nova has a task workflow system</a> for automating building and running. You can associated servers, tasks, and more with individual projects; Nova’s project settings are considerably more comprehensive than I’ve seen in other editors. You can even set up remote tasks</a>. Nova has a serviceable Git client built in, too. Like VS Code, Nova uses JavaScript for its extension API, and it has built-in Language Server Protocol</a> support—it’s a superbly solid foundation. Beyond that, some smaller features have become table stakes for modern GUI editors, and Nova handles them with aplomb. “Open Quickly” can jump to any file in the open project, as well as search by symbols or just symbols in currently open files; it has a command palette; you can comprehensively edit keybindings. It has multiple cursor support for those of us who like that, and a “mini map” view for those of you who like that, although know that you are wrong. Nova’s selection features include “Select all in scope” and “Select all between brackets,” a command I often use in BBEdit and miss dearly in Code. (Both Nova and BBEdit select between brackets and braces, although BBEdit also selects between parentheses.) This effectively becomes “Select between tags” in HTML, a nice touch. There are a few other commands like “Select all in function” and “Select all in scope” that I didn’t have any luck in making work at all; a little more documentation would be nice. That’s worth an aside. Panic has created a “library” of tech note-style articles about Nova sorted by publication date rather than an actual manual, and it’s not always easy to find the information you want in it. I know this is just what a technical writer would say, but I’d dearly like to see a human-organized table of contents starting with the editor basics and moving to advanced topics like version control, server publishing and extension authoring. The Zen of Language Servers</h3> A lot of Visual Studio Code’s smarts depend on the implementation of a “language server” behind the scenes: language servers offer almost spookily intelligent completion. For instance, take this PHP snippet: if ($allowed) { $response = new Response(405); $response-> </code></pre> If you have the Intelephense</a> PHP language server plugin, Code understands that $response</code> is an instance of Response</code> and, after you type the ></code> above, offers completions of method names from the Response</code> class. Right now, Nova’s mostly limited to the language servers</del> completion systems Panic provides, and they’re… not always so smart. In that snippet above, Nova starts by offering completions of, apparently, everything in the open project, starting with the variables. If I type “s,” it narrows things down to methods that begin with “s,” but it’s all methods that start with “s” rather than just the methods from Response</code>. The “Jump to Definition” command shows a similar lack of context; if I highlight a method name that’s defined in multiple places, Nova shows me a popup menu and prompts me to choose which one to jump to, rather than introspecting the code to make that decision itself. But, this is a solvable problem: there’s (I think) no reason someone couldn’t write an Inteliphense plugin for Nova. If Nova’s ecosystem takes off, it could be pretty formidable pretty quickly. Walk like a Mac</h2> Even so, LSP support isn’t Panic’s biggest selling point. Unlike Sublime Text or VS Code, Nova isn’t cross-platform: it’s a Mac-only program written to core platform APIs. Is that still a huge draw in 2020? (Is it instead a drawback?) You can definitely see a difference between Nova and BBEdit on one side and Sublime and Code on the other in terms of resource usage. With the two Ruby files shown in the screenshot above loaded, I get: VS Code: 355 MB, 6 processes</li> Sublime Text: 338 MB, 2 processes</li> Nova: 101 MB, 2 processes</li> BBEdit: 97 MB, 1 process</li> </ul> Code is an Electron-based program, although Microsoft famously puts a lot of effort into making it not feel like the black hole a lot of Electron-based apps are. Sublime uses its own proprietary cross-platform framework. In fairness, while us nerds like to harp on resource usage a lot, if your computer’s got 16G or more of RAM in it, this probably isn’t a big deal. You notice Nova’s essential Mac-ness in other ways. Its preference pane is, like BBEdit’s, an actual preference pane, instead of opening in another tab like Code or just opening a JSON file in a new tab (!) like Sublime. And while all editors better have first-class keyboard support—and Nova does—a good Mac editor should have first-class mouse support, too, and it does. You notice that in the drag-and-drop support for creating new tabs and splits. Nova’s sidebar is also highly customizable, possibly more so than any editor I’ve regularly used. (Yes, Emacs fans, I know you can write all of Nova in Lisp if you want. When one of you does that, please get back to me.) Unlike BBEdit, though, Nova doesn’t have a Mac-like title bar, or a Mac-like outline view of the project files, or Mac-like tabs. (Well, BBEdit doesn’t have tabs at all, which turns out to be a great UI decision once you have a dozen or more files open, but never mind.) This isn’t necessarily bad; people often say BBEdit “looks old,” and it’s hard not to suspect that what people mean by that—whether or not they know it—is that it looks like the long-established Mac program it is. Nova is relying less on “we have a Mac UI and the other guys don’t” than on “we have Panic’s designers and the other guys don’t.” Make no mistake, having Panic’s designers counts for a lot. What may be more disappointing to old school Mac nerds is AppleScript support: none whatsoever. It doesn’t even have a vestigial script dictionary. Again, this may not be something most people care much about; personally, I hate having to write AppleScript. But I love being able to write AppleScript. BBEdit’s extensive scriptability is one of its hidden strengths. Nova’s Node-based JavaScript engine is probably more powerful for its own extensions and certainly more accessible to anyone under the age of 50, but it may be hard to call it from external programs. So is it worth it?</h2> That probably depends on where you’re coming from. If you loved—or still use—Panic’s older editor, Coda, this is a no-brainer upgrade. If you used Espresso</a>, a Coda-ish editor that always seemed to be on the verge of greatness without ever reaching it, Nova may also be a no-brainer for you. If you’re a fan of Sublime Text, BBEdit, TextMate, or another editor that doesn’t have native Language Server Protocol support, you should definitely try Nova. Sublime and TextMate have more plugins (especially Sublime), but many extensions seem to be languishing (especially TextMate). BBEdit never had a great extension ecosystem to start with. All of these editors have strengths Nova doesn’t, but the reverse is also true, and Nova may catch up. If you’re an Emacs or Vim power user, we both know you’re just reading this out of academic interest and you’re not going to switch. C’mon. If you use Visual Studio Code, though, it’s way tougher to make the case for Nova. Code has a vastly larger extension library. It has the best support for LSP of any editor out there (LSP was developed for Code). Despite being Electron-based, it’s pretty high-performance. Code doesn’t have an integrated SSH or FTP client, but it does have an integrated terminal and task runner and Git client. If you don’t object to using an editor that isn’t a “perfect fit” with the Mac UI, Code is very, very good… and it’s free. I don’t object to Nova’s pricing model—$99 up front including a year of updates, $49 for future years of updates—but I can’t help but wonder if Panic should have gone with super aggressive introductory pricing. Also, I saw more than a few suggestions on Hacker News about how there should be a Code-to-Nova extension translator; I’m not sure automatic conversion would be practical, but a guide on manual conversion seems like an excellent idea. For my day job of technical writing, I’m going to stick to BBEdit. (One day I’ll write up an article about why I think it’s the best “documentation as code” editor on the market.) For programming and web editing, when I was working on both a Ruby and a PHP project—the former a Rails learning exercise, the latter an obstinate “I am going to write a modern PHP app without using a framework” exercise—I kept trying Nova’s betas and then switching back to Code for Inteliphense and, I swear to God, MacVim</a> for Tim Pope’s amazing rails.vim</a> plugin. I suspect Nova could duplicate both of those, but I’m not sure I want to be the one to do it. (Also, while Panic has decent reference documentation</a> for writing extensions, I’d like to see a few simple end-to-end walkthroughs for those of us who look at a huge list of reference topics and don’t know where to start.) But Nova isn’t just pretty, it’s powerful, and has a lot of promise. The editors I’ve been comparing it to have been around since 2015 for VS Code, 2008 for Sublime Text, and 1992 (!) for BBEdit; it’s not reasonable to expect Nova to blow past them in every respect right out of the starting gate. Even so, they are Nova’s competition. Catching up fast is an essential requirement. So: yes, I’ve bought Nova, and I’m rooting for Panic here. I’ll come back in a year and report if I’m willing to stay on the update train. To support my writing, consider a tip on Ko-fi.com. Signs of the old Apple 2020-03-20T12:07:53+00:00 For a long time, there’s been two competing narratives about Apple’s pricing: They’re “premium”: sure, they’re expensive, and yes, you pay for the brand name. But when you compare them with products of equal quality, counting not just feature specs but design, materials, and build quality—so an iPhone with a Galaxy S, a MacBook with a Dell XPS—they’re rarely unreasonably higher, and the user experience of macOS and iOS is arguably worth it. </li> They’re “luxury”: there’s nothing about a Mac, an iPhone, or an iPad that truly makes it better than its competition; the only reason anyone buys Apple products is for the supposed status of owning them and showing them off. </li> </ol> For most of my Apple-using life, I’ve unsurprisingly been in the first camp. But for the past, oh, let’s call it five years, Apple has been making this…difficult. Phone prices have climbed faster than their sizes, especially at the high end;</li> Their vaunted hardware engineering and design teams have become tethered to form over function, leading most infamously to the “butterfly” keyboard that prioritized thinness at the expense of not only typing feel but reliability, but also to the Touch Bar (a solution in search of a problem) and the paucity of ports;</li> Laptop prices have also climbed fast, possibly because prioritizing form over function turns out to be expensive;</li> While iOS 7 made significant UX changes, those changes sometimes came at the expense of clarity (e.g., little to no visual cues as to where tap targets are), and later development seems fairly stagnant;</li> While Apple has always had a “we know what’s best for our users” attitude and has never been super chatty about their decisions, they’ve just made themselves look arrogant and unresponsive a few times, most infamously over battery management.</li> </ul> But the two products announced this week show off two old-school Apple moves we haven’t seen in, well, a while. First, the new MacBook Air. They’re bringing the Magic Keyboard to it, as we all assumed (hoped?) they would, fixing the biggest problem with their laptops of the last few years. They’ve upped the starting SSD from 128G to a more respectable 256G, and upgraded to a modern Intel processor that can turboboost up to twice as fast as last year’s Air—and that’s just the base processor: you can get up to a quad-core i7. And they lowered the price. For the last few years Apple’s been far more likely (especially with Macs but arguably with the iPhone, too) to improve a product some ways while making it worse in others and raising the price, playing right into that “luxury” narrative. But this wasn’t always Apple’s modus operandi. For many years, their classic move was upgrading a product while keeping the same price point: you still paid $1499, but now the CPU was faster, you had more RAM, you had a bigger hard drive, and so on. But the new Air is back down to the semi-magic $999 point (and Apple loudly trumpets that it’s $899 with the education discount). To be fair, we’ve been seeing more of this from Apple in the last couple of years—there was a modest price drop moving from the iPhone 8 to the XR, and while high-end iPads climb to nosebleed pricing territory, the low-end model has gained power while dropping in price—but there was a multiyear stretch when “get more for the same money” became “get mildly aggrieved for another couple hundred.” Second, the new iPad Pro—but especially the new iPad Magic Keyboard, which has a trackpad, a USB pass-through port, an amazing cantilever design like no other tablet keyboard case I’ve ever seen, and a wheeze-inducing price point of $299–349 depending on size. This is also a classic Apple move: the damn thing is so expensive you can’t help but think of it as overpriced, but there’s literally no competition for it. You are not going to be able to get a keyboard case that nice from anyone else. (Amazon will fill up with knockoff versions for $99 or less, and they will all be pretty bad by comparison, because that’s the way of things.) Personally, I’ve grown frustrated with the iPad’s software limitations at both the OS and app level. I wrote about that in “The iPad needs more focus on the little things</a>” last August, which concluded with, Don’t get me wrong: I’m not giving up my iPad. Who knows what iPadOS 14 will bring? But in the meantime, I confess I’m watching what happens with the next MacBook Air revision pretty closely. </blockquote> Well, that update is here, and frankly, they knocked it out of the park. It’s pretty much everything I was waiting for. And yet, if the iPad Magic Keyboard lives up to its promise, it’s the keyboard case for the iPad that I’ve always wanted. I still do love the iPad, and the Magic Keyboard might make it the best portable computing device for me ever—in terms of hardware. Unfortunately, hardware won’t fix the frustrations I have with the platform. So now, I suppose I’m watching what happens with the next major iPadOS revision pretty closely. For the time being, though? The portable computer for me—and, I think, for anyone who wants the best laptop—sure seems like it’s the new Air. To support my writing, consider a tip on Ko-fi.com. Notes from the Road 2019-10-13T19:54:07+00:00 For a while about a decade ago, I was doing tech blogging; I never rose to “A-list blogger” level, but I probably hit B-list for a couple years, based on who was linking to me and the “yeah, I think I’ve heard of Coyote Tracks” comments from bigger names. I used to joke that while you weren’t reading me, people you were reading were reading me. I never figured out how to monetize those eyeballs, as they say, I suspect because I worried I wasn’t big enough to go after any real ad networks. The blog did lead to my tech writing job at RethinkDB and a middle-age shift into that career path, though, so it paid off well enough. But I’d been “blogging” since 1998, even though nobody had that term back then. My first try was for some kind of online column, as I recall, just personal essays; I doubt anyone read it, and I didn’t keep up with it. For a decade or so after that, though, I kept a journal called Coyote Cartography, the name I resurrected for this blog. That started up while I still lived in Florida, where I’d taken to driving long distances for no reason other than to do it. I used to joke about practicing “driving zen” instead of walking zen</a>, but it wasn’t entirely a joke. I did a lot of thinking when I drove, and once I got out of traffic and found myself on relatively rural roads, it brought a measure of peace. I kept doing that occasionally after I moved to California, at least for a while. Occasionally, I’d write up wherever I’d ended up and post it in the journal with the title “Notes from the Road.” Geography makes it harder than I’d expected to drive out of the San Francisco Bay area and get somewhere new and interesting than it is to drive out of the Tampa Bay area and get somewhere new and interesting. Also, I’m old enough now that driving ten hours for no particular reason can be more tiring than relaxing. Over the last few years, it’s been rare for me to go much farther than Sacramento, even as I’ve had the resources to take “staycations” as a way of recharging. (Those staycations tend to be in the Sacramento area, as I’ve come to love it despite its brutal dry heat summers—and also, I confess, because the hotels there are so much cheaper than anywhere near San Francisco.) But a couple weeks ago, I’d started to feel unaccountably melancholy and maudlin, a kind of free-floating mild depression I’d become familiar with off and on. In the past, I’ve self-medicated with St. Johns Wort, which I suspect may be as much placebo as anything else; while I’ve grown to like beer, wine and spirits more now than I did when I lived in Florida two decades ago, that’s a dangerous road to go down. So I decided to take to the literal road, heading back to Big Sur for the first time in years. Last weekend, I ended up in Modesto after a slightly too late start, so I didn’t see much of Modesto. (I don’t know if there’s anything to see there, but sometimes overlooked towns can surprise you.) And, as I type this, I’m in Ukiah, at Black Oak Coffee Roasters. Ukiah</a> is the largest city in Mendocino County, which is still only about 17,000 people. Like a lot of Mendocino, there’s a mix of old school agriculture and hipster counterculture around the town. This is a coffee shop you’re going to go into and find great local-roasted coffee and people working on laptops; it’s not going to be abuzz with conversations about social media startups. It feels kinda Bay Area in a good way, but also feels fully of its place. San Francisco insists it’s “northern California,” but a glance at the map tells you it’s clearly in the middle. This is northern California, and it’s a noticably different pace. You start feeling it at the northern edge of the Bay Area, in Sebastopol and Santa Rosa and Healdsburg, although those are still close enough to San Francisco to make everything just a little more fraught in both good and bad ways. Going north of Healdsburg requires effort. It rewards you, as long as you like the kind of place Ukiah is. I do, which I’d almost forgotten. It all came back instantly when I turned down its main street. (There is a Main Street, but the real main street is State Street, one block over.) It’s the sort of town I’d love to live in, if only I could install a teleporter. For one, my job isn’t keen on full-time remote employees. For another, as solitary as I can be, I’m not antisocial—I’m bad at being social, but that’s not the same thing. I wouldn’t want literally all of my friends to be a day trip away. The last time I was in Ukiah, at least a decade ago, I was at this coffee shop, too. But back then it was called the Coffee Critic. Apparently they went under back in 2010 or 2011, and two locals—one of whom had been a head roaster for both Ritual Coffee and Counter Culture Coffee, two pretty big names in the snooty coffee world—bought the shop. They’re doing light roasts in that “third wave” style, but also doing dark roasts, blends, crazy flavors for lattes and mochas, all to fit in to the local market. Frankly, this is too far from home to be out this far on a Sunday at a quarter to five; it’s a three-hour haul back. But because of that time, now that I’m here I’m going to find some place for dinner. Ideally I’ll have enough time to walk around, too, and won’t get back deathly late. I’d planned to get out and do some writing today; I don’t know if that’s going to happen. Other than, of course, this. But maybe that’s okay. To support my writing, consider a tip on Ko-fi.com. The iPad needs more focus on the little things 2019-08-12T11:36:07+00:00 I’ve been using an iPad Pro instead of a laptop for going on two years now, and have definitely spent more time on it than I have on my personal Mac during that time. Name a major writing app on the iPad and I’ve almost certainly not just tried it but given it a serious spin. We’re talking an 80,000-word novel in Scrivener</a>; short stories, multi-part novellas and blog posts in Ulysses</a>; a screenplay in Slugline</a>; random bits and bobs in Drafts</a>. I’ve made cover art on the iPad. I’ve created shortcuts for arcane conversion and batch processing. I am not an iOS guru, but I don’t think I’m overselling myself if I say I’m a power user. One could argue that the iPad has become my main computer, too. (Maybe “had”? I’ll come back to that.) It’s always been hard to explain why the Mac works “better” in some subjective way than a Windows PC. It’s not one huge thing; it’s the sum of small, seemingly inconsequential things that add up to a nicer experience. The iPad often feels that way compared to the Mac (or PC), because so many big ticket items—document management, windowing, security—have been radically rethought for the better in iOS. The problem is the little ticket items, if you will. To get into that, I need to talk about writing on the iPad. It’s become received wisdom among a certain set that the iPad is great for long-form writing. Here’s tech pundit turned novelist Matt Gemmell back in 2016 talking about using iOS Scrivener for writing novels</a> (he later switched to Ulysses</a>). MacStories’ editor Federico Viticci, who calls his iPad his “main computer,”</a> literally wrote a book called Writing on the iPad. Podcaster and writer/blogger Jason Snell has blogged about his setup</a>, which he’s been using in some form or another since 2014. The problem—for me, but I would argue I am not a unique, special snowflake among writers—isn’t the writing, it’s the editing. Editing is what you do after the first draft. Rewrite paragraphs. Move text around. Run a spell check. Change words across the entire document, or even multiple documents. Look at the first part and the last part of the story (or article or whatever) together to make sure you’re staying consistent and not contradicting yourself. And the iPad is just not good at that stuff. Let me show you. On the Mac, the arrow keys behave the same way in every editing app, and the modifier keys (nearly) always behave the same way, too. On the iPad, though, it’s not merely that behavior isn’t consistent app to app—it’s that most editors get at least one basic operation just bonkers wrong. Tap the up arrow repeatedly, and at some point the cursor jumps to the start of the line. Option-up jumps to the start of a paragraph and stops rather than continuing to move up. Option-down moves with weird hitches. On the Mac, Shift with any movement operation performs the same operation but selects the text; on the iPad, that’s usually true, but not a given. I’ve seen at least one app that doesn’t let you use up and down arrows when holding down Shift, and several that don’t handle Shift with another modifier key, like Option or Command. What in blue blazes is going on? Mac apps are built on the crufty and old AppKit framework, while iOS apps are built on the shiny and new UIKit. And that’s the problem. UIKit’s shiny and new text components suck lemon-flavored poop balls. AppKit may be uglier, but it’s simply more powerful. iPad developers are on their own to implement things Mac developers get for free, and, the results are…not great. Okay, the arrow keys are quirky. But come on, you can live with that. But what about running spell check? I don’t mean “check as you type.” I want to turn off the red squiggly lines when writing first drafts, and check the document for spelling errors after writing. For each possible misspelling, I can skip that one instance, ignore the word for the rest of that review, or add the word to the dictionary. I can turn on grammar checking, which isn’t great but catches duplicate words, and once in a blue moon catches an actual grammar mistake. But on the iPad, it’s “check as you type” or nothing—and literally only check as you type. If you open a document full of Lorem ipsum, it will be blissfully, stupidly squiggle-free. There’s no way to ignore words or add words to the dictionary. (The Mac lets you do that even in check-as-you-type mode by right-clicking a word.) What if I’m writing an article of a few thousand words full of technical terms? How about my 110,000-word science fiction novel Kismet, with invented city names and in-world jargon like “totemic” and “cisform?” Some of my complaints have fixes in iOS 13. Touch (but not keyboard) text selection is getting overhauled; apps can run in multiple windows once compiled against the iOS 13 SDK, which should address both “see two places in the same document at once” and “let me have a notes document and the main document open in the same app at once.” But the final editing that I’m going to go through with this blog post—a spell check and a pass through Marked</a> with its keyword highlighting turned on—that needs a Mac. You use Marked by having the same file open in both it and your editor simultaneously; I’m not sure that’s even possible in iOS. Look, I love that the iPad is rethinking so many big things about the computing experience. And I get the pushback about how the iPad is not a “laptop replacement” in the sense of letting you do what you did on your laptop the same way you’ve always done it. What I’m talking about here are little ways in which the iPad is not different than the Mac, but objectively worse. And here’s the thing: despite the common wisdom that iOS is wired and macOS is tired, I’m using iA Writer right now on the Mac and it’s just as good as it is on the iPad—and I can do all those Mac-only things like use Marked with it. What are the iPad-only things I can do with it that don’t have Mac equivalents? There’s no share sheet, but there’s an export command—and to open it simultaneously in Marked, I just dragged the document icon in the title bar to the Marked icon. The iOS counterpart would be “tap the share icon, tap ‘Share…’, tap ‘Marked’ if it existed.” I don’t have Shortcuts, but I have the Services menu and Automator. And AppleScript. And Keyboard Maestro. The Mac turns out to be pretty good at this application interoperability thing, huh. “Yeah, but the iPad is young, and you need to cut it some slack.” For how long? The iPad is nearly a decade old, and iOS is even older. I’m not asking why we can’t still can’t sideload even signed and notarized apps, or install non-toy development environments, or even just make Chrome our default browser. All those things are good questions, questions that, if the iPad is truly the future of computing, Apple needs to deal with. But I’m just asking why I can’t add a word to the system dictionary. I’m asking why, when I connect a Bluetooth keyboard, I can’t expect consistent behavior from the fucking arrow keys. It’s nice that Apple focuses on big moments of wonder and delight on iOS—but in truth, the Mac can still be pretty wonderful and delightful, too. I’m happy I let myself be surprised at how nice using an iPad as a main computing device can be. But I suspect some of the more partisan iPad users would be surprised at how nice using a Mac is if they let themselves. I love how those big-picture Computing Experiences are being rethought on the iPad, but it’s past time for iOS to go after the prosaic bits and bobs that the Mac had nailed before the turn of the century. Don’t get me wrong: I’m not giving up my iPad. Who knows what iPadOS 14 will bring? But in the meantime, I confess I’m watching what happens with the next MacBook Air revision pretty closely. To support my writing, consider a tip on Ko-fi.com. Review: Brydge 12.9″ Keyboard Pro 2019-07-12T10:43:43+00:00 I take my 12.9″ iPad Pro to the office every weekday, where I sneak in writing during lunch breaks. On weekends, some weeknights, and even the occasional work-from-home Thursday it travels to coffee shops and microbreweries around San Francisco Bay. For the past couple of years, when I’ve traveled the iPad has been my sole computer. I may use it more than I use my iMac. So when Brydge announced they were making their keyboard for the new iPad Pros, I jumped. It’s beloved of (some) heavy iPad users. Mine came at the end of May, and I’ve been using it for over a month now. I absolutely see why people like it. To my surprise, though, I’m not a convert. Let me be clear: if you want “laptop running iOS,” there’s a lot to recommend it. Unlike most iPad keyboard solutions, you can set it in your lap. You can adjust the “screen” to any viewing angle. It’s terrific. This is stuff I miss when I just pair the iPad with the Magic Keyboard. The Brydge keyboard itself is about as thick as the keyboard part of a MacBook Pro. So thin! Well…actually, no. That was the first (literally) big problem. The MBP just has a screen on top of it, but the Brydge has an iPad. The “BrydgeBook” they form together is thicker and weighs more than a 13″ MacBook Pro, let alone a MacBook Air. If you carry this thing around with you, you notice the weight penalty. The iPad clips to the Brydge with little rubber clamps that act as hinges. It’s a clever design, but they grip the iPad so tightly it’s difficult to get the iPad out quickly and even harder to get it back in and lined up just so. Maybe I’d get better at this with practice, but I’ve been practicing for over a month. I also worry about the stress this puts on the iPad’s corners and sides; “opening the lid” means gripping one edge of the iPad and lifting it up, rotating the hinges—which take a lot of force to move, as they need to be tight enough to hold the weight of the iPad up at an angle—by using the iPad as a lever. But I still haven’t talked about the keyboard part of the keyboard, have I? I like the real inverted “T” for the arrow keys and the dedicated home button; I don’t like the dedicated Siri key. The keys themselves feel a little mushy yet a little too resistant at the same time. Maybe it’s the stiffness, maybe it’s the gumminess, maybe it’s the way the keyboard itself is recessed more deeply than other laptops, but I miss letters more often on the Brydge than on any keyboard I recall using in years. I don’t want to make it sound like it’s a bad keyboard. It’s not. I just don’t like it as much as the Magic Keyboard, or the older Apple bluetooth keyboard, which has some of the same gumminness, at least when compared to the Magic Keyboard, but less resistance. I don’t even like it as much as the polarizing “butterfly” keyboard (talking just about key feel, mind you, not reliability). Before getting the Brydge, I thought “If you want a MacBook, just get a MacBook” was a dumb criticism. Now I’m not so sure. A BrydgeBook is the only way to get a laptop that runs iOS (er, iPadOS), and if that alone is enough to get you to buy it, I totally understand! But it’s a bulky, heavy laptop with a so-so keyboard, and opening and closing its “lid” puts stress on your iPad in ways it almost certainly wasn’t designed to handle. If what you truly want is a laptop, and you’re not wedded to iOS, look at the 2019 MacBook Air. If what you truly want is iOS, and you’re not wedded to the specific things you get with a laptop that you don’t with any other iPad keyboard solution, well…think twice about the Brydge. My previous—and now again current—travel setup is the iPad in the non-keyboard Smart Folio cover and a Magic Keyboard in Waterfield Design’s slip case, toted around in a Tom Bihn café bag</a>. The BrydgeBook is way faster to pull out and start using, and I can use it in my lap rather than needing table space or a lap desk</a> (which I use in the living room). But my iPad retains its essential iPad-ness. I don’t have to take the keyboard out at all. I can leave the keyboard where it is and pick up the iPad instantly. It feels true to the iPad’s intent. And If I bring my Compass Pro</a>, I can stand the iPad up in portrait mode, which turns out to be pretty damn cool for writing sprints. And I have what I consider to be the best travel keyboard ever made. I’m not sure what I’m going to do with my Brydge. While they do have a return policy, it’s not the most generous one out there, and I’m past my own return window anyway. I may come back to it later and give it a second round. If it continues to not stick, I’ll try to sell it; as I noted, a lot of people love it, and my lack of enthusiasm doesn’t mean it might not be perfect for someone else. To support my writing, consider a tip on Ko-fi.com. Random (but not angry) thoughts on 'Game of Thrones' 2019-05-17T22:58:26+00:00 (This will have spoilers for previously-aired episodes. Avert your eyes if you care.) “Everyone hates this final season!” I’ve noticed. Entitled nerd rage has been a thing over the last few years. “Oh, come on. Benioff & Weiss [the show’s creators] are terrible hacks!” Are they? They not only story-edited your favorite episodes, but they wrote most of the episodes that fans and critics both love. “Battle of the Bastards” (the second to last episode of season six), “Hardhome” (the third to last episode of season five), and “The Rains of Castamere” (the infamous “Red Wedding” episode, second to last of season three)—all written by Those Guys. “Okay, but the show ran got terrible when they ran out of George R.R. Martin’s books!” Are you sure? The show started blending in original work as early as season five, and had just about fully parted ways by season six. You can make the argument it was better in earlier seasons, but there were terrific episodes in later seasons that didn’t have much in the way of blueprints from the novels. “You’re not saying you like where this is all going, are you?” I like what I think they’re trying to do, I just don’t think they’re doing it well. They’ve said that they knew the “story beats”—the major plot points—of the final season as far back as season three. As shocked as people seemed to be by Daenerys’s tyrant turn, she’s had a cruel streak for as long as she’s been a major character, sometimes pulled back from villainy only by the counsel of others. Cersei says in the first season, “When you play the game of thrones, you win or you die,” but most of what we’ve seen since then—culminating with Dany’s tragic arc into supervillainy—suggests that, to borrow a line from another story, the only winning move is not to play. If there’s an overriding theme to this epic, it’s more old-fashioned than it lets on: a lust for power corrupts. (This is arguably why the Night King turned out not to be the Final Boss: he wasn’t playing the game of thrones, after all. Thematically, the final battles must involve those that are.) “If you’re not blaming the show ‘not doing it well’ on the writing, what are you blaming it on?” Oh, it’s the writing. It’s not the major story beats, it’s the banana crazypants pacing. The show has jettisoned years of magnificent character work to become a dragon-themed roller coaster, all because the creators decided they had to wrap things up in “70–75 hours” total. And so, so many of the things fans and critics have been bitching about might have worked with more breathing room. No weird time compression between events! Time to develop Dany’s descent into madness beyond a weird voiceover montage in the “previously on…” segment! No pressure to have characters make stupid decisions just to get them in position for the next plot event! No shoving critical plot and character moments off-camera! Explaining why the hell Jaime spent six seasons on a long redemption arc only to apparently throw it away because “Sorry, I just remembered I’m still awful after all!” (Actually, I don’t think that last one is explainable.) The most maddening thing is that I’ve never found an explanation for why they made the choice to race toward the ending as fast as possible. Would have making seasons seven and eight both 10 episodes, for 80 hours total, truly have been that onerous? “So can they stick the landing in spite of all this? Where does it go from here?” I don’t know, and a few seasons ago that would have been exciting. This season it fills me with sigh. I expect them to go for a “you didn’t see that coming!” note, and I expect it to be something that’s defensible based on the arc of the entire series—but no, I don’t think they can stick the landing. “Any predictions?” Dany on the throne. Arya enters. Jon enters. Dany rises. Greyworm rises. Arya draws a sword. Jon draws a sword. Greyworm readies his spear. Cut to black. Credits. To support my writing, consider a tip on Ko-fi.com. Medium thinks it's a brand 2019-04-13T12:40:36+00:00 There’s a lot of reasons people are down on Medium, Ev Williams’ ongoing whatever-the-hell-it-is. It’s a platform! It’s a publication! It’s a platform for publications! It’s a clean, clutter-free reading experience, except for all the clutter! There have been a few great stories written about this; my favorites are reporter Laura Hazard Owen’s “The long, complicated, and extremely frustrating history of Medium</a>” and acerbic typographer Matthew Butterick’s “The Billionaire’s Typewriter</a>.” (He occasionally updates this, most recently linking to Owen’s article.) Butterick critiques Medium’s design from an ethical standpoint, which turns out to be bang on point with Medium’s ultimate underlying problem: Medium thinks it’s a brand. What most bothers me when I click on a Medium link these days isn’t the increasingly dispiriting design (nothing says “great reading experience” like fixed-position bars at the top of and bottom of the window imploring you to sign in and/or pay money and/or download the app). No, that annoyance has been eclipsed by watching ever more articles on Medium go behind a paywall. You can only read three “premium” articles for free a month, just like The New Yorker, like The Wall Street Journal, like The Atlantic. And indeed, when you go to Medium’s upsell page, they’re pushing articles written by the sorts of authors you’d expect to see in a New Yorker issue: Roxane Gay, Dave Eggers, even Margaret Atwood, and namedropping publications that offer a selection of “curated” articles on Medium. So, their value proposition is: Publications you’ve heard of also publish their stuff on our site, like they do on Apple News! Think of us like a Readers’ Digest for hipsters. (Some of this content may be original, but a glance at Medium’s page for New York magazine shows that the only difference is that everything they put up on their own site for free is behind a paywall on Medium. Which, to be fair, is fucking brilliant if they can get away with it.)</li> Authors you’ve heard of published what may or may not be original content on Medium at least once! This piece from Margaret Atwood is from 2017, and she wrote a couple other things here in 2015, but it’s Margaret Goddamn Atwood, people!</li> By paying us, you get access to a whole world of other content from people you’ve never heard of, “curated” through the editorial mechanism of those authors clicking a checkbox saying “put this behind a paywall and let the checks roll in, please!”</li> </ul> Butterick asks what he dubs “the $132 million question”: Who’s going to pay $50 every year for this? People dissatisfied with the unlimited free clickbait available elsewhere? Gulls, rubes, and saps? Dogs with credit cards? </blockquote> While the aspiration to become the premier literary journal of our time by aggregating blog posts seems relatively new, this belief in its own essential brandness, its brandosity, forms Medium’s original sin. If you have a Medium account, your “front page” when you sign in isn’t a list of new articles from people you follow, like a Tumblr dashboard or, heaven forbid, an RSS reader. Goodness, no! Instead, it’s a Netflix landing page: articles they want to push at you, articles popular across the network, articles “recommended” based on your history. The design changes over the years, but the fundamental notion that you go to Medium™ to read Medium™ Stories remains. What makes a story a Medium™ Story? Who the hell knows? Medium™ surely doesn’t. They can’t. They have no control over their own content. Can you imagine Automattic deciding that because dozens of well-known authors run blogs on WordPress.com, they should charge $50 a year for access to blogs hosted on it? Hey, you can get three free reads a month to get a sense of what the WordPress editorial voice is like! This is essentially what Medium is doing, except that you get only one theme and don’t get to give your blog a title. (There was a point you could create a “publication” on Medium, which meant “give your blog a title,” but that’s gone, along with the ability to use custom domains. Remember: Medium thinks it’s a brand.) In what’s probably their thirteenth or fourteenth pivot at this point, Medium has brought back the idea of edited “magazines” with staff writers hosted there. They’ve done this before and the results were, what’s a good word, ruinous, but Owens quotes Ev Williams’s statement to Bloomberg in December 2018: “We are going to significantly increase our investment in original editorial in the next year, and we are absolutely not going to pull the football away this time, Charlie Brown.” I don’t know. Maybe there’s something I’m missing. If you’re paying for Medium, I’m genuinely curious why, and if you think you’re getting value out of it. Also, I’d like to know if you’re a Golden Retriever with an Amex. To support my writing, consider a tip on Ko-fi.com. Dan Olson on Patreon 2019-02-05T17:48:10+00:00 So, Dan Olson has an interesting thread</a> (note: requires Twitter login, thanks, Twitter!) on Patreon’s realization that they’re going to have to make a lot more money to make back their VC investment. Dan’s reading of Patreon CEO Jack Comte’s “Patreon needs to build new businesses, services and revenue line to be sustainable” is “we need to tack more junk on”; this may indeed be what happens, but I’m not sure it’s entirely what they mean. I actually wrote about this over a year ago</a> (apologies for the link to the Medium version, but I don’t seem to have it anywhere else currently, which I’ll have to fix when I can): To earn that $450M valuation Patreon has, they’re going to have to double revenue every year for the next four or five. Wouldn’t a great way for them to start making serious bank be to start landing creators who can get a few hundred thousand patrons instead of just a few thousand? As of this writing (December 2017), just six creators have more than 10,000 patrons. The top of the “long tail” curve Patreon has just isn’t that far above the bottom. This is the part of Patreon’s business that I suspect investors are most keen on changing. It’s great that Patreon can get Amanda Palmer now, but they’re going to need to get Imagine Dragons. I don’t mean “the next” Imagine Dragons, either. I mean an existing artist who can bring a bazillion fans with them. And to do that, going after Financially Successful Creators™ as they’ve defined it now isn’t good enough. They can’t just go after people they think Patreon can bring to the next level. They’re going to have to go after people who are already making six- or even seven-figure incomes from their art. They have to be able to say, hey, if you take a chance on us, we can give you the same income with fewer middlemen. </blockquote> I wrote that in response to Patreon’s quickly-rolled-back fee structure change, and since they rolled it back virtually the hour that I published the article, I think my conclusion kind of got buried. But you know what? I still think I was right. Dan Olson may be right that Patreon is going to feel obligated to junk things up for “little” creators (and patrons) in order to extract more money out of them, but the way companies like Slack and Dropbox have thrived is by focusing on the enterprise space and being kind of…let’s say “lackadaisical” about supporting mere consumers. It’s just that, as I wrote, the big value unicorn in Patreon’s space isn’t General Electric. It’s Beyoncé. I don’t think Patreon is necessarily going to eat itself, in Olson’s words–but creators with just a few dozen, or even “just” a few thousand, patrons may not be ones Patreon is structured to serve much longer. To support my writing, consider a tip on Ko-fi.com. A digression about Facebook 2019-02-05T12:54:43+00:00 Around this time last year, I was contacted by a recruiter from Facebook wanting me to apply for a technical writing position there. We talked for a little bit, I got the job information; it sounded pretty interesting, and the folks I know here in Silicon Valley who work for FB generally love it. They take care of their employees as much as modern tech companies do, and that’s not meant as a slight; while FB is more prone to the “constant frat party vibe woo” stereotype than possibly any other company as large as it is, they pay incredibly well, and have generous benefits with respect to bonuses, vacation, insurance, and retirement. And, at this point, I’d been out of work for a few months; part of that was by choice, in that I didn’t start looking for new work right after being laid off from Realm in September of 2017, but I’d just had a couple dispiriting rejections after trying to get back into the workforce and was feeling a little desperate. But…I didn’t feel good about working for Facebook. This is not about politics, per se; it’s about the way Facebook aggregates, synthesizes, and applies data. It is not an exaggeration to say that they’re the biggest intelligence-gathering organization in the world. We willingly tell them who our friends are and what groups we belong to; they can infer information we didn’t explicitly offer, based on who our friends associate with and what groups they belong with. They know what ads you’ve clicked on, even ones that aren’t on Facebook, since they supply advertising networks to other web sites and increasingly run free wifi at coffee shops and other businesses (just log into your Facebook account to connect). They know what you’ve searched for on Facebook, but they kind of know what you’ve searched for anywhere else, because they served you ads based on the search keywords you used and, hey, you’re logged into Facebook, so it’s you, welcome back! They know things you don’t tell your friends, or at least don’t tell all of them. They probably know if you’re gay, even if you’re still in the closet. They know you’re a science fiction fan, or a Golden State Warriors fan. They know you’re a furry. They also know you’re an alcoholic, or that you have a gambling problem, or that you have an STD. You’ve never explicitly told them any of those things, sure, but they’ve designed their platform to be one giant automated private investigation service…all in the service of giving you better, more targeted ads. They’d tell you that their mission isn’t just to serve you ads, of course, it’s to “connect the world.” But that makes it a little worse in some ways, doesn’t it? That gives them a philosophical backing for their ends-justify-the-means mentality. Isn’t connection good? Does that end not justify virtually any mean? Can’t any problems just be written off as collateral damage? Well, no. No, they can’t. Again, this isn’t about politics, per se, but on a meta level, it kind of is: Facebook wants us engaged, and we get engaged by clickbait. We’re engaged when we’re outraged. We’re engaged when we see which of our so-called friends are so very, very wrong about whatever’s got us fired up. Facebook has “connected” us with people we probably didn’t really need to stay in contact with beyond the occasional Christmas card. We think we’re expanding our social circle tenfold, but too often we’re fraying it, click after click. So I called the recruiter back and said that I couldn’t pursue the position. Nothing that happened since then has made me feel that was a bad call. Every month seems to bring a new story about Facebook’s essentially unethical behavior. And it is hard to overstate how much reach and power Facebook has in our economy and our society right now; the claims critics made just a year or two ago that seemed bombastic and ridiculous keep being proven right. It’s frankly not a good, healthy place to be, either as an employee or a customer. So far I’ve avoided deleting my Facebook account, because there are still people who I will literally only hear from if I remember to check FB (which I increasingly do not, for the record). It is so woven into the fabric of hundreds of millions of lives that the notion that someone you care about is not seeing your Facebook posts seems almost absurd. But I’m not checking very often, and I don’t expect that to change. It’s not impossible that 2019 will see me deleting the account entirely. To support my writing, consider a tip on Ko-fi.com. Let's talk about the Tumblrpocalypse 2018-12-04T16:57:52+00:00 Did you hear Tumblr’s getting rid of all the adult sites? Yes, the news is going around. Man, if only Apple wasn’t so prudish! Come again? This is all because Apple pulled the Tumblr app in mid-November after they found child porn on the site. Apple did pull the app because of that, yes, but there’s no evidence Apple is insisting Tumblr get rid of all NSFW material across the entire site as a condition to get back into the App Store. Besides, Apple has a “17+” rating category for apps, which Tumblr has been in since early 2013</a>. There’s no sign that they’ve been purging other apps in that category. But we keep hearing about how strict Apple is! Walled garden and all that. They keep cracking down on user-generated content. Apple’s actual guidelines</a> prohibit services “that end up being used primarily for pornographic content,” so sure, there’s not going to be a Pornhub iOS app any time soon. But “incidental NSFW content” is explicitly (ahem) allowed. Tumblr’s NSFW stuff is more than just incidental. Arguably, but Tumblr’s iOS app has been on the App Store since 2009–almost since there was an App Store to be on. Tumblr said Apple found child porn hosted on Tumblr in a “routine audit”; the word routine implies they audit a random sample of Tumblr sites through the iOS app at least semi-regularly. So it’s damn unlikely it took nearly ten years for Apple to be prudishly horrified by a naughty catgirl pinup. Well, if it’s not Apple’s fault, why would Tumblr do this? It’s going to kill their site deader than a doornail. Were doornails ever alive? It’s just an expression. Right. Well, okay. “Tumblr is for porn” has become received wisdom, but there are conflicting reports as to just how much porn is there. In 2012, Tumblr creator and then-CEO David Karp estimated</a> it at 2–4% of the blogs. A web analytics firm a year later</a> estimated it at 11.4%, and a study in 2016</a> estimated it at a mere 1%, but estimated 22% of the audience was there for the porn. That’s all over the map. Yeah. The analytics firm used “explicit domain names” as a marker for porn production, which is likely to overestimate, and the later study classified Tumblrs as porn if they could be found by “a large number of search engine queries containing pornographic keywords,” which I suspect underestimates. The chances are that Karp’s estimate was likely the best. While it’s an old estimate, I doubt the percentage of porn Tumblrs increased under Yahoo’s watch, given the Great Tumblr Porn Crackdown of 2013</a>. Let’s keep it on the high end and say 5%, though. No way. There are millions of pornographic Tumblrs! It’s easy to lose track of just how big the numbers involved are on an absolute rather than percentage basis. There are about 250 million Tumblr users. Suppose only a quarter of them actually post, and only 5% of those post porn. That’s still millions of pornographic Tumblrs. What about that figure of 22% of the audience being there for the porn? Is that suspect, too? This is really difficult to quantify, because the vast majority of Tumblr users who look at some NSFW content don’t look exclusively at NSFW content. Also, thanks to Tumblr’s reblog feature, you may see NSFW content you don’t explicitly (stop it!) intend to see; that 2016 study estimated more users saw porn that way than saw it by following Tumblrs they’d categorized as pornographic. So the real question isn’t how many people see NSFW stuff on Tumblr, the question is how many people will stop using Tumblr if they stop seeing NSFW stuff on it. So what’s the answer? No idea. I guess we’ll know in a year. I still think it’s gonna die. look at sites like LiveJournal and MySpace. Once people start leaving, they don’t come back, especially if the creators they follow aren’t there. That’s the million-dollar question, right? They’re going to take a big hit initially, but they probably figure it’ll be balanced out. But there’s a real chance that the big hit gets followed by a slow slide. You sound pretty sanguine about this. Doesn’t Tumblr making this move bother you? It does. I have a soft spot in my heart for Tumblr; my old tech blog</a> wouldn’t have taken off if it hadn’t been hosted on Tumblr (and probably wouldn’t have survived a few initial “Fireballings” when John Gruber linked to it). I’ve always thought it was underrated as a pure blogging platform. And, yes, I think it’s worthwhile to have a place to share NSFW content. But the bottom line comes down to the bottom line. Tumblr is on its third owner at this point and still largely resists monetization, and current owner Verizon is not going to keep running it as a social good. As risky as it might seem to bet that kicking off the pronz will increase ad revenue, it wouldn’t be a bet they’d make if Tumblr as it is now were a sustainable business. Do you think it’ll work? Maybe? I mean, if you’re asking if I think Tumblr will “die in a month,” or even a year, absolutely not. LiveJournal, MySpace, and Digg are all still around. Fucking Ask.com is still around. Tumblr could have an indefinite life ahead of it as an irrelevant artifact of internet history. Come on, give me an actual prediction. Okay, here’s the thing. Tumblr hasn’t really changed much in years, and that’s a risk. Making an unpopular policy change that drives high-follower-count users off the site is also a risk. Tumblr can survive those as long as there isn’t anything else that does the job it does, but both of these moves open up space for disruption. But as usual with disruptors, we shouldn’t expect The Thing That Replaces Tumblr to look like Tumblr. It might not even be just one thing at all. In fact, LiveJournal’s decline might be a good study: they, too, opened up space for disruption through a combination of site stagnation and stupid policy decisions, but their users didn’t end up all migrating to some LJ-but-better service. LJ was ultimately rendered irrelevant by the one-two punch of Twitter and, ironically, Tumblr. So what do you think the lesson is that anyone trying to disrupt Tumblr should draw from all this? Money first, naughty catgirl pinups later. To support my writing, consider a tip on Ko-fi.com. The new iPad Pro 2018-11-16T15:28:03+00:00 So I bought a new iPad yesterday. After a lot of waffling, I made some seemingly counter-intuitive choices: The 12.9″ model (up from my original 9.7″)</li> Only 64G of storage</li> WiFi only, not cellular</li> </ul> The rationale for the first one is simple enough. I’ve been using my iPad more than my laptop, and moving to the bigger screen pushes it that much closer to my only-needed portable computing device. Over the last two years I’ve knocked down nearly all the “showstoppers” that keep me from doing my personal work on the iPad, although there are still clunky points–many of which are more due to constraints in iOS. So if that’s the case, why go with the big iPad but the baseline version? That’s crazy! Well, okay. It is, sort of. But when I checked my previous iPad just a few days ago, you know how much of its 128G storage I was using? 30G. And the three biggest apps were Grim Fandango (which I haven’t played in a year), Garage Band (which I don’t use), and iMovie (which I don’t use). By either deleting or “offloading” a few of the biggest apps, I’m now using less than 24G. The reality is that I don’t edit media, I stream it. When I travel, I’m more likely to have podcasts or books with me on planes than movies. And I keep a lot of documents in cloud storage. I’m not a big photographer, but even if I start seriously using iCloud Photo Library, the iPad isn’t going to need to keep all or even most of the photos on it–that’s what the point of “cloud” is, right? Honestly, if Apple had had a 128G version for just $50 more, like they do with the iPhone XR, I’d have taken it. But they’re playing their stupid storage pricing game, as usual, and I’m not. As far as the cellular radio goes: I’ll be honest. That was tough. I’m still not positive I made the right call. I end up using cellular tethering a fair amount; I’m using it right now, in fact, typing this at lunch at the office (I don’t want to connect the iPad to the corporate network). Doesn’t that make me a perfect candidate for this? Well, sure. But it’s a $150 option now, and after paying that upcharge, then it’s either add it to my cell plan for another $10 a month or pick a “pay as you go” option. A lot of folks do that, treating the cellular radio as insurance and almost never using it. Well, okay, but if you almost never use it, you have less reason not to just put up with the inconvenience of tethering. I use it enough that I’m still considering taking the iPad back within the 14-day window and exchanging it. But it just feels like a lot to pay to save five seconds–yes, those five seconds could add up to a minute or two a week. But even so. Again, if this was a $50 upgrade, I’d have taken it almost without thinking about it. So how am I feeling about the new iPad? After less than 24 hours, I love it. I’ll see how it goes in real world usage, but it’s pretty awesome paired with the Canopy and Magic Keyboard (it is so close in width to the Magic Keyboard it almost looks like they were meant to go together). Do I worry I haven’t “future-proofed” this purchase? A little. But I think in practice I’ll be more than okay. Updates from the future: After a little waffling, I exchanged it for the 256GB iPad Pro, still no cellular (and still in Space Gray), on November 20, 2018. In mid-2020, I exchanged it again, this time for an iPad Air. Why? Well, same processor and a lot of the same accessories, but smaller and lighter, which helps differentiate it from the M1 MacBook Air. As of April 2024, I still have it! To support my writing, consider a tip on Ko-fi.com. Low cognitive load blogging 2018-11-14T20:40:53+00:00 Hey, did you know I used to be a blogger? Okay, it’s not quite fair to say that I’m no longer a blogger; I’ve made about a half-dozen posts this year. But that’s way down from the original Tumblr-hosted Coyote Tracks</a>: in the earlier parts of this decade I was at least managing a few posts a month, and occasionally even a few posts a week. Ironically, this post is going to join the rest of this year’s flock as another post about blogging. The truth is that I simultaneously miss it and don’t want the cognitive burden of committing to it again. I’m edging toward making this microblog my “real, canonical” blog; after all, it can accept long posts like this one, it crossposts to Twitter and can be followed via Mastodon, and it would let me quietly move my main web site off WordPress onto…well, frankly, I don’t know yet, although I’m perversely considering saying to hell with “generators” and moving to pure hand-coded HTML. But the main advantage that moving to the microblog—as well as moving to the static site not-a-generator—would give me is freedom from that cognitive load. Okay, too strong: a reduction of that cognitive load. I don’t have to worry about templates more complicated than what BBEdit</a> handles for the main site, and the journal can be updated with any Micropub</a> client (including, of course, Micro.blog’s own client). The million-dollar question: if I go ahead and make this change, will I actually start blogging again? When I think, “I should write about the problems I see looming ahead for Apple,” or “maybe I should write about why I’m considering going iPad-only for portable computing despite that last thought,” or “I feel a little like ranting about how ridiculous it is to deride Nancy Pelosi as a toothless centrist,” will I actually do it? I don’t know. But I know that making it easier for myself to get there probably can’t hurt. To support my writing, consider a tip on Ko-fi.com. Apple's topsy-turvy iPhone lineup 2018-09-13T14:20:57+00:00 I started thinking about this yesterday, and originally was going to say crazy, as in irrationally expensive. William Schuth</a> expresses this well: My strategy had been to buy the mid-tier spec of the best iPhone offered. My iPhone 6S Plus cost me $849 at launch; the mid-tier XS Max is $1,249. That’s a whole Apple Watch worth of price inflation in three years. </blockquote> The more I think about it, though, the more I think it’s not as simple as “Apple jacked up the price of the best phone a lot.” They did do that, no question. But they also made the “less-best” phones a lot better. In the iPhone 6, 6S and 7 years, the calculation was pretty straightforward: Get the normal model</li> Pay extra for the Plus model, which got you a bigger screen, bigger battery, and better camera</li> </ul> Last year, though, the calculation changed a little: Get the normal model (iPhone 8)</li> Pay extra for the iPhone 8 Plus, which got you a bigger screen, bigger battery, and better camera</li> Pay even more for the iPhone X, which got you Face ID, an OLED screen, a bigger battery, the Plus’s camera, and an edge-to-edge screen bigger than the iPhone 8 but not as big as the iPhone 8 Plus</li> </ul> This year, though, things are even weirder. All of these phones now have Face ID and an edge-to-edge screen. So: Get the presumably normal model, the iPhone XS, which is “iPhone X with some bumps” (like most “S” model years)</li> Pay extra for the iPhone XS Max, which gets you a bigger screen and bigger battery (but the same camera)</li> Pay less for the iPhone XR, which still gets you a bigger screen and bigger battery than the iPhone XS, but drops back to an LCD screen and a slightly worse camera</li> </ul> So while Schuth’s heuristic ostensibly leads to the $1,249 XS Max, this isn’t the same scenario as we had with the iPhone 6, 6S and 7 where the difference between normal and Plus was obvious, nor is it like last year’s scenario, where the iPhone 8 and iPhone X were starkly different. This year, you have to really want that OLED screen and dual lens camera to make the XS worth it, and you have to really want the Galaxy Note-sized screen to make the XS Max worth it. In many (albeit not all) ways, the XR is the true successor to the Plus versions of years past, and it’s priced like it. The mid-tier XR is $849. I wonder how this is going to affect iPhone sales next year. Does the ASP go up, because of the Max, or down, because of the XR? I’m betting the latter is at least possible. Unlike the iPhone 8 vs. the iPhone X, the XR provides a huge chunk of the ooh cool new shiny of the iPhone X, it’s available in unique colors, and it’s not a grimace-inducing price. (Well, no more than the Plus phones were, at the least.) This is my upgrade year (I’m on an iPhone 6, no “S”), and it’s going to be a tough decision for me. To support my writing, consider a tip on Ko-fi.com. Form over frolic: Jony Ive’s quest for boring perfection 2016-12-16T14:30:52+00:00 Right now I’m sitting in front of a 27″ iMac. It’s the best computer I’ve ever owned, with a 5K display, high color gamut, 24 gigs of RAM and 512 gigs of SSD storage. It’s beautiful and minimalist, just like every iMac they’ve released since they switched to aluminum in 2007. It’s also the least modifiable desktop computer I’ve ever owned. This trend also goes back to that aluminum iMac, in which — like today’s—only the RAM is user-upgradeable. (Since 2012, even that’s no longer true of the smaller 21″ iMac.) It’s hard not to ask: why is thinness the priority in all of Apple’s designs? You know the answer: Jony Ive. It’s clear by now that he would like everything Apple produces to look as close to a pure pane of glass as he can make it, with minimal, unadorned metallic frames, as close to unbroken and symmetrical as functionality allows. And Ive’s team is perfectly willing to sacrifice functionality in pursuit of this goal. A female Lightning port is fractionally thinner than a female USB-C port, and now you know why the iPhone will never get USB-C ports. Sorry. You’re lucky the one-port MacBook’s one port isn’t a Lightning port. (I have it on good authority that was under consideration.) This often gets portrayed as a choice between staying chained to legacy hardware and forging ahead to the future. But if you were using Macs a decade ago, do you remember the way the power indicator light on a Mac, both desktop and laptop, used to slowly pulse when it was asleep, as if it were slowly breathing? Or the way batteries on laptops, both replaceable and permanent, used to let you check charge levels without turning on or waking up the machine. Or, as recently as last year, the way power plugs changed color to show charging state. All of that—along with the illuminated Apple logo and, now, the cheerful startup chime—has gone away. All the price of progress, right? A couple years ago, Shawn Blanc published a book about “how to make good things great” called Delight is in the Details</a>. That phrase captures an essential paradox: we want our products to stay out of our way in everyday use, yet products that convert us from merely satisfied customers to fans have little touches that call attention to themselves in just the right way. When I start my Mazda, its display lights up with the words “Zoom Zoom” for just a few seconds. It’s stupid, but after six years it still makes me smile. “Little touches that call attention to themselves” are the opposite of Ive’s guiding aesthetic. He creates beautiful objects you can appreciate as works of art. You can’t help but marvel at the lengths to which his team will go to make a perfect fusion of glass and metal, to craft UIs that appear to directly manipulate data, to make the hardware disappear while you’re using it. Under Ive’s direction, Apple delivers works which are closer to the science fiction future than any other major consumer electronics company. And yet his designs are relentlessly whimsy-free. There won’t be a moment that catches you off-guard and makes you smile. Ive’s work never aspires to make you giggle with delight. Software doesn’t escape his penchant for austerity, either. The Ive era of software UX has been about flattening, removing, relentlessly stamping out skeuomorphism. The “traffic light” window controls are just circles now; the swirling barber pole progress bars are simple blue, with a subtle pulse; we don’t even get the little puff of smoke when we pull icons off the dock. I’m surprised the iOS icons still jiggle-dance when they’re in rearrangement mode. I’m not sure that it’s fair to say that we’re seeing a software analog to Apple’s quest for thinness, but I’m not sure it isn’t, either. I’d hardly be the first one to complain about a perceived drop in software and UX quality, or to question whether Apple’s being a little too aggressive in dropping legacy ports. Yet it feels like that’s always been part of the deal, right? We’re taking away the floppy drive, or only giving you these weird USB ports, or sealing the battery in, but look at how cool we can make this thing now! It’s not like anything else on the market. It’s fun. This iMac is the best computer I’ve ever owned, but nothing about it screams fun. The quirkiest thing about it is my mechanical keyboard</a>, something Apple would never dream of making on their own these days. (So gauche.) Yes, but you keep talking about the Mac line. The future is in iOS! Despite revealing myself in past posts as a Mac partisan, I think this is not only true but, overall, good. I’m a fan of that science fiction future, and it’s not one in which I see many people sitting down in front of 27″ monitors and keyboards for their computing needs—even if the monitors are holographic and the keyboards aren’t physical. But man, talk about the “pure pane of glass” ideal, right? The argument Apple is implicitly making is that computers—especially the computers of the future that the iPad typifies—are appliances. Appliances can be beautiful, but they shouldn’t exhibit frippery. They should be focused. We should prefer the Kitchen-Aid stand mixer to the plastic knockoff that does twice as much at half the price, because it won’t do any of those things well and it’ll fall apart in a year. (Besides, you can do all those things with the Kitchen-Aid, anyway; you’ll just need to buy some dongles.) That’s all true. Maybe Ive knows best. But if you showed me a table with an iPad Pro, a Surface Pro, and a Surface Book on it and asked me to rank them in order of Cool Factor, I’d be hard-pressed to put the iPad at the head of the line. Microsoft isn’t trying for tiny-quirk delight, which is just as well (“It looks like you’re trying to add personality to your UX! Can I help?”), but they’re sweating small, thoughtful details. Apple sweats the details of manufacturing processes. That’s great, but it’s not the same thing. Maybe—just maybe—a little frippery is okay, even if it adds a half-millimeter in depth to a product, or adds a touch of (gasp) skeuomorphism to the UI here and there, or allows a slightly less restrained, tasteful pigment on the anodized aluminum case. Injecting a bit of fun, even weirdness, to their computers in the late ’90s helped pull Apple back from the brink. It may be time for another injection. Being Kitchen-Aid is a fine goal, but you know what? They sell that stand mixer in nearly three dozen colors. To support my writing, consider a tip on Ko-fi.com. Bubbles, Baseball, and Mr. Marsh 2016-12-07T16:15:48+00:00 Deep stories and ugly truths in American politics When I went to high school in Florida, I was in one of the last years that had to take a class called “Americanism vs. Communism.” I had the dumb luck to get a teacher named Bob Marsh, an iconoclastic sixty-something motorcyclist and science fiction fan who told the students like me who were also sf fans about his friend Joe Haldeman</a>. While there’s a common joke I hear even from today’s high school students about American history classes ending at World War II, we learned about the Cold War, about Korea, about Vietnam. We learned about Castro and Kruschev and Mao, but also about Watergate and COINTELPRO</a> and HUAC</a>. It was, improbably, a pretty good class, I suspect almost entirely due to Mr. Marsh. Despite the name, the class’s story wasn’t about how Americanism, whatever that was, opposed communism. It was about how liberal democracy opposed authoritarianism. That sense of “liberal” has gotten conceptually muddled over the years, particularly in post-war America. (Call it “classical liberalism” if you must, although that phrase is even more easily coopted.) This is the point: Bernie Sanders and Paul Ryan might not agree on much, but Ryan has never, to the best of my knowledge, advocated for a return to monarchy; Sanders has never once suggested outlawing private industry. They would both agree that the individual liberty and representative democracy thing is, on the whole, a pretty good idea. They are both pretty firmly standing for liberal democracy and against authoritarianism. That’s a foundational ideal of America. We’ve failed to hit it a lot through our history, but we’ve still done better than many other countries. The name “Americanism vs. Communism,” though, tells us another story, a story that’s been pervasive in America in the post-World War II era. This story tells us that if we want to oppose authoritarianism, we need only worry about “the left.” It doesn’t tell us that “the right” has its own kinds of authoritarians. To some people, it even implies that Nazis were socialists (it’s right in the name!), and that fascists were liberals. The name “Americanism vs. Communism” tells us, maybe, to let down our guard. On John Gruber’s podcast “The Talk Show</a>,” guest Merlin Mann said of the 2016 presidential election: “It’s not that my team didn’t win. It’s that maybe I just don’t understand baseball anymore.” Merlin and I went to the same small Florida college at more or less the same time. (We all totally knew he was going to be a professional podcaster.) I’m pretty sure he also took an AvC class. We probably share a roughly similar, and from appearances similarly inadequate, understanding of baseball. Before the election we were inundated with think pieces about how “the left” was wildly misinterpreting the appeal of nationalist populism. No no no, we were told, it’s not racism and misogyny and homophobia. It’s the rage, the deep story, the message to people who felt they were being not merely left behind but that “the elites” were letting other people “cut in line” ahead of them on the way to the American Dream. We’re still constantly hammered with the idea that if you’re in a city you’re in a bubble, if you’re liberal you’re in a bubble, that we just need to get out of that bubble and listen to real, non-bubble America. The deep story may be about all that. But it’s also about how gay marriage devalues “real” marriage. How letting transgender folk use public bathrooms puts “real” men and women in danger. How we should watch, register and deport immigrants and build a wall around our borders. The racism and misogyny and homophobia isn’t incidental. It’s not a byproduct. The deep story is about tribalism. Here’s an ugly truth: some of the country doesn’t believe that America belongs to people who aren’t in their tribe. That tribe is white, straight (at least openly), and Christian. It’s gotten bigger over the years—it didn’t used to include the Irish, or Italians, or Catholics, or women—but every inch of expansion has been fought, bitterly and grudgingly. Other tribes can live in America, maybe, but theirs comes first, and everyone else is here at their forbearance. Another ugly truth is this: some of the country considers not just welfare, not just social programs, but basic justice and legal protection to be a zero-sum game. Her marriage means less if you can get married. The sign on the restroom door means less if you can go through it. The police are here to protect me from you. And when it comes to actual tangible costs, they would rather everyone get nothing than risk paying to help you. The third ugly truth is this: those people are in power now. Despite my sarcastic streak, I’m a natural optimist. I’m not going to claim there’s much of a silver lining here, though. I believe that the oft-maligned millennials—and even us Generation Xers—will pull us back on track. I don’t think this is the end of the great American experiment, that representative democracy is at its end, that America is doomed to become a mashup of The Handmaid’s Tale and Idiocracy. But I think it’s going to get worse before it gets better. And I don’t know how much worse. I wonder what Mr. Marsh would have said about all this, back in that Americanism vs. Communism class. I think he might say the problem isn’t bubbles. It’s not who’s listening to who in the present. It’s who’s listening to the past. America has always been at its worst when we’re encouraged to turn against one another, and at its best when we move toward ensuring that liberty and justice truly is for all. I think he might also say this. Liberal democracies can vote themselves into authoritarianism. Voting themselves back out is much harder. That seems obvious to me, but I never did understand baseball. To support my writing, consider a tip on Ko-fi.com. Text Editor Intervention 2011-05-04T12:30:00+00:00 “Don’t write a text editor; you’re reinventing fire.” — Ben Straub</a> As I’ve mentioned here before, I’ve been a TextMate user for the last few years, albeit increasingly reluctantly. TM has always been a mix of sheer brilliance and stone cold stupid, and while the former outweighs the latter, when the latter pops up it really gets in your face. (Undo character by character, anyone?) We first started hearing about TextMate 2 in early 2006</a>, and as people will always respond if you point out that it’s now 2011, the author never gave an ETA other than “after Leopard.” All well and good, but if your dad walks out one Thanksgiving saying he’ll be back “sometime after Christmas” and it’s now five years later, when your little sister tells you “he didn’t say how long after Christmas” she’s maybe not facing reality. If your dad pops up to make a blog post once a year saying he’s still working on it, he is just possibly not facing reality, either. So. A lot of old TM users have been looking for the Editor That Will Be TextMate 2.0 By Default. This has in turn given rise to new editors all aiming for that ETWBTM2BD spot. I’ve been following a few of them sporadically. Sublime Text</a> is coming to the Mac! It supports (some) TextMate bundle operations to make it easier to move over to it! And so does Kod, and even though it doesn’t actually do anything yet it’s open source! Let’s not forget Panic’s Coda and MacRabbit’s Espresso, which are both marketed as quasi-IDEs aimed specifically at web developers and—again—sport limited compatibility with bits and dribbles of TextMate’s bundles. Other than Sublime Text, all of these editors are dead now: Coda has been replaced by Nova, Espresso is abandonware, and Kod and Vico were both basically stillborn.</aside> A few days ago, word started spreading around of yet another new editor, Vico! Like Sublime Text and TextMate, Vico is a shareware program developed by just one guy and sold for about $50. This one, though, is a native Cocoa editor that uses vi</code> keybindings, uses Nu</a> instead of Vim’s scripting language, and is partially compatible with TextMate bundles. Wait. Now we’re trying to make Vim pretty? Really? I love you all, but it’s time to stage a text editor intervention. Put down the mouse, back away from that download link, and take a deep breath. First off: if you are a Mac user and compatibility with TextMate is an absolute must-have, let me ask you two questions. Is TextMate 1.5 still working for you? Can you keep living with its limitations? If you answered both those questions “yes,” our work is done here. Go in peace. Otherwise, I’m going to make a radical prescription. I like both Coda and Espresso, but right now I can’t recommend either one unless you’re mostly doing static HTML files. Your decision is between three editors: BBEdit, MacVim and Cocoa Emacs. Seriously. This assertion will piss off BBEdit fans, but BBEdit is the least powerful of those three. Its color schemes are minimal, its “codeless language modules” are underpowered, and it can’t do syntax-aware indenting. And it’s $99</s> $49, whereas the other two are free. But it’s the only one of the three which is a true Mac program. In its basics, it behaves like every other Mac text window you’ve ever seen and just adds lots of great stuff onto that. MacVim and Cocoa Emacs make a few concessions to the Mac environment, but you gotta learn Vim or Emacs to use one or the other. But all of these editors are very extensible. Either out of the box or with an afternoon of dorking around, you’ll have everything you miss from TextMate. They all have TextMate-style snippets. (Ironically, BBEdit had them years before anybody else; the extensions for Vim and Emacs are both consciously modeled on TextMate’s.) All of them have folding and multiple windows and pane splits. All of them have quick navigation between files in a project hierarchy. If you’re the kind of nerd who knows what Emmet</a> is, all of them have add-ons for it. And, oh yeah: they all understand that tabs are a terrible way to manage multiple open files once you have more than four or five open at once, and all can handle files much bigger than TextMate can dream of. Those of us who cling tenaciously to TextMate do so because of its amazing bundle system; you really have to spend time digging into it to understand how powerful it is. BBEdit, Vim and Emacs all require you to learn a weird scripting language (AppleScript, VimL, and Lisp, respectively); TextMate lets you write in any language you can script a Unix shell with, from bash to Ruby. The flip side of that, though, is that TextMate’s bundles are—with few exceptions—limited to passing standard input and environment variables to a shell script and getting standard output back. With BBEdit and the Ugly Unix Twins, you can attach scripts nearly everywhere, with deep access to the underlying editing engines. I keep trying to love BBEdit once more but at best it’s an affable fling. I still prefer TextMate for writing Markdown prose (like, say, this blog post); for code, I’m spending more time hanging out with MacVim these days. It’s not as pretty as Vico, but it retains Vim’s top-notch support for displaying multiple files at once. (None of the native Mac programs except Coda can display more than one file per tab or physical window.) And Vim’s modal editing really is faster once you get used to it for coding. Yes, it takes a damn long time to start making sense. But the first time you watch someone cut a <div></code> block in an HTML document by putting the cursor somewhere in the open or close block tag and typing three characters—or typing one command for “delete every line after each line that contains this pattern”—you have to at least concede there’s something to it. Here’s my baseline test: can I tell the editor I want most code to be indented by 4 spaces but YAML by 2? In 2011, this should be dead simple—but it knocks all but our Fab Four right out. (In theory, Sublime Text can do this, but dear God, you thought Vim configuration was arcane? Welcome to the 500 Preference Files of Bartholomew Cubbins.) Okay, sure. You’re an editor junkie like me. I understand. I’m not saying you can’t keep watching the other programs. Play with ’em in your spare time. One of them might end up being something that blows our collective doors off. It’s okay. We can all hope. (Maybe dad has just been out on a smoke break all this time, too.) But in the meantime, you gotta get work done. Either pony up money for BBEdit, pony up time for MacVim or Emacs, or stick with TextMate. To support my writing, consider a tip on Ko-fi.com.

Coyote Tracks

"How are you leveraging AI in your technical writing?"

What you're allowed to do

“Trim Silence”: Threat or menace?

Is AI helping you to do work, or just to do stuff?

Emacs and Shortcuts

Apple wins me over again, but it’s getting tougher

The Emacs for everything obsession

Collaboration or privacy: pick one

If not the Mac, what?

Seriously, Markdown

It's the heat and the humidity; also, the fascism

Dr. Neckbeard, or how I learned to stop worrying and love Emacs

Maybe the waiting is actually the easiest part

The robots have come for your em dashes

What makes an app feel “right” on the Mac?

Having to think about essential computerness

Just slap an LLM on the front end—how hard can it be?

The Reluctant Technologist

The pressure to stay genteel

Platforms are not neutral

On evacuations and hurricanes

Generative AI comes for imaginary friends

How much of yourself do you put online?

Conservatives are the real snowflakes

It could be burnout

Creative writing and AI's failure modes

Judge the AI that exists, not the AI you imagine

A year back in Florida

You're So Vain, You Probably Think This App Is About You: On Meta and Mastodon

Twitter, failure modes, and your favorite bar

Thoughts on Leaving California

BBEdit 14, and why you should care

Leaping forward</h2> Here’s something cool:</p>

Panic’s Nova, ten months in

A quick unofficial Apple Music Spatial Audio FAQ

A quick unofficial Apple Music Lossless FAQ

The Peril of “No Politics”

The Mac and the iPad aren’t meeting in the middle yet

A brief chat about Chuck Wendig, the Internet Archive, and bad information spread in good faith

Panic's Nova text editor (a review)

Signs of the old Apple

Notes from the Road

The iPad needs more focus on the little things

Review: Brydge 12.9″ Keyboard Pro

Random (but not angry) thoughts on 'Game of Thrones'

Medium thinks it's a brand

Dan Olson on Patreon

A digression about Facebook

Let's talk about the Tumblrpocalypse

The new iPad Pro

Low cognitive load blogging

Apple's topsy-turvy iPhone lineup

Form over frolic: Jony Ive’s quest for boring perfection

Bubbles, Baseball, and Mr. Marsh

Text Editor Intervention