Glyph

he/him

You probably heard about me because I am the founder of the Twisted python networking engine open source project. But I’m also the author and maintainer of several other smaller projects, a writer and public speaker about software and the things software affects (i.e.: everything), and a productivity nerd due to my ADHD. I also post a lot about politics; I’d personally prefer to be apolitical but unfortunately the global rising tide of revanchist fascism is kind of dangerous to ignore.

posts: https://blog.glyph.im/
disclosures: https://blog.glyph.im/pages/disclosures.html
code: https://github.com/glyph
patrons: https://www.patreon.com/creatorglyph

Public Key

npub16y8yvp4qyfg03ljv5ahvcz0l3l9ggkmry6fszeek5a0x5gtgvd4q0pd455

NIP-05 Address

[email protected]

Profile Code

nprofile1qqsdzrjxq6szy58clex2wmkvp8lclj5ytd3jdycpvum2whn2y95xx6spz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0ds6mp0jc

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub16y8yvp4qyfg03ljv5ahvcz0l3l9ggkmry6fszeek5a0x5gtgvd4q0pd455

Show more details

Last Notes

2026-03-29 04:36:03 CEST

- reply

(I did not actually do this. But it is an advertised feature of chrome, and I am pretty sure it would "work", as much as any AI-based solution "works".)

2026-03-29 04:35:31 CEST

I am writing a blog post about why it is bad to use AI. It is extremely heavily sourced. I have a tedious automation problem formatting my citations. No problem, I think. I will write a computer program. The computer program does not work, because websites are blocking simple computer programs in an effort to block AI. Solution? Simple. Browser comes with AI embedded, browses like a human, has all my cookies. Just ask the AI. It sails through the primitive anti-AI measures easily.

2026-03-18 01:06:41 CET

- reply

Thanks, I appreciate that. And I understand your perspective. I think it would be best to table this for now and have a longer conversation when there is maybe a little less bullshit happening in the world such that we are both better-regulated and able to deal with nuances.
Personally I find it kind of exhausting to have to include five paragraphs of qualifications about the harm that they do before every fact that I share about Apple, but I also understand why I need to.

2026-03-17 20:20:50 CET

- reply

not something I have any interest in convincing you of. the only reason I talk about this stuff is to try to help the free desktop community properly understand what you're up against. if the model of the world is just "apple is acausally evil and makes inferior products; people buy their devices only because they're stupid" then there's no hope of ever competing with them realistically

2026-03-16 03:24:52 CET

- reply

I mean, technically, yes, the devices do still exist, but I don't know anyone who uses one willingly, even *very* budget-conscious folks. 15 years ago a windows shitbox was largely the same experience as an iPad or a chromebook at a competitive price point with some bonus ability to run older casual games and entirely serviceable. Nowadays it's a hideous albatross that can barely boot Windows 11 and is nowhere close to what you can get in other categories for about the same price.

2026-03-16 03:17:20 CET

- reply

x86_64 windows laptops are still kinda heavy, kinda hot, unreliable, sluggish, but most of all the main thing I hear from actual Windows users is that sleep just doesn't work at all? Among mac users "my laptop got hot in my bag and it was dead when I got to work" does happen but it's the kind of problem that would make one suspect a hardware problem and bring it in for repair, on Windows it seems to be a weekly occurrence on most machines, particularly those priced in the shitbox category

2026-03-16 03:13:49 CET

- reply

I'm far from an expert over the entire domain, but from my casual interest in it… I don't really think so? The student shitbox laptop has largely been replaced by the "gaming laptop" category, which has a ton of problems right now. The two biggest being "Windows" and "Intel"

2026-03-16 02:05:01 CET

- reply

But if you have people with zero tech experience in your life, who have a kid who doesn't really know what kind of computer they need… I'm not going to tell you that you should never recommend Linux to such a person. But at the *very least* you cannot be recommending that they go bargain hunting for mystery-meat laptops that will "probably work with Linux". You need to find a company like System76 or Framework that will actually help them out if the dang thing breaks.

2026-03-16 02:00:17 CET

- reply

If you think that you can compete with this with a bespoke Linux installation on a few old ThinkPads, you need to figure out a way to provide *all that other stuff* to the people who will be using them. And I wish you would! If you ran a charity campaign to raise money to scale up such an effort for a few local school districts in a particular region, I'd probably donate to it!

2026-03-16 01:56:35 CET

- reply

But if you're trying to source a 50-machine bulk order for a CS extracurricular program, with a uniform hardware profile so that students have a consistent experience, then no, you cannot reliably do that by going around to garage sales and rummaging through bargain bins. You cannot afford to repair all of these units (which WILL have a failure rate several times the average for a new machine) yourself. You can't even afford to troubleshoot them and manage the RMA process.

2026-03-16 01:53:43 CET

- reply

Yes, you could personally get a more powerful computer by getting a refurb 16GB M1 MacBook Air somewhere by bargain hunting. But you will need to hunt; right now on the official refurb store the cheapest MacBook Air is $929. If you're shopping on eBay, now you've got a machine with a ton of wear cycles on the SSD, and dubious amounts of damage.
If you, personally, have the time & energy for that, it *IS* a better choice.

2026-03-16 01:51:09 CET

- reply

Out of all of these I have the most experience with category 4. I have set up labs full of Linux computers on many occasions. I've also done the same for macs. I won't say that macs are universally superior but there are TONS of things about imaging, configuring, provisioning, and authenticating macs that are vastly superior to Linux. If it's to teach a topic that isn't programming or sysadmin, like say graphic design, macOS has huge, huge advantages for legibility to the instructor.

2026-03-16 01:44:46 CET

- reply

What is interesting about the device is not that you *should* buy it—the whole value proposition is that it is a very cheap, but also kinda bad, MacBook—it's that people *will* buy it. A lot. It fills a market gap. The only products that this is positioned against are Chromebooks and iPads; cheap refurb Linux machines are not in the same product category for most potential buyers, and I think the fact that Linux fans do not understand the different categories are endemic to why Linux struggles.

2026-03-16 01:39:43 CET

Macbook Neo Hot Take™, take 2. Earlier I was annoyed at tech reviewers who should *really* know better giving a *really* myopic assessment of its gaming potential. Now I'm seeing another bad take on Fedi, which is "all you Apple shills love this stupid thing, but a cheap Linux laptop would work better, don't buy it". I am much more sympathetic to this but it appears to be missing what is interesting about this device and why people are talking about it at all.

2026-03-15 18:05:43 CET

"Is the macbook neo good for Gaming" the Influencers howl, looking at multivariate equations including the price of RAM and the number of teraflops that a GPU can produce
you dolts, this machine is about to be in the hands of like ten million children. its price point means it is highly likely going to be The Computer for an entire generation. it already plays a huge catalog of casual games just fine. the demo units in the apple stores are all playing Apple Arcade titles with no difficulty.

2026-03-13 07:21:35 CET

- reply

I have ADHD. Which means I am experienced in this process of self-denial. I have time blindness. I run an app that tells me how long I've been looking at other apps, because if I trust my subjective perception, I will think I've been looking at YouTube for 10 minutes instead of 4 hours. Every day I need to deny my subjective feelings about how using software is going, in order to function in society.

2026-03-13 07:11:47 CET

- reply

The very fact that things like OpenClaw and Moltbook even *exist* is an indication, to me, that people are *not* making sober, considered judgements about how and where to use LLMs. The fact that they are popular at *all*, let alone popular enough to be featured in mainstream media shows that whatever this cognitive distortion is, it's widespread.

2026-03-13 06:57:16 CET

- reply

1. YES THEY ARE.
They are vibe-coding mission-critical AWS modules. They are generating tech debt at scale. They don't THINK that that's what they're doing. Do you think most programmers conceive of their daily (non-LLM) activities as "putting in lots of bugs"? No, that is never what we say we're doing. Yet, we turn around, and there all the bugs are.
With LLMs, we can look at the mission-critical AWS modules and ask after the fact, were they vibe-coded? AWS says yes https://arstechnica.com/civis/threads/after-outages-amazon-to-make-senior-engineers-sign-off-on-ai-assisted-changes.1511983/

2026-03-13 06:55:11 CET

- reply

But, as Cory puts it:
"""
It is nuts to deny the experiences these people are having. They're not vibe-coding mission-critical AWS modules. They're not generating tech debt at scale.
"""
I had a very visceral emotional reaction to this particular paragraph, and I find it very important to refute. Here are two points to consider:

2026-03-13 06:45:03 CET

- reply

I'm open to a future where we do some research and figure out the limits of how AI influence works, and where the safety valves are, and also the extent to which it's *fine* that AI can influence our views because honestly many different kinds of stimuli can influence our views, not least of which is each other. But it sure looks right now like it has a bunch of very dangerous feedback loops built-in, and it's not clear how to know if you're touching one.

2026-03-13 06:43:41 CET

- reply

Now, for rhetorical effect, I'm obviously putting this fairly dramatically. Cory points out that people have been doing this *to each other* mediated by technology, in emergent and scary ways, with no need for AI. He shows that people prone to specific types of delusions (Morgellons, Gang Stalking Disorder) have found each other via the Internet and the simple availability of global distributed communication has harmed them. But obviously that has benefits, too.

2026-03-13 06:39:54 CET

- reply

For me, this is the body horror money quote from that Scientific American article:
"participants who saw the AI autocomplete prompts reported attitudes that were more in line with the AI’s position—including people who didn’t use the AI’s suggested text at all"
So maybe you can't use it "responsibly", or "safely". You can't even ignore it and choose not to use it once you've seen it.
If you can see it, the basilisk has already won.

2026-03-13 06:38:00 CET

- reply

More to the point though in this metaphor where you're getting a potentially-infected scrape at work, we are living in the pre-germ-theory age of AI. We are aware that it might be dangerous sometimes, but we don't know to whom or why. We are attempting to combat miasma with bloodletting right now, and putting the miasma-generator in every home before we know what it's actually doing.

2026-03-13 06:35:36 CET

- reply

If I could use another inaccurate metaphor, AI psychosis is the "instant decapitation" industrial accident with this new technology. And indeed, most people having industrial accidents are not instantly decapitated. But they might get a scrape, or lose a finger, or an eye. And an infected scrape can still kill you, but it won't look like the decapitation. It looks like you didn't take very good care of yourself. Didn't wash the cut. Didn't notice it fast enough. Skill issue.

2026-03-13 06:32:34 CET

- reply

Cory also correctly points out that "AI psychosis" is probably going to be gatekept by medical establishment scicomm types soon because "psychosis" probably isn't the right word and already carries an unwarranted stigma. And indeed, I think the biggest problem with "psychosis" as a metaphor is going to be that the ways in which AI can warp our minds are mostly NOT going to be catastrophic psychosis, and are not going to have great existing analogs in existing medical literature.

2026-03-13 06:29:50 CET

- reply

The "critic psychosis" thing is tedious and wrong for the same reasons Cory's previous "purity culture" take was tedious and wrong, a transparent and honestly somewhat pathetic attempt at self-justification for his own AI tool use for writing assistance. It pairs very well with this Scientific American article, which points out that pedestrian "writing AI tools" influence their users in subtle but clearly disturbing ways. https://www.scientificamerican.com/article/ai-autocomplete-doesnt-just-change-how-you-write-it-changes-how-you-think/

2026-03-12 04:18:06 CET

- reply

I think I can follow you up to the 5th 'type', specifically the parentheses. I am _so_ curious what you're doing now

2026-03-04 18:52:58 CET

- reply

I have struggled my entire life to relate to incurious people and one of the biggest challenges of my life remains understanding those who do not seek to understand

2026-03-04 18:04:49 CET

- reply

it’s definitely ideologically aligned. I didn’t read it too deeply (partially because i have been just generally distracted, but also) because it echoed another one of my faves, https://blog.nelhage.com/post/computers-can-be-understood/ particularly “building mental models” .

2026-03-04 07:51:02 CET

I've seen a lot of people talking about the practice of code review in a way that I consider unrealistic lately. So here are my thoughts on what code review is—and isn't—for: https://blog.glyph.im/2026/03/what-is-code-review-for.html

2026-03-01 20:39:21 CET

the AI alignment problem is entirely a smokescreen designed to distract from the capital class alignment problem

2026-02-27 21:22:44 CET

never have my biases been confirmed so hard
https://www.gelliottmorris.com/p/new-poll-democrats-real-problem-isnt

2026-02-20 19:04:55 CET

- reply

One way to look at this is to say "oh, algorithmic feeds make people more racist" but the way that attitudes are being measured, the entire way that attitudes *work*, is actually showing something different here: what algorithmic feeds do is *allow racists to efficiently find each other*. "platforming" in this context is not allowing people to hear racist ideas, it is allowing people to *build a command and control network for white supremacist violence*.

2026-02-20 19:03:21 CET

RE: https://hachyderm.io/@mekkaokereke/116103661147175039
Mekka's take on the methodological implications on the (lack of) cross-tabs on this study are on point, but there's another thing to look at here: our definition of "platforming". So much discussion of "platforming" is conducted from the perspective of "are these ideas dangerous, is it OK to let people hear these dangerous ideas". That's not what is happening. The speech acts involved are not "conveying ideas" and letting people analyze them.

2026-02-15 01:47:55 CET

- reply

I don't feel like I delayed _so_ long on this that you had time to move.
But sure, if I manage something, you'll see it. Remember to like & subscribe

2026-02-04 20:52:24 CET

- reply

(Also also: very excited to see if you write something about CORS, your style would be EXACTLY what I need every time I bump into this problem again :))

2026-02-04 20:43:24 CET

- reply

(Also, speaking of misremembering things, I vaguely recall that there are some reasons why it might not even *send* the preflight at all? Caching, or something? Also extremely hard to figure out, although this might be more of a 'networking' thing than a 'browser' thing)

2026-02-03 21:38:52 CET

- reply

silent and confusing failure modes. it's annoying to set up a browser in such a way that you can even see the headers at all, let alone exactly what the browser is parsing, how it's matching, etc. No way to articulate to any debug tool I've ever used, "I expect *this* to work, tell me why it isn't"

2026-01-31 01:20:17 CET

Okay, the “wrapped” mania has officially gotten _completely_ out of control. I just received an official electronic mail from a medical equipment manufacturer, which read, in part:
“Your 2025 [redacted] is a personalized year-end recap that turns your [HIPAA protected data] into an inspiring story of progress and motivation you can be proud of. It's a one-of-a-kind narrative tailored to you”
If you are on a tech or marketing team who is writing copy like this: CALM DOWN.

2026-01-24 20:29:18 CET

- reply

Contract terms slowly grow liability limitations, forced arbitration and other customer-hostile terms. Irregular expensive upgrades with big new features designed to be exciting for users are smoothed out into nice investor-palatable annual recurring revenue. Even EULAs, as gross as they are, have the rough edge of needing to "accept" in order to use the software, mostly tumbled down into terms of service, which you implicitly re-agree to whenever you touch anything

2026-01-24 20:25:45 CET

- reply

Users' ability to decide to choose something else is a form of market discipline; and market discipline is — from the perspective of a corporation — pain. So there is constant subtle pressure to adjust the environment to avoid inflicting this pain, and to adjust the corporation itself to build scar tissue around points where it might be applied so it won't affect their behavior.

2026-01-24 20:23:42 CET

- reply

This is an aspect of @npub1fdr…lvhs 's enshittification: market actors' inherent resistance to discipline. When users experience software as a *product*, where users must consciously make a choice to purchase new versions, the the people *within* the software-making apparatus who care about things like care and craft and details and delight, can advocate for those things on that basis. "We should make this good, so that the user's next purchase decision is informed by that experience"

2026-01-24 06:56:52 CET

- reply

happy to answer any more specific questions if they come up!

2026-01-24 06:48:28 CET

- reply

any feature that requires a backend will prompt you to get one, so it’s not entirely surprising. technically you’ve got one if you’ve ever bought a single app, but of course you can disable many of its features if you want.

2026-01-24 04:26:06 CET

- reply

this implies to me that you have done something that requires iCloud and it is trying to complete the setup for you. did you set up a HomeKit device, maybe? or did you buy an app on the Mac App Store? I have several VMs with no iCloud at all and it has never bothered me like this

2026-01-07 02:22:26 CET

- reply

read the toot, thought, "ugh this is just so silly, we do NOT need this many prefixes, functions are fine, what could this possibly be" clicked through to the PEP, instantly thought "dammit, I *would* use this all the time"

2025-10-07 21:22:40 CEST

- reply

but the scary part of this article is the bit where vibe coding *reads* to corporate interests as an alternative to open source, in that you can externalize your infrastructure development and maintenance costs onto OpenAI's VC investors instead. slightly higher overhead per developer, but no need to deal with pesky human beings who might start to agitate for more resources, so the reduction in hassle is worth the cost

2025-10-07 21:20:43 CEST

- reply

corporations have been reluctant to "give back" because while it can produce good press, if you start "giving back" to the "community" too much, then the cost savings you got from externalizing your complement starts to erode; if you're going to give money to some tech, might as well own it

2025-10-07 21:19:46 CEST

- reply

bluntly, POSIWID: "open source" is a social system for corporations to externalize infrastructure costs, and, materially, not much else. there are of course principles involved and possible future benefits, but *today*, that's mostly what is going on in the "community"

2025-10-07 01:50:53 CEST

- reply

But, filmmakers making films today barely even know what old-school analog glitching actually looks like and even when they do, they can't integrate it into digital tech in a way that doesn't look silly or anachronistic. So they invent imagined (one might say "hyperreal") visual effects that kinda look like what we might *imagine* a partial digital glitch might look like. if a computer got "hacked" or was having trouble with some communications data like a call or video.

2025-08-25 18:27:36 CEST

Should I do some long-form writing about the relative merits of ActivityPub and ATproto (and the ideological positions of their adherent factions), as I have been able to understand them thus far?
I have felt for a while like there is a lot of noise on this topic, and obviously after this weekend that feeling has intensified, but I'd really like a read on whether y'all would think that my contribution would cut through it or add to it.

2025-04-24 05:57:03 CEST

Once again I am asking what are these things actually good for https://bsky.app/profile/glyf.bsky.social/post/3lnjrqdyzl22i

2025-04-16 21:27:09 CEST

Watching this shit rolling out while watching the Fediverse's MAU gently decline month over month on FediDB and every Bsky stat estimate showing explosive growth is just depressing. https://mastodon.social/@grishka/114348720237318718

2024-12-19 20:27:41 CET

- reply

"number" is a broad category as defined by this test, which notably includes '⅕'

2024-11-18 06:12:35 CET

- reply

I remember when Tim Berners-Lee did this as a keynote

2024-11-06 00:11:16 CET

- reply

I do have a Parallels install, as well as CrossOver, and both work pretty well these days. I haven't played a game start to finish on one of them yet, but I think I could. But… the performance does still leave something to be desired.

2024-10-29 23:47:04 CET

- reply

bluesky funding, the latest Musk outrage lead to a huge boost in bluesky user count, while mastodon stayed flat. Plus there has been a huge amount of posting about plushes, which comes across as poorly timed at a moment where mastodon is so badly losing the race for twitter refugees

2024-09-08 00:57:28 CEST

- reply

but also it's hard to find precise, technically-correct terminology that captures these vibes properly

2024-09-08 00:57:11 CEST

- reply

yes that was kinda my point, there's no such thing as a "compiled" language, and rust actually does have a runtime, I mean, they even have a page in the documentation about it https://doc.rust-lang.org/reference/runtime.html

2024-09-08 00:53:00 CEST

- reply

interesting that "compiled" and "with a runtime" are the opposite ends of the spectrum here :). (I get what you're saying, but it is _possible_ to do this with Python, and a real shame that "rewrite in a different language" is actually an easier path than just using existing tooling)

2024-09-07 06:55:56 CEST

- reply

Even in the best case, Rust is more expensive and difficult to maintain, not to mention "non-native" to the average customer here (who presumably knows enough Python to dip into helping out with maintenance if Ofek or Frost wanders off.

2024-08-05 18:40:30 CEST

- reply

yeah I get that this is a tough line to walk. I am still over there too for a complex mix of reasons, although I don’t use it much any more. And sometimes I log in and the horrors on display shock the conscience in exactly the way you just felt. I just want to scream at somebody.
On the other hand, gamergate was a decade ago, so it’s not like it was all sunshine and roses when I was over there regularly. The floor was just tilted to a less obvious degree.

2024-08-05 18:30:14 CEST

- reply

like I also still think that on balance the tradeoffs of the Fediverse are far better but listen to any Black user of both platforms and they will tell you racist abuse is worse here, as well as some clear structural reasons why that’s not an accident, it is due to designed properties of the system. It’s not a pure win.

2024-08-05 18:27:26 CEST

- reply

while I understand the desire to keep up the pressure to divest from the mechanism of fascist propagandizing, I think it is important to keep that pressure gentle and empathetic. Consider: you and I both live in the USA, a country whose crimes *vastly* exceed the troubling dalliances of Twitter as a platform. Twitter has advocated for torture but the good old US of A has actually *done* it. So why are we still here? Cost of moving is high, no alternative is perfect. Same as for Twitter.

2024-06-18 00:21:18 CEST

- reply

I absolutely do not have the energy for a big fight about this and thus have not been yelling about it publicly, but, sigh, YES. Just pass the bylaws changes, there are clear problems that need addressing here, and if we need another round of changes later, sure, whatever, we can refine this mechanism. No need to delay another year or whatever to start improving things.