[object Object]

infrastructure admin for awful.systems, where the jank is a feature (tm)

lemmy: self (npub1k5y…8y2p)
writefreely: https://gibberish.awful.systems/self/

Public Key

npub1xle0wdgqh6470rg5a4jrql3kgruvnd9v4vft686jvec9k0qmde0qvp27vc

NIP-05 Address

zzt_at_mas.to@momostr.pink

Profile Code

nprofile1qqsr0uhhx5qta2l8352w6eps0cmyp7xfkjk2ky4arafxvuzm8sdkuhspz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0ds4akasq

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub1xle0wdgqh6470rg5a4jrql3kgruvnd9v4vft686jvec9k0qmde0qvp27vc

Show more details

Last Notes

2026-04-18 12:27:04 UTC

- reply

please, computer touchers, sneak into art school and befriend even a single artist
art and the process that goes into making it isn’t at all like what you think it is
if you think art can’t serve a function or that it doesn’t take a fuckton of planning and work, you’ll be fucking shocked by what you’ll learn working alongside artists
what you learn will make you a better person

2026-04-18 11:27:16 UTC

- reply

every long-running personal project I’ve ever done has been built on code I consider beautiful
I design fucking languages and frameworks on the basis that I’ve found a more aesthetically pleasing way to do something, because if I find it beautiful then it’s probably worth doing
the only time I write ugly code is when I’m forced to by my employer or when I’m solving a problem that shouldn’t exist
what does that tell you about your code, asshole?

2026-04-18 11:14:36 UTC

- reply

I’m very curious
when a mathematician calls an equation or a method to solve it beautiful, what do you think they’re referring to?
when I doodle lambda calculus in my little notebook and line up all the transformations just so, do you imagine I’m doing it for any reason other than because it looks and feels aesthetically nice? my brain likes it even though it’s not useful.
when I write my Haskell like an equation on a whiteboard and my lisp like a sentence, what do you imagine I’m doing?

2026-04-18 10:52:09 UTC

- reply

we exist in a world with poetry
where the way words are organized is part of the beauty
how can the same not apply to code?
do you not have eyes?

2026-04-18 10:47:53 UTC

- reply

like, it’s actively depressing to me that there are people who program who have never looked at code and found it aesthetically pleasing
we really shouldn’t let anyone like that dictate what code is, because it’s clear they don’t understand. I don’t let a finance bro choose what I read (they’d tell me to ingest some airport lounge cult self-help nonsense) or what art hangs on my walls (“I don’t know, what’s expensive?”), why would I let the computer equivalent tell me about beautiful code?

2026-04-18 10:38:26 UTC

as much as I hate the industry surrounding it, I cannot imagine hating my own craft so much that I’d ever reduce it to this
it feels unfair that I know so many people who agonize over beautiful code and suffer from imposter syndrome but this little fucker from nvidia can argue against the existence of beauty in what we do because capital demands it and his disdain for what he does reinforces it
code is art, and LLM code is ugly, regardless of what capital wants https://nondeterministic.computer/@mjg59/116424732252818630

2026-04-17 22:56:48 UTC

- reply

“it’s just a column in a database” said presumably a full grown adult whose ability to live under capitalism is a column in their bank’s database

2026-04-17 22:10:42 UTC

- reply

“it’s just a column in a database, aren’t you being dramatic” thanks fucker, believe it or not I know how software works and I know perfectly well the type of horrid crap we can build on top of a simple backend if it’s encapsulated in the right type of system

2026-04-17 21:08:49 UTC

- reply

also, is it not plainly fucking horrid that your account creation experience on linux of all fucking things is getting worse due to an unjust, unconstitutional (in the US) law? are you not pissed that muscle memory and scripts will break and command line account creation will get much worse due to this? no objections from the fuckers who campaign against any improvements to the status quo on this one? no? only when doing so fucks over someone else? how weird!

2026-04-17 21:06:08 UTC

- reply

if your proof is that I should trust that the people currently complying will suddenly grow a backbone and it’s actually very punk rock to have a database field with PII in systemd: go fuck yourself

2026-04-17 21:04:32 UTC

- reply

feel free to fuck off with the “it’s just a field in a database, distros will make it a simple age gate during account creation” horseshit unless you come with an explanation of why that’ll work in a regulatory landscape where porn sites with age gates are currently under legal threat from states where an age gate isn’t a sufficient proof mechanism, to the point where some of them have started partnering with companies like (Thiel-backed) Persona for identity verification

2026-04-17 19:49:30 UTC

“software can’t just ignore laws it doesn’t like” it literally can. corporations do it constantly and I really doubt any of them will drop linux if it doesn’t comply with a set of godawful fascist age verification laws. historically one of the forms of pushback against unjust laws is to show some basic fucking solidarity and do nothing to assist in their enforcement because it really isn’t practical to sue everybody, but unfortunately solidarity is alien to most of these computer fuckers

2026-04-15 19:08:30 UTC

¿desayuno? ¿por que no desaydos?

2026-04-14 08:35:17 UTC

- reply

if LLMs are just AI tools that are safe in the hands of experts, then a ban only ensures you’ll get the quality you claim exists
if they’re just tools then banning LLMs is like banning emacs: if you don’t spew config noise across the repo (and you really shouldn’t for most projects) and your output isn’t fucked up in some identifiable way, then nobody will know what you’re using
if on the other hand I’m right, then banning LLMs is the only sane choice if you’d like to keep doing FOSS at all

2026-04-14 08:29:03 UTC

“but if we ban LLM code people will just lie about not using LLMs” good. when the AI corps and the proprietary software houses pull the most choreographed decapitation in history and try to sue your project out of existence, you’re better off if you can show evidence that you didn’t willingly incorporate their crap into your codebase.
also, if the high quality commits the boosters won’t shut up about exist, a ban on anything that looks like poorly understood slop is a good filter for the rest.

2026-04-14 00:21:40 UTC

post solutions to linux problems I don’t have and am not asking about in this thread

2026-04-09 19:25:51 UTC

- reply

some asshole doing free marketing for anthropic decided my post was worth screenshotting as an example of what’s wrong with the AI critics but they utterly failed to engage with any of the points I made, as if pushing back on this obvious shit at all is the problem
anyway here’s a citation for the Firefox thing because no that wasn’t just me talking, the grift really is that fucking obvious
https://media.mas.to/media_attachments/files/116/376/387/634/086/125/original/c8a39fa88e738909.jpeg

2026-04-09 18:00:20 UTC

“why won’t the AI haters admit claude mythos is good?” when it turns out the exploits it found are utterly overblown (ie the firefox exploit that only works on a custom build with the sandbox disabled amongst many other non-exploitable bugs), were found at extreme expense, and required a ton of human staff to verify (just like with existing non-LLM techniques), why won’t you admit this is a grift? why won’t you admit you’ve been falling for the same grift since 2019?

2026-04-06 19:49:31 UTC

your engineering job now involves smoking 3 packs of cigarettes a day and bragging about how much it costs and if you don’t you’ll be left behind

2026-04-01 20:49:41 UTC

Serious Sam seems like the kind of guy who sends money to his grandma (complimentary)

2026-03-22 03:10:07 UTC

- reply

at this point I have no remaining patience for the “if you don’t like it just fork it” crowd. patches don’t fix any of this. the only thing that does is to refuse; to get angry and organize. none of these people deserve your advocacy or your loyalty. they do not give a shit about technical merit, the FOSS ecosystem, your contributions, or you. there is only one goal, and it’s very clear what it is.
end

2026-03-22 03:08:13 UTC

- reply

they are fully aware that said laws disproportionately punish individual online creators. they must be aware of these things, as they are presenting themselves as experts on the downstream effects of age verification when they state that it’s just like an age gate that can easily be bypassed.
you must understand that the unfairness is the point. you must understand that you are the individual, marginalized online creator who will be punished. that is the point of all this.
5/

2026-03-22 03:05:00 UTC

- reply

when Dylan M Taylor (the author of the age verification code in systemd, Ubuntu, and Arch, and a defender of Google’s dreadful new restrictions for Android apps) and others in his wake compare his age verification implementation with an age gate on an adult site, they know full well that adult content online is a gray area rapidly verging towards illegal as US states and other repressive regimes implement age verification laws.
4/ (Taylor cont.)

2026-03-22 02:58:30 UTC

- reply

systemd has always been a lever of power; there’s no other reason to create an ecosystem of its shape. the person who grasps that lever is Lennart Poettering, and it always has been. I don’t need to write much on this; I watched the “oh shit” moment last week when systemd started accepting slop code, and again when an age verification mechanism was imposed on every systemd user and distro on Poettering’s final word. this was always the social structure on offer, enforced by a rigid ecosystem.
3/

2026-03-22 02:52:50 UTC

- reply

Bryan Lunduke has always treated FOSS like a right-wing political project. he has also always misrepresented his goals as technical ones. the current moment is very convenient for him: he is using the valid need for an alternative to systemd to funnel people into fascist-controlled software ecosystems like Devuan and Artix. under no circumstances will Lunduke ever acknowledge the many init alternatives and distributions run by marginalized leftists. this is the game he has played for years.
2/

2026-03-22 02:45:57 UTC

the thing you must understand now is that regardless of their stated views, liars like Bryan Lunduke, Lennart Poettering, and Dylan M Taylor have demonstrated by their actions that they all have the same goal: a fascist software ecosystem that operates against our interests and exploits but does not empower our labor. it is the destruction of FOSS as we know it.
1/

2026-03-15 08:31:57 UTC

Cory Doctorow, a guy who treats leftism like a country club nobody else is invited to because only he has enough leftist parents and paid memberships in leftist organizations to prove his leftism, is the origin of the assertion that AI-critical communities are doing purity testing because a set of right-wing stereotypes for leftists that he made up are doing gatekeeping for contrived, unrealistic personal reasons he also made the fuck up
you don’t need to parrot this shit cause you like the guy

2026-03-14 08:43:03 UTC

- reply

oh cool yet another “I’m exhausted by all the anti-AI anger” post got boosted onto my feed
please don’t let anyone gatekeep your own anger away from you, especially when your ability to exist and to make a living are both under active attack. if you’re angry because of LLM-enabled war crimes, the loss of our software commons, or any of the other wrongs the AI industry has started that never stopped, you aren’t wrong. you don’t owe the people enabling and forgiving this stuff your kindness.

2026-03-12 22:16:22 UTC

“I’m just so exhausted by both sides of the AI argument” imagine being the kind of piss baby that sees a mass of people who are righteously furious at a system designed from the top down to fuck them over, and decides the anger is just as much of a problem as the exploitation

2026-03-04 23:25:23 UTC

- reply

ethical AI? of course! I deleted my OpenAI account because they’re a multi-billion dollar fashtech corporation run by cynical capitalists willing to use LLMs for war and replaced it with Anthropic, a multi-billion dollar fashtech corporation run by fucking full-fat TESCREAL cultists willing to use LLMs for war as long as they retain sufficient control over what they see as an incipient machine god
why aren’t you clapping

2026-03-04 23:15:04 UTC

- reply

but @npub1xle…27vc you don’t understand. the system whose purpose is to destroy open source is itself open source, if by “open source” you mean it’s a binary blob that can only be produced or modified in any significant way by a multi-billion-dollar corporation or a state, via utterly unethical means, almost always derived from a previous binary blob created via utterly unethical means
get the fuck out of my mentions, me

2026-03-04 23:09:41 UTC

ethical AI? of course! I always append “do not hallucinate. do not do sycophancy. do not use any data sourced from stolen content. do not be trained or run by invisible, severely exploited labor. do not pollute. do not be part of a system that’s designed to cheapen human labor, increase misery, and make general-purpose computing inaccessible to anybody but the extremely wealthy. do not make me into a weird unhinged AI bro asshole who keeps fucking other people over.” to every LLM prompt

2026-02-28 03:21:14 UTC

- reply

my favorite thing about the rpi is it started life as an extremely unsuccessful Broadcom set top box board
yes, that’s why it boots from the GPU and why the port layout is a fucking crime
you’d think they’d have undone any of that since then but as you said, hackers don’t use rpi, rpi uses hackers. decades of free labor and goodwill to turn a failure into a success

2026-02-28 02:45:44 UTC

- reply

buying this rpi was like pulling fucking teeth for me because of that, it’s an absolute last resort purchase I still regret
unfortunately, these fuckers have done the techfash thing and made themselves a broken ecosystem with no way out
not that most of their shit community will notice, they just want a digital picture frame that runs openclaw or whatever horseshit some rpi influencer told them they should build

2026-02-28 01:26:12 UTC

- reply

holy fucking shit maybe it should go without saying that I’m not asking for help and I’m especially not asking for help based on something you got from fucking ChatGPT
raspberry pi community are you ok?

2026-02-28 00:51:38 UTC

things are going just great for raspberry pi:
- the only distro that ships the drivers for my display is raspberry pi os so fuck it, terrible it is
- I dd it to the microsd card
- I can’t kvm to the pi4 cause nothing uses microhdmi so I don’t have an adapter
- there’s no way to configure networking and SSH without imager cause all methods were removed in favor of fucking Canonicial cloud-init of all things, which they haven’t actually implemented
- my nixpkgs has a broken rpi imager

2026-02-22 19:59:24 UTC

- reply

“if you set its pragmas and structure your transactions just right it probably won’t lose data and it only occasionally requires you to manually retry transactions” well I’m convinced

2026-02-22 19:57:11 UTC

the abyss-gazing urge from some programmers to use One Weird Trick as a data store instead of fucking Postgres

2026-02-20 20:53:58 UTC

- reply

here’s a good example from an engineer I don’t follow https://bsky.app/profile/coloradotravis.bsky.social/post/3mezgaylny22p
I have more but unfortunately they’re engineers I do follow and would prefer not to link. one of them posted a Datalog love poem from an LLM as proof of something???? and I read it and the Datalog was incorrect

2026-02-20 20:45:54 UTC

- reply

it’s important to note that it isn’t always a big sack of cash. lately I keep seeing this pattern happen with engineers:
- “as an AI skeptic I finally have empirical proof that LLMs are good/useful/thinking/feeling <posts slop>”
- “uhhh are you ok? I checked the LLM output you posted and it doesn’t make any sense if you dig in at all and the citations are all fake”
- “this is empirical proof and you’re being emotional.”
this is engineer brain. Doctorow isn’t an engineer, so sack of cash it is.

2026-02-20 20:03:19 UTC

oh someone finally offered cory doctorow a big enough sack of cash to do a cory doctorow? I’m shocked

2026-02-16 21:01:20 UTC

- reply

but speaking of rolling out half-baked shit to users who didn’t ask for it, attached is all the documentation I could find for whatever the fuck the New Classic Window item is
the New Classic Window item disappeared when I disabled extensions.ml.enabled so it’s related but fuck if I could tell you how
https://media.mas.to/media_attachments/files/116/082/322/953/463/730/original/cca133d4cca04342.jpeg

2026-02-16 20:59:53 UTC

- reply

let’s look at mozilla’s marketing blog: https://blog.mozilla.org/en/firefox/ai-window/
>Completely opt-in, you have full control, and if you try it and find it’s not for you, you can choose to switch it off.
haha no, I did nothing to opt into this. I opted out as strongly as I could and it still showed up to displace a menu item I frequently use. disabling it in my version of Firefox required yet another dive into about:config.

2026-02-16 20:57:52 UTC

here’s a worked example of why nobody believes firefox’s AI kill switch will be anything but a marketing wedge
these showed up in my File menu, displacing the New Private Window item and fucking up my muscle memory, in spite of having browser.ml.enabled set to false. the setting for this is a new nonsensically named option, extensions.ml.enabled. a merry fuck you to the guys who showed up in my replies telling me to just disable the config flag I already disabled months ago.
https://media.mas.to/media_attachments/files/116/082/337/971/150/735/original/3921f57832c1c9c2.jpeg

2026-02-11 23:35:41 UTC

- reply

unfortunately, this now appears to be paywalled
https://media.mas.to/media_attachments/files/116/054/618/324/857/772/original/428b37052c5707b9.jpeg

2026-02-10 09:01:19 UTC

- reply

come on guys, I gave you such a softball with this one. I didn’t even talk about how your stated ethics are neoliberal claptrap desperately searching for a non-paradoxical definition of a good cop who can never exist.
all you gotta do is tell us explicitly you won’t work with ICE and give us an example of the consequences for a public institution that violated your ethics policy.
yes you can share that exact information; you’re receiving public money for your services.

2026-02-10 08:36:54 UTC

- reply

in the context that these are public institutions and that you’ve been very public in supporting them with partnerships and technology: are you able to show me a case where an organization you partnered with violated items #5 or #6 in the link you posted and forced you to end the partnership?

2026-02-10 08:30:05 UTC

- reply

this is just a page that says “unable to decrypt message”
more seriously, instead of giving a generic “this is not the case” in response to aral’s post that included receipts of the partnerships you’ve entered, are you able to confirm in writing that you will not provide services or technology to ICE or its affiliates? if one of the American police departments you have partnered with has become affiliated with ICE, have you done anything to end that partnership?

2026-02-07 23:27:21 UTC

- reply

“LLMs are natural language compilers”, brought to you by the same kids insisting their product is “the operating system for the web” because nothing means anything if you ignore all implementation and engineering details

2026-02-07 23:25:09 UTC

- reply

it’s always a bit depressing when I find out about a new pocket of mediocre tech jackasses posting twitter crap on masto. all of the guys posting “LLMs are like compilers for natural language” should have their CS degrees yanked cause they’ve proven they don’t meet the academic requirements for a CS undergrad.

2026-02-07 22:28:02 UTC

>When you create a tab group, we use generative ML to understand the tab titles and suggest a name for the group. We use generative AI to understand what other tabs are open and suggest related tabs be added to the group.
why do Firefox users hate our generative AI features? all we do is spin up your GPU or many of your CPU cores and download a large database of plagiarized work to do trivial shit that was already solved better by an edit distance or your choice of relevancy algorithms

2026-02-03 21:20:17 UTC

- reply

in brief, please do not interact with Jake or the mozilla dev account if you may suffer reputation or career damage as a result. anything you post directed towards Mozilla as a company will be framed as a personal attack on a Mozilla employee. please consider the amount of power their company’s money and position affords them. as you can read in the full thread, even relatively innocuous posts will be framed in the worst light possible if Jake disagrees with you.

2026-02-03 21:16:44 UTC

- reply

also disgusting is how Jake frames the accusation. he left an emoji in so he could claim it was a joke, but it was posted in a context where a joke isn’t appropriate. the other poster isn’t Jake’s friend. the “pretty grim” follow-up, “I’m not going to debate this further”, and blocking the original poster make it pretty clear it wasn’t tongue in cheek at all.
Jake was hoping to shape drama. unfortunately for him, we can read.

2026-02-03 21:10:51 UTC

- reply

because Jake didn’t like the form of the feedback directed towards an official account, he decided to frivolously associate that other poster with a false rape accusation. if that other poster’s account is associated with their real name or their work, they now have this accusation associated with their identity.
this is a fucking gross, dangerous thing to do in response to what is, in reality, a garden variety piece of angry feedback.

2026-02-03 21:07:29 UTC

- reply

here’s the quoted post in context. while again it’s not something I would post, I will not be the one to tone police the feedback directed towards a billion-dollar company that is, in fact, fucking us over while pretending to represent us.
if your response to this is that the rape comparison should never have been made, please consider the amount of power Jake has in this interaction.
https://media.mas.to/media_attachments/files/116/008/737/071/553/562/original/f57811c1816eb0dc.jpeg

2026-02-03 21:04:06 UTC

- reply

that unfortunately brings us to the danger. here’s the bits of that earlier thread that didn’t fit in the screenshots.
while the quoted post isn’t one I’d make personally, I’d like to point out something important: it’s directed towards an account that represents Mozilla, a billion-dollar grift-shaped nonprofit. Jake isn’t tagged or mentioned anywhere near that post.
Jake has turned a crass piece of corporate feedback into a personal attack.
https://media.mas.to/media_attachments/files/116/008/723/605/177/578/original/0a33490539ef4e8f.jpeg
https://media.mas.to/media_attachments/files/116/008/723/693/945/942/original/ff7d35617382178a.jpeg
https://media.mas.to/media_attachments/files/116/008/723/783/590/727/original/b9b6962150e17b99.jpeg

2026-02-03 20:49:33 UTC

- reply

part 2 of that last one. “you’re making a mountain out of a molehill” / “show me where I accused you of being irrational” is a pattern some may recognize on a personal level
https://media.mas.to/media_attachments/files/116/008/666/471/958/391/original/5180fcbda9bb0097.jpeg
https://media.mas.to/media_attachments/files/116/008/666/555/959/376/original/e4f0a23748300a59.jpeg

2026-02-03 20:48:11 UTC

- reply

if you’ve been boosting the mozilla account because you think they’re listening to us, please note that anti-AI mastodon users are, by Jake’s admission, categorically being ignored by mozilla.
the attached screenshots also serve as a second example of “show me where I said that”. as I said, he loves this move.
https://media.mas.to/media_attachments/files/116/008/661/071/668/902/original/d8cf51ce12252f0a.jpeg
https://media.mas.to/media_attachments/files/116/008/661/143/795/615/original/b87c28e1905614fa.jpeg
https://media.mas.to/media_attachments/files/116/008/661/215/890/238/original/a91361756eabda1d.jpeg

2026-02-03 20:44:03 UTC

- reply

attached is a thread where Jake willfully misunderstands an obvious use of scare quotes and argues the other party to exhaustion over it.
I’m very certain this type of behavior worked to his benefit when he was paid to stonewall web standards committees. it’s designed to kill conversation — and dissent — dead.
there’s lots of examples of this particular behavior; I chose this one because it’s fresh and easy to screenshot
https://media.mas.to/media_attachments/files/116/008/644/599/129/150/original/3e618575aee59110.jpeg
https://media.mas.to/media_attachments/files/116/008/644/717/869/579/original/be7a2e9bccacc7f7.jpeg
https://media.mas.to/media_attachments/files/116/008/644/845/285/244/original/35d9d6346760d5a5.jpeg
https://media.mas.to/media_attachments/files/116/008/644/926/913/348/original/56792ec70a6082cc.jpeg

2026-02-03 20:39:26 UTC

- reply

to start this where it must start: the (former chrome, current mozilla) developer relations person in question loves the “tell me where I said that” move. it’s something they’ve tried in every negative interaction they’ve had on fedi.
it works as follows: Jake picks out a minor part of your criticism and insists you tell him where he said those exact words. this isn’t how human conversation works, so you can’t. if he needs to willfully misunderstand your post to do it, he will.

2026-02-03 20:35:08 UTC

here’s some unfair warnings if you’re considering boosting or interacting with that one mozilla dev account or mozilla in general really
if you have interacted with them and gotten an icky feeling, you’re not alone
later posts in this thread will be CWed. there is danger in interacting with the account in question. many of these posts feature someone named Jake; as you will see later, Jake and the Mozilla account are the same person in every way that matters.

2026-01-27 18:34:12 UTC

“zzt how dare you accuse Mozilla employees of suffering from very obvious LLM psychosis”
please “enjoy” the State of Mozilla, one of the least hinged things I’ve seen released with pride by an established corporation: https://stateof.mozilla.org/ (epilepsy warning)
note the 2001-style not-Flash intro. the fake captcha. the timecube-esque prattle about AI threatening humanity. “DO NOT ACCEPT DEFAULT SETTINGS”. the ASCII art cop finger in the footer of the main page. all of it. what in fuck is this.

2026-01-10 01:00:13 UTC

ed(1) is no longer the standard editor and should no longer be shipped. the new standard is gTextFucker, a graphical-only wordpad clone with indecipherable keyboard shortcuts that change constantly, which gradually grinds to a halt as it runs because it keeps trying to send messages to the dbus receiver org.gnome.fuck.text which doesn’t exist on your system

2026-01-07 01:40:28 UTC

my brain generated the shitpost “linux for dogs” and then my other brain almost immediately started designing a small embedded linux collar computer with a sensor suite and wifi and now I want linux for dogs

2025-12-23 23:11:24 UTC

if you frame the fascist takeover of an open source project by a major war profiteer and associated exodus of the vast majority of that project’s marginalized contributors (who were the ones keeping the lights on for most of it, as per usual) as drama, maybe you can stay the fuck out of my replies
no I’m not going to CW this or any other post as “nixos drama, future of the project” because I don’t reflexively try to protect fucking monsters

2025-12-04 17:18:28 UTC

there’s an ending in cultist simulator where you accidentally pour too much of your passion into being a middle manager at an office job and forget about the murderous cult you lead, and I think about that a lot

2025-11-10 00:27:55 UTC

“uhm you can’t say a password manager is cooked just because they’re encouraging LLM code contributions and their official mastodon account keeps saying factually incorrect things”
consider fucking yourself. there’s little margin for error in security-critical software. if a team institutes a policy that’s been proven to produce vulnerable code and then says a bunch of things that, read uncharitably, appear to point to them not knowing what the fuck they’re doing, they deserve the backlash

2025-11-09 00:32:17 UTC

- reply

glad to help!
and I think you’re exactly right about the thought process behind the ladybird advocacy. it’s frustrating that servo doesn’t get the same level of marketing and funding, though I realize it’s an uphill battle for a project that isn’t planning to be a lever of control. I’m also not fully convinced that even a well-funded servo project can save the web before one of its custodians does damage that can’t be reversed (ie Google and browser attestation), but I’m willing to try

2025-11-08 22:17:52 UTC

please for the love of fuck don’t have a parasocial relationship with a password manager, an encrypted mail provider, or any other security or privacy software vendor
you absolutely don’t need to defend the honor of a project you like. you especially don’t need to defend a project that’s provably doing something dangerous
when I call out keepassxc or proton or whatever in the strongest terms possible, I do it because it’s important. I do it to create awareness, not fear.

2025-11-05 01:50:05 UTC

- reply

I keep trying to justify wanting to buy a PiDP-10 as a furniture purchase… maybe your furniture budget has room for a PiDP-11?

2025-10-12 09:07:28 UTC

GNU GIMP (1998): we can’t change our terrible fucking name, we’ve had it for years. imagine the confusion. imagine the work.
the Coq theorem prover (1989): we’re Rocq now because we got tired of the joke and we trust you’re too busy doing mathematics to give a shit

2025-10-04 22:19:44 UTC

lmao @ jeff vaguetooting me of all people
no you don’t get the presumption of good faith when you use an unhoused person for marketing purposes and then change your story to make yourself barely the hero what the fuck
(for anyone who missed it, https://mas.to/@zzt/115306806916386973; jeff has since changed his story to a version where he gave his friend $19k and did a bunch of research months before passing it all through an LLM for no reason, which doesn’t make sense) https://infosec.exchange/@codinghorror/115309311041264788

2025-10-02 21:57:24 UTC

jesus christ jeff. this is “I went online and printed off some resources for you about how to stop being unhoused! you’re welcome!” but you then went on to use the encounter to promote slop machines to your audience
please can we stop acting like these neoliberal birthday boys have any idea how to solve problems? jeff’s altruism ends when the tax breaks and the marketing end. that’s why his millionaire ass couldn’t give his “friend” access to a couch or a mailbox. https://infosec.exchange/@codinghorror/115279844443067032

2025-10-01 12:23:32 UTC

- reply

guys will literally make a terrible web framework and turn it into a fascist pipeline before they go to therapy

2025-10-01 12:16:15 UTC

- reply

please "enjoy" the suckless developers doing a torchlight hike in coordination with the nazi Unite the Right rally. one of their core devs says this was just a coincidence! two of the links are that same guy ranting about fucking cultural marxism!
https://tilde.team/%7Eben/suckmore/
there's more but if you don't see it from this, you don't give a fuck about fascism, you're here to do evangelism
7/

2025-10-01 12:11:44 UTC

- reply

Urbit is fractally wrong; every part of it that claims to be original isn't, every time it claims to be distributed it's a lie, and when it claims to work it's almost definitely broken, though you wouldn't know cause it was written by an asshole trying to convert Hacker News types to neoreaction and it worked
ask me about Urbit, I have so many notes
the Chromium fork is Brave, a browser by an asshole that claims to be privacy-focused but which is there to push LLMs and cryptocurrencies
6/

2025-10-01 12:08:11 UTC

- reply

no I won't dig up links, type "hyprland toxic" into a search engine of your choice. other parts of Wayland fucking suck too.
but speaking of fucking sucking, XLibre is a fash X server that also barely works, because the point of fash tech isn't that it works, the point is that it's metaphorically and sometimes literally a weapon. see also the defense contractor currently taking over Nix, but see also the OG fash language, Urbit, the worst functional programming language I've ever known
5/

2025-10-01 12:05:16 UTC

- reply

the fash papertrail for Ladybird really shouldn't have to go much farther than them receiving money from fascists in a group that included other fascists but here's more anyway:
https://possum.city/notes/a504c0e7vwdj000h
I legitimately do not know how this shitty browser got so popular on fedi, but Andreas controls it so tightly he pretty much single-handedly switched all future development to Swift over his own community's objections
anyway, hyprland is the most toxic part of Wayland
4/

2025-10-01 12:01:25 UTC

- reply

also publicly coordinating with DHH on Twitter are the guy who started Tailwind (terrible Fisher-Price CSS) Adam Wathan and the leader of the Ladybird web browser Andreas Kling; both are actively brainstorming with DHH ways to make their projects more hostile to non-fascists. while the Tailwind guy turning fash might be surprising, Andreas has a bunch of fash shit on his Twitter and a history.
now's maybe the time to stop advocating for that unserious non-working web browser
3/

2025-10-01 11:49:16 UTC

- reply

DHH and the fascists taking over Nix being in league might seem too convenient to be true, so here's some links. these are just analysis, but if you click through to Twitter you're going to get hit in the face with bare bigotry so fair warning:
https://fedi.slightly.tech/@joepie91/statuses/01K6AHA0BNBBX616JGDJHYET36
https://fedi.slightly.tech/@joepie91/statuses/01K6E3G2Z19M82BWRB3SPADHP9
https://mstdn.games/@KFears/115286316219357517
the tactics the fascists used to take over nixpkgs and RubyGems are old, and they worked. now that they know their tactics work, they'll be coming for other projects.
2/

2025-10-01 11:45:47 UTC

- reply

by popular demand, here's an answer key for my fashtech vaguepost, in order of urgency.
the package repositories currently being taken over are RubyGems and NixOS' nixpkgs. it's quite well known that DHH, a Shopify board member, is behind the corporate takeover of RubyGems. he's also coordinating openly on Twitter with the defense contractor Anduril and other actors to ensure that nixpkgs meets an even worse fate. this is the culmination of a long project to make Nix fascist.
1/

2025-09-26 20:27:03 UTC

why not embrace fashtech! they’ve got:
- terrible Fisher-Price CSS
- the most toxic part of the Wayland ecosystem, and that’s saying something
- the worst part of the X11 ecosystem, and that’s saying something
- suckless tools and they all fucking suck
- corporate control over the packages you wrote, for your own protection
- distributed functional programming that doesn’t function or distribute

2025-09-10 12:03:27 UTC

it’s fucked how every time I use a tool that calls itself opinionated, the opinions are always some annoying dev’s terrible defaults that they want to make into an industry standard and not anything of value, like how transphobes can fuck off forever
if your priorities include dictating code formatting and exclude keeping bigots out of your project, your priorities are fucked

2025-09-04 18:20:50 UTC

I can refuse to pay $80-$100 for games longer than your studio can remain solvent

2025-09-04 06:21:15 UTC

so the founder of the federated blogging platform Ghost is one of the AI bros that’s tricked themselves into thinking the slop machine makes them 10x more productive in spite of all evidence to the contrary
if your site’s built on Ghost, it might be worth checking how much of the code you’re relying on is slopcoded https://john.onolan.org/.ghost/activitypub/note/751e39e3-365e-4398-830b-7cdf7ceb2882
edit: since Ghost links don’t seem to work properly, it’s John’s post in this thread https://stefanbohacek.online/@stefan/115142592402628516

2025-08-19 21:46:05 UTC

I feel like a chihuahua complaining about small yappy dogs, but why in the fuck is an obsession with lambda calculus such a red flag now
(this isn’t a subtoot about any of the CS researchers I follow that do good work in lambda calculus, but their fans have an above average probability of being weird fascist freaks)
(also a note to the lambda calculus fans who aren’t weird fascist freaks: yes I know, me too, see also the joke about chihuahuas complaining about small yappy dogs above)

2025-08-18 10:18:06 UTC

the state of infosec:
- it’s insecure, it can’t be secure, it isn’t even a good idea
- here’s evidence of wrongdoing that stops just short of being a smoking gun
- the people who make it have admitted it’s not secure
- “nah that’s just FUD”

2025-08-12 16:11:19 UTC

- reply

I’m kind of surprised it wasn’t initially popular — I remember that the users on my Lemmy instance loved the post back then, and they still reference it now.

2025-07-12 20:05:30 UTC

- reply

I know I’m asking for a bad time, but I’d love to find out what the “FPGAs aren’t real hardware, they’re just emulation” elitists think the outer machine is. because emulation’s definitionally one Turing machine acting like another, right? so uh, where’s the other machine? like, you can configure an FPGA’s fabric to function like a wide variety of Turing machines, but those muxes and sram can’t be the outer machine, right? cause if so, all hardware is emulation.

2025-06-19 17:38:47 UTC

can someone please fund my film idea
a fictionalized Wes Anderson, played by film auteur Wes Anderson, and a fictionalized Paul W. S. Anderson, played by schlock film auteur Paul W. S. Anderson, each direct alternating scenes of a movie in which they try to fucking kill each other
Anderson vs Anderson. no matter who wins we lose

2025-06-18 19:34:53 UTC

- reply

supposedly uv2nix will handle conversion but:
- it’s by the same author who abandoned poetry2nix, the previous tool that worked fine
- it’s barely documented
- do I even want uv? it honestly feels like it’s destined to be replaced too, and I’m not seeing a lot of evidence that it or uv2nix are very good for what I’m doing

2025-06-18 19:28:36 UTC

- reply

as a pure consumer of python packages (I convert them to Nix packages, I use them in development environments, I don’t usually author Python packages directly) I’m out of the loop on what was wrong with poetry and feel like uv’s broken my ability to use Python packages at all, because I’m honestly too exhausted to learn the intricacies of yet another Python package installer

2024-12-07 00:46:24 UTC

- reply

for anyone who wants to learn more about how serious this type of therapy is: https://en.m.wikipedia.org/wiki/Proton_therapy (and yeah, it’s more properly called something along the lines of proton beam therapy rather than proton laser therapy, but something tells me this fucking monster of a former executive might not have any real medical training)

2024-12-07 00:42:51 UTC

- reply

> “We’d have times when you’d deny proton laser therapy for a kid with seizures and the parent would freak out,” said the former executive.
holy shit what. this is the therapy they use as a last resort for brain tumors — that kid was seizing because their tumor was killing them
someone please tell me that former executive followed their little anecdote up with “The Aristocrats! haha just kidding imagine being this much of a fucking monster”

2024-12-04 03:37:04 UTC

- reply

I like that! I’ve been meaning to try a non-gaming x68k setup on my MiSTer, but there’s so much to dig into. Human68k had 3 GUIs — VS (the one in your photo), SX-WINDOW (https://en.m.wikipedia.org/wiki/SX-WINDOW), and Ko-Window (Motif). I think I like the look of VS the best so far!

2024-11-29 21:12:25 UTC

- reply

huh, it looks good and I will check it out next time I’m doing documentation outside of an org or latex toolchain and also outside of a work mandate to use something terrible

2024-11-29 21:05:58 UTC

- reply

what I need open source evangelists to understand is that I already know about libreoffice

2024-11-16 22:14:19 UTC

i never thought the leopard language model would generate *my* face

2024-11-14 20:46:17 UTC

- reply

I’ve been so close to defederating from lemmy.world and telling our regulars from there to just make an account on our instance
it’s interesting that a lot of the bullshit reports I’ve been getting lately are for “incivility” from lemmy.world users — but we don’t have any civility rules, nor will we ever, cause they invariably get used against rightfully angry marginalized people

2024-10-27 04:18:41 UTC

- reply

see my favorite part of the movies is when he does weird clown shit and leans way more into it than most horror movie clowns do, so I was legitimately hoping Art was played by someone with clown experience