Notices by Phantasm (phnt@fluffytail.org)
-
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Thursday, 15-Oct-2026 02:33:13 JST 
Phantasm
@snacks @twinspin6 They have other issues.
japanese-hiragana-no-homo.png -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Thursday, 15-Oct-2026 02:27:41 JST
Phantasm
@snacks I don't know about banning, but they did the smug "are you having issues" popup. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Wednesday, 14-Oct-2026 20:23:16 JST
Phantasm
@snacks Traffic fingerprinting. The thing where the video starts, plays for exactly one minute and then stops with 403 has been in their playbook for years. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Wednesday, 14-Oct-2026 03:02:36 JST
Phantasm
@p
> I'm still not certain whether you're talking about workstations or servers or what.
Neither. Hypocrisy and difficulties of maintaining patchsets for outdated and complicated pieces of software leading to giving up on doing so. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Wednesday, 14-Oct-2026 03:00:54 JST
Phantasm
@p
>> Popularity is not what I'm taking about. It's usability.
>"No one will use it" is popularity.
The two terms are entangled, you can't have a software be popular without it being usable (or having no other competition). By a substantial enough decrease in usability you lose popularity. "No one will use it" is not popularity in the way I meant it, it means as in "No one will use it, because it is unusable."
And to remove ambiguity even more, "no one" is a hyperbole meaning negligible amount to the point of irrelevance. No one daily drives Windows 3.1 in $current_year is a valid statement, despite it being very likely false because at least 1 person on the whole planet probably still uses it every day. And if you think that is not valid, we are arguing whether "No one" or "99.9% of people using computers" is big enough difference to mean something in the statement.
>> If Linux had that, its server market share would be way lower than what it is now.
>This is a popularity thing again.
Again, an unusable piece of software cannot be popular in a space with competition.
>Sometimes it means "faster". It's a matter of degree, though: you sit down at a machine and we've written most of the system software in C because it's easier to write a correct C program than the equivalent in assembly [...]
Yes, that is the point I was trying to make. That certain compromises have to be made, and at least in the context of OP (Linux) a lot of what can be called "bloat" are tradeoffs of complexity and security surface for making something fast. There's no denying some of it can be removed with little consequences, but a lot of the complex stuff cannot.
Going back to the 9front example, those tradeoffs don't have to be made because it is a project that doesn't necessarily need them. 9front doesn't need shared kernel-userspace ring buffers for async IO (io_uring) to push tens and if not hundreds of gigabytes per second around. Does Linux need them? Debatable. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Wednesday, 14-Oct-2026 01:33:42 JST
Phantasm
@SuperDicq @icedquinn
>The real workflow intended for git (sending each other patches through email and having the actual repository decentralized) most people don't know about.
The intended workflow also sucks ass. It only works if you have a single person managing a whole subsystem of a project where all the patches to it go against a common base for everyone for that subsystem. Once you start mixing different startpoints from a moving branch, it completely falls apart. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Wednesday, 14-Oct-2026 01:23:40 JST
Phantasm
@PurpCat "People are doing more without asking the annoying community for help. Here's why that is a bad thing." -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 23:45:39 JST
Phantasm
@icedquinn @bajax @demha @sun
>the cryptobros are all about zaps
Nostr basically turned into Xitter 2.0 with the engagement bait for Zap farming. Barely anything genuine happens there at this point. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 21:33:58 JST
Phantasm
@newt @mangeurdenuage In one universe he took the "a few months in federal prison" pill and in the other one he took the easy way out pill. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 11:04:31 JST
Phantasm
@p The amount of people that will do so will be so small that the number of them will be irrelevant. And the longer the "out of date" period is, the harder it will be to maintain the patch sets, provided that the upstream patches aren't "tainted" either, which would be debatable as well at that point. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 10:42:49 JST
Phantasm
>so just a full vibe, done with the LLM that can’t make a web browser, C compiler, rsync features that aren’t riddled with bugs, or JavaScript runtime
Even humans can't make a web browser, C compiler or a JS runtime without a bunch of bugs either. I think the list of Firefox/Thunderbird/Chromium CVEs every few weeks and the miscompilation bug reports in gcc and LLVM found during testing all the time speak for themselves.
RT: https://mas.to/users/zzt/statuses/117429179239146994 -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 10:35:57 JST
Phantasm
@p
>fuck popularity as a benchmark for anything, it's a terrible reason for anything.
Popularity is not what I'm taking about. It's usability. In my synchronous IO example in order to remove complexity, you get a system that is now unusably slow when you do real work with it. An example of that are the giant kernel locks still present in NetBSD and OpenBSD, making it hard to push maybe more than 25 Gbit out of very capable systems. If Linux had that, its server market share would be way lower than what it is now. I couldn't push more than 40Gbit on OpenBSD with virtio drivers which have barely any overhead on 8 cores for example.
You can have simple systems, but simple does not always mean fast enough and sometimes you need a lot of complexity to make something fast or reliable. 9front is small, simple and without bloat. Is it fast? Not really, but its goal is running a small grid and desktop/development use, not serving thousands of requests per second so it does not matter. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 09:50:44 JST
Phantasm
@p
>what if people used systems that weren't bloated monstrosities?
Eliminating bloat reduces the exposure area, yes. But it doesn't eliminate it. At the same time, try making a race-free filesystem API without reinventing it as a DBMS with per-fd views. Even an API that does stat/open/close/read/write and only that has inevitable races and fixing them requires complex monstrosities that guarantee atomicity.
Similarly, you can remove a lot of "bloat" by removing racy async IO in the kernel and do it synchronously, but what you will end up with is something slow people don't want to use. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 09:22:31 JST
Phantasm
I will forever use the pre-slop version of <insert important and large project> is one of the most hypocritical copes ever. Not only will a fraction of the actual people saying that even attempt to do it, but they will inevitably reach the junction of "Do I run an outdated piece of software with a large vulnerability/vulnerabilities, or do I maintain my own patches." At which point the two answers are no and no, so back to running the "slop version" it is.
Have fun applying over a year of security fixes to Linux and core libraries btw. No, even BSDs aren't safe either. :) NetBSD allows LLMs as long as the patch is approved by a core maintainer, OpenBSD already has Claude commits in it via tmux and FreeBSD has an "I don't care" stance. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 07:59:19 JST
Phantasm
@mikemccaffrey I don't know, since Microsoft bought Mojang (which Notch sold), the game has turned into a heap of disjointed ideas and useless cruft. It's as if it was lacking something called "direction", which a certain previous owner would provide. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 07:59:18 JST
Phantasm
@sun @mikemccaffrey It really shows in 1.13 and newer versions. Pillagers and related are the most useless mobs ever, while being very annoying. Phantoms are just there to annoy you to bed and nothing else. Raids serve no purpose other than to annoy you. The cave generation is so bad and boring since 1.18, it is unbelievable. Villager trading got so overpowered to the point of being able to skip most of the game with it, which got nerfed later for obvious reasons.
But hey, at least it has foxes now, which I had in Mo' Creatures in 1.2.5 Technic Pack. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 03:05:59 JST
Phantasm
@icedquinn Github/Gitea/Forgejo? Yeah, it desperately needs a clown emoji react added. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Tuesday, 13-Oct-2026 01:51:14 JST
Phantasm
@icedquinn Qt has GUI testing support in _some_ capacity, but it is not amazing. And every other toolkit is less complete, so you can guess how that is elsewhere. The webdev solution is to start a headless browser and simulate actions by finding the element in the DOM and interacting with it. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Monday, 12-Oct-2026 20:51:40 JST
Phantasm
@papush_ Negativity/drama sells and its hard to find people immune to that. For reasons beyond my comprehension, people want to be angry at something. -
Embed this notice
Phantasm (phnt@fluffytail.org)'s status on Monday, 12-Oct-2026 03:11:08 JST
Phantasm
@kaia @snacks
image.png
All GNU social JP content and data are available under the 
