Phantasm

Professional thread derailer.
Bad posts ahead.

Alts when my instance is broken: @[email protected] @[email protected]

pfp and banner shamelessly stolen from Alessandra Maria

76 69 84 39 83 32 65 76 76 32 76 79 86 69 32 76 65 73 78

Public Key

npub1er26g8v2gdquxkyk43usmlk7gaqhle08p378ld3ldfd6p5yrt8usmfarq5

NIP-05 Address

[email protected]

Profile Code

nprofile1qqsv34dyrk9yxswrtzt2c7gdlm0ywstluhnsclrlkclk5kaq6zp4n7gpz4mhxue69uhhyetvv9ujuerfw36x7tnsw43qz9thwden5te0wfjkccte9ekk7um5wgh8qatzpj09z2

Publishing to

relay.ditto.pub

relay.mostr.pub

Author Public Key

npub1er26g8v2gdquxkyk43usmlk7gaqhle08p378ld3ldfd6p5yrt8usmfarq5

Show more details

Last Notes

2026-07-29 12:17:33 CEST

As the Overgoddess of the Speakeasy and Protectrix of the Speakeasers
Holy larp

2026-07-29 11:10:25 CEST

- reply

@nprofile…d73q @nprofile…tsag PHP is the quintessential bodge it together until it works language. It may not be slop, but it is the source of most we slop ever created.

2026-07-29 10:31:16 CEST

- reply

@nprofile…tsag @nprofile…3wz5 There's nothing especially bad in that FEP, so I'll make it compatible according to the FEP. Pleroma's Actor in-memory cache will hopefully do the heavy lifting of keeping the DB queries away.

2026-07-29 10:07:45 CEST

- reply

@nprofile…tsag @nprofile…3wz5 Yeah, I've had a couple issues with that fep as well. It's not bad, but could be better.
https://socialhub.activitypub.rocks/t/fep-82f6-actor-statuses/5310/20

2026-07-29 09:46:05 CEST

- reply

@nprofile…tsag @nprofile…3wz5
adding mood to Pleroma
Could work nicely with this which I'm working on: https://codeberg.org/fediverse/fep/src/branch/main/fep/82f6/fep-82f6.md

2026-07-28 22:27:09 CEST

- reply

@nprofile…6wcx They host literal CP on Fedi all the time. And when you email them about it, they go "Hey, we just provide WAF and tunneling for them. Go talk to their hosting provider and domain registrar. Plus we support free speech btw." But they pulled the plug on KiwiFarms, because it was too inconvenient for them.

2026-07-27 15:33:21 CEST

- reply

@nprofile…z753 @nprofile…076k You are already using that terrible blob built on top of a subpar tech. The blob is ActivityPub and subpar tech is HTTP. Anything newer than HTTP/1.1 is overly complicated.

2026-07-27 12:28:47 CEST

- reply

@nprofile…076k Welcome to Saw, this puzzle requires you to leave the room without reinventing the wheel in the corner. It is a perfectly functional wheel. Just go!

2026-07-26 21:10:06 CEST

- reply

@nprofile…cext @nprofile…t3h7 Pleroma-tan is tall and having to scroll for half a page to get some information about Pleroma isn't ideal. And I'm not doing her dirty by making her ~200px tall.

2026-07-23 20:34:49 CEST

- reply

@nprofile…qj99 @nprofile…tsag I've read the whole blogpost hours before even replying here. There is nothing much to talk about as it mostly is fluff I've read and heard tens of times already with unfortunate facts sprinkled in.
The only point worth talking about is right at the end and that is their plans of enforcing said policy. And as I've expected, unless a project is visibly mostly vibecoded, the enforcement will come down to vibes. Which I expected as the policy isn't enforceable without being at least somewhat subjective. That said, I'm personally not a fan of rules/policies that cannot be enforced with near 100% accuracy. But it is their non-profit, their forge hosting, they are free to do whatever the members want.
At least the policy is defined and not completely vague unlike their blanket "crypto related projects" ban they even lie about in the banner that pops up when you are logged in today. If you wanted my opinion on this whole situation, that is it. Nothing more, nothing less. I have no comment on the part you quoted, simply because I mostly do not care about what it talks about.

2026-07-23 14:25:21 CEST

- reply

@nprofile…tsag @nprofile…qj99 @nprofile…mvm4 Especially in the case of "add feature X to project Y that was refused with: No use case/We don't do that." If the project doesn't want it, might as well fork it, get an LLM to do most of the work and be done with it. Not like the changes will get upstreamed anyway, so code quality doesn't matter as much for a personal fork.
This case is where the "making a better world" argument falls apart, I get my feature I want and everyone else gets nothing (likely not even the patches).

2026-07-23 11:01:54 CEST

Why the lie in the banner? It is not "cryptocurrency projects", but "cryptocurrency related projects".
image.png
https://upload.fluffytail.org/media/d0/4f/90/d04f9003b3c08d3a8d7b6b6c329dd21c7a7b40e5534e7603e80844651358f8d1.png?name=image.png

2026-07-23 10:05:12 CEST

- reply

@nprofile…gunk
gnujak-windows-linux-skill-issue.png
https://upload.fluffytail.org/media/3b/6c/e4/3b6ce4c3fbdf7961ea94c67cab7373c0228d020b3725e4fe9f0a52f7a5a387d9.png?name=gnujak-windows-linux-skill-issue.png

2026-07-23 09:34:16 CEST

- reply

@nprofile…tsag @nprofile…muls Kinda glad that whenever I go to any pub or restaurant no matter how fancy it is, the first thing a server gives me is a paper menu.

2026-07-22 21:52:55 CEST

Pleroma propaganda #cofe
d6f6bd6b-8788-4971-a0b6-d1ffde31e4e0.webp
https://upload.fluffytail.org/media/41/69/c1/4169c1435c6c6ae789da8a619bd0a6413fba6c8a38356e5e730dd150420a6e86.webp?name=d6f6bd6b-8788-4971-a0b6-d1ffde31e4e0.webp

2026-07-22 18:24:37 CEST

- reply

@nprofile…9683
I learned today that I can modify something called the PATH variable in my VSCode Terminal window and put my frequently used scripts in the place I added.
But be sure to include the previous contents of PATH to your new one, otherwise your VSCode Terminal will stop working properly.
Here's the zsh configuration:
OLD_PATH="$PATH"
PATH="/User/youruser/frequent_scripts:"$OLD_PATH""
# don't forget this
export PATH
#SysAdmin #DevOps #AlwaysLearning

2026-07-22 16:35:08 CEST

- reply

@nprofile…r2m8 @nprofile…04yf @nprofile…nptm @nprofile…ad74 Makes sense if that is true. Garron desperately wants Mastodon to be a Twitter clone with very similar (bad) UX.

2026-07-22 16:23:42 CEST

- reply

@nprofile…04yf @nprofile…nr0z @nprofile…e40t @nprofile…nptm Spin up a Gitea instance, and disable search engine indexing with robots.txt, it is not that hard. Once scrapers find you, geoblock them with geoip2, or kill off website access without an account. If you aren't hosting a large project, you can afford to be bold with countermeasures.

2026-07-22 16:20:34 CEST

- reply

@nprofile…nptm @nprofile…04yf @nprofile…ad74 There have been few users requesting the same in Pleroma, and I don't get it. Either have it as a reply option, or a dedicated button. A quote and repeat aren't even remotely close in purpose imo.

2026-07-20 21:32:54 CEST

- reply

@nprofile…ehgy @nprofile…72y8 You can theme gtk unless it is libadwaita, the issue is that flatpak does not properly work with system themes. You need to also install the same theme as a flatpak.
Besides that, there is no use case for theming anyway.

2026-07-20 11:58:48 CEST

- reply

@nprofile…em79 Here's a nickel admin, get yourself a real hardware for the real software.

2026-07-20 11:54:50 CEST

>get real software w/ real hardware
>runs Mastodon which eats gigabytes of RAM
Makes sense, you need _real_ powerful hardware for _real_ software

2026-07-18 20:36:24 CEST

>This project argues that trade is the origin of most problems: from climate change to slavery, waste and pollution, corruption, crime, you name it.
>I sure do love sometimes eating exotic food imported from other parts of Earth.
<Do you know that by eating imported food via trade you are causing climate change, corruption and slavery?
>:meowHuh:

2026-07-17 12:29:55 CEST

- reply

@nprofile…3utp @nprofile…t3h7 @nprofile…nptm @nprofile…5vpa @nprofile…tsag
>Not to mention the JSON-in-PostgreSQL being pain in the ass every time something wrong happening to it
That's just ActivityPub being ActivityPub. You can't normalize a barely structured document into columns. Best you can do is pull out certain normalized values into separate columns and have a hybrid schema. Which the indexed jsonb stuff basically does anyway.
I think that is also where the whole rewrite Akkoma DB backend for Cassandra came from.

2026-07-17 12:21:52 CEST

- reply

@nprofile…3utp @nprofile…nptm @nprofile…5vpa @nprofile…tsag It's not as tangled as Pleroma and easier to hack on because of that. Just a simple federation POST request goes through like 10 different modules in Pleroma.

2026-07-17 10:31:47 CEST

Out of all the friends and relatives that said they will delete soycial media and various unencrypted chat apps if chat control passes, guess how many of them actually did. Zero.

2026-07-17 09:20:25 CEST

- reply

@nprofile…tsag It's an amazing filter as well. There seems to be a correlation between the anti-AI cult and software not worth using.

2026-07-17 00:05:26 CEST

- reply

@nprofile…nptm Coincidetally Pleroma also uses application/json which isn't perfectly correct either.

2026-07-16 23:34:02 CEST

@nprofile…nptm Can't find the ActivityPub developer's guide I think you wrote at some point.
But if you wrote it and it has a section on WebFinger, could you add the expected Content-Type headers for /.well-known/webfinger responses? The RFC specifies application/jrd+json, but I would be surprised if developers actually read relevant RFCs these days.
Ran into this issue right now where badgefed sets Content-Type on WebFinger responses to application/activity+json, which makes sense in the AP context, but is completely wrong.
For reference Pleroma supports:
application/jrd+json / application/json
application/xrd+xml / application/xml

2026-07-16 19:14:36 CEST

I like how people are freaking out about Torvalds not being in the anti-AI cult, but being oblivious to the fact that an nVidia engineer has been using LLMs in the kernel for like a year and is responsible for stable releases along with gkh for more than a decade.
This whole ordeal will die in a week and everything will go back to the same state anyway.

2026-07-15 14:35:22 CEST

>How do you upgrade from CentOS 7 to Rocky 8, or Rocky 8 to Rocky 9? The answer is always the same: The project does not support in-place upgrades of one major version to another major version. You need to reinstall to move to the next major version.
Common L

2026-07-13 21:58:13 CEST

- reply

@nprofile…q5n4
>XMPP
I think that is a waste of effort. The moment the classic "Could not decrypt this message" pops up, they are gone. Matrix would probably be better since the clients don't suck as much, but Synapse is kind of a pain in the ass.

2026-07-13 21:55:35 CEST

- reply

@nprofile…q5n4 Or recommend literal honeyposts like Tutanota.

2026-07-13 21:38:00 CEST

- reply

@nprofile…smhc @nprofile…jcp5 @nprofile…hfjf @nprofile…94n6 At least for you the posts loaded. When I look there, half the time they won't even load. Last time I couldn't even find glussy's profile.

2026-07-12 16:00:17 CEST

- reply

@nprofile…pgx0 @nprofile…yj9c @nprofile…8cpz @nprofile…9683 Also by using their RH's "developer" license for personal stuff, you are agreeing to the same EULA like enterprises which gives them the authority to request access to all your RHEL servers for "auditing purposes" :meowWink:

2026-07-12 15:58:32 CEST

- reply

@nprofile…pgx0 @nprofile…yj9c @nprofile…9683 @nprofile…8cpz It is paid-only by default :D Unless you use derivatives like: https://fluffytail.org/objects/40249859-3748-46e7-ba3b-0176b57f1e6b Which kinda exist at RH's mercy.

2026-07-12 15:19:50 CEST

- reply

@nprofile…pgx0 @nprofile…yj9c @nprofile…9683 @nprofile…8cpz At the risk of getting Pete angry, it's redhatware (a downstream of it to be completely correct). Good luck getting a usable desktop from a mostly server only OS btw.

2026-07-12 14:07:22 CEST

- reply

@nprofile…smhc @nprofile…9683 @nprofile…8cpz The comments are classic Loonix discourse as well...
image.png
image.png
image.png
gnujak-windows-linux-skill-issue.png
https://upload.fluffytail.org/media/2b/60/01/2b600199af51eff99bca69ea487d0cc4e2ab24d9346a9da63cb108737fca7586.png?name=image.png
https://upload.fluffytail.org/media/34/03/85/340385f41e776f9f2b036da83926d21299ce21f152b9bebdeaa2cb9ee8f037e0.png?name=image.png
https://upload.fluffytail.org/media/03/d1/4f/03d14ffc7b1719eeb982789a68a5b94d803f4554d18debaab7b0b3ac575c0c22.png?name=image.png
https://upload.fluffytail.org/media/3b/6c/e4/3b6ce4c3fbdf7961ea94c67cab7373c0228d020b3725e4fe9f0a52f7a5a387d9.png?name=gnujak-windows-linux-skill-issue.png

2026-07-12 13:17:08 CEST

- reply

@nprofile…yj9c @nprofile…9683 @nprofile…8cpz Even there you aren't free from the memery of Debian and Canonical.
Everyone hates me for this, but all I need to do to upgrade between releases on my Loonix servers is install two packages, run 2 commands, one of which checks whether my current system configuration is even valid on the next release, maybe update said configuration and reboot. 5 minutes of my time and I have to do it only once every 10 years.

2026-07-12 12:23:17 CEST

- reply

@nprofile…ca3n It's just malware that does nothing it advertises to do properly, inconveniences every visitor by bruteforcing hashes and scrapers don't care anyway as that level of compute is literally free on various cloud platforms.
Besides the fact that it's whole premise is "getting rid of AI scraping", while being written by someone that uses LLMs. Quite ironic.
Besides all that, I'm not waiting a minute and a half to read an email thread or documentation page.
Record_2025-05-30-14-14-00.mp4
https://upload.fluffytail.org/media/45/aa/32/45aa32101cee9d6bf412342166be4ebf3325b971d2bade9d37845678290fb52c.mp4?name=Record_2025-05-30-14-14-00.mp4

2026-07-12 11:43:30 CEST

Even Fedora has this cancer :blobcattableflip:
Screenshot_2026-07-12-11-42-53-39_5dca195d7d09adf043e42a8f3ad62b19.webp
https://upload.fluffytail.org/media/78/d1/0e/78d10ee4f7aeb0e81882905935d7c64a19e82e9d66bce3fdd6148c7b0c140e63.webp?name=Screenshot_2026-07-12-11-42-53-39_5dca195d7d09adf043e42a8f3ad62b19.webp

2026-07-12 11:03:56 CEST

- reply

@nprofile…9683 @nprofile…8cpz Upgrading from 25.10 to 26.04 has Python3 issues still :)
Upgrading anything that is Ubuntu or Debian based between releases might as well be a dice roll.

2026-07-11 22:27:04 CEST

- reply

@nprofile…em79 https://en.wikipedia.org/wiki/Real_life

2026-07-11 22:25:36 CEST

- reply

b91307b0-b7f6-49fa-8519-f4431273f064.webp
860efc62-8e48-4a20-ad37-87bf5df888a9.webp
13ce92d8-2fcc-412b-8959-d0003f04e98d.webp
82cbe3d7-bc6a-4e97-affc-756ffd91e5de.webp
https://upload.fluffytail.org/media/70/c6/65/70c66547b35b7bee83ffa16741542f648fb8d5455d808359b14361295a44f72a.webp?name=b91307b0-b7f6-49fa-8519-f4431273f064.webp
https://upload.fluffytail.org/media/6a/71/35/6a71358737381eb308b16dd0a642738cecc741f16f0866230a922e19f3b76bc2.webp?name=860efc62-8e48-4a20-ad37-87bf5df888a9.webp
https://upload.fluffytail.org/media/9d/43/47/9d4347b4baf96197b2d55fe13763da3778897f8e13d674b59123cfa2a6cdfec1.webp?name=13ce92d8-2fcc-412b-8959-d0003f04e98d.webp
https://upload.fluffytail.org/media/78/d0/17/78d0170bc881f3d5eb5174533e2b51650356178fcf9b78f0baeeb3b8e12c4fe1.webp?name=82cbe3d7-bc6a-4e97-affc-756ffd91e5de.webp

2026-07-10 21:43:39 CEST

c4400a38-f695-4a9a-a0a5-ac106d0225e9.webp
5cafe455-a904-4313-8e20-736204161788.webp
e91b0fd0-ee38-4c27-aa8e-2c9d5cbb52c9.webp
832e7b7d-c9be-4827-abf6-82a7c62408b1.webp
8b5dea39-2440-4050-8a6b-71079632db28.webp
0ae27482-e26b-475b-93e1-aa68325069aa.webp
https://upload.fluffytail.org/media/a4/d1/19/a4d11941a4252dab77ead3a0191156c9bd088cb08d73ea2aaf4060e3b781fd73.webp?name=c4400a38-f695-4a9a-a0a5-ac106d0225e9.webp
https://upload.fluffytail.org/media/39/8a/d9/398ad9841dcd4b4bd1bfa7c58ecec57d2f47bed7bb4dc932d2ee1dce384bd309.webp?name=5cafe455-a904-4313-8e20-736204161788.webp
https://upload.fluffytail.org/media/37/8b/8d/378b8d5bb5d65ea5ffd14c9d6b58257976c4bd1b7acf59f4d401759ea1c2afb8.webp?name=e91b0fd0-ee38-4c27-aa8e-2c9d5cbb52c9.webp
https://upload.fluffytail.org/media/92/e0/25/92e0257cfc3236b7393fe0ba63c3ab8dd9d5fee6b4ecfa3a04222828078ba494.webp?name=832e7b7d-c9be-4827-abf6-82a7c62408b1.webp
https://upload.fluffytail.org/media/2f/ee/72/2fee72248a9e9c0580233df0cbe500cef339bc5d6945bdb717da7b5243f8ce11.webp?name=8b5dea39-2440-4050-8a6b-71079632db28.webp
https://upload.fluffytail.org/media/81/2e/9c/812e9c42e1e6bacf2a720527de95b63559376accb177a0d050f2947ed320ab7d.webp?name=0ae27482-e26b-475b-93e1-aa68325069aa.webp

2026-07-09 23:26:31 CEST

- reply

@nprofile…6xru @nprofile…fq95 That probably comes next (mandatory ID verification)

2026-07-09 16:46:44 CEST

- reply

@nprofile…thtd open sores and free copeware. One consistently kills itself with drama, the other one uses a license as a cope for feeling superior.

2026-07-09 16:37:12 CEST

- reply

@nprofile…thtd Yeah, both lunny and silverwind use Claude. The maintainer with the snowflake avatar isn't super happy about it since some of the reviews he does have nonideal (broken) design, but that's about it. There was tension between lunny and the snowflake for years already, nothing new.
The whole Gitea codebase is full of nonideal design and Forgejo just made it worse. Gusted which is now pushing this Codeberg policy was a Gitea maintainer for like 2 months with <300 commits and then suddenly went to Codeberg when the drama started (he's now in the Codeberg presidium).

2026-07-09 16:31:20 CEST

- reply

@nprofile…thtd That's just Gitea now :D

2026-07-08 22:58:48 CEST

- reply

@nprofile…d73q I wouldn't be so optimistic. If they can introduce unenforceable policies to make a statements that one is right up their alley. They even have the same core argument (killing the planet).

2026-07-08 22:03:46 CEST

- reply

@nprofile…cext Or Linux, they won't be able to do anything with it anyway since nothing works.

2026-07-06 14:36:59 CEST

>Sushi has been ported to GTK4.
Even Japanese cuisine isn't safe from gnomemification

2026-07-04 23:03:47 CEST

>heat pump
>look inside
>reversible fridge
#nevent1q…ty4c

2026-07-04 22:25:12 CEST

- reply

@nprofile…hu5l
>Let's end fast tech
Unfortunate phrasing.

2026-07-04 13:40:55 CEST

- reply

@nprofile…sw8j The FS is unmountable with a bad main superblock at least. Only way to recover is to use a reserve superblock and hope the trees are fine.

2026-07-04 13:15:52 CEST

Broke btrfs by testing hibernation award.

2026-07-02 14:32:27 CEST

- reply

@nprofile…cext It's also a reflection of humanity, because social media in general is a total dumpster fire at least for the last 3-5 years.

2026-07-02 14:27:58 CEST

- reply

@nprofile…cext In the grand scheme of things, it doesn't matter at all how either of those are ran. What matters is the content and Fedi is lacking in that regard in many ways. Creatives and artists are gone, shitposters are half gone. What we have though is people shitting on everyone's timelines with <insert dumb politics/AI discourse>, and GNU/Retards being retarded as they've always been.
The quality ratio has gone down on Fedi noticeably, while on Twatter it seemingly went up and that's what matters.
pleroma-social-media-arguing.png
#nevent1q…tlf4
https://upload.fluffytail.org/media/db/02/72/db027261df9409cd685c9ec936401d4127eb3bd48f930eb74db6d4d835039105.png?name=pleroma-social-media-arguing.png

2026-07-02 12:15:35 CEST

- reply

@nprofile…ehvw @nprofile…tsag VSCode and friends or Sublime.

2026-07-02 12:12:16 CEST

- reply

@nprofile…tsag Didn't expect you to be a JetBrains user :D

2026-07-02 11:57:03 CEST

JetBrains apparently freed most of their IDEs for non-commercial use. Kinda unexpected, I guess the 500 different AI IDEs are cutting into the revenue. Only GoLand and PhpStorm are paid only now.

2026-06-29 22:14:45 CEST

- reply

@nprofile…t42d @nprofile…9683 Memory-holing Pleroma is normal for Mastadon for years. Even Funkwhale with its 5 users is there.

2026-06-29 22:03:30 CEST

Remember to fill this absolutely not answer-biased questionnaire out that definitely is not made to get specific results. (Answer truthfully.)
cc @nprofile…9683
dumb-one.png
missing-deletion.png
missing-no.png
#nevent1q…fpnr
https://upload.fluffytail.org/media/a3/68/b6/a368b6874dbddd8758f6d3619ef961c79169ec6b230b35cf4b6db9d09f892579.png?name=dumb-one.png
https://upload.fluffytail.org/media/ce/d9/e3/ced9e3c4b35123b7f88614da35734f1c399b6a645ac09dd87c456401f4f542ef.png?name=missing-deletion.png
https://upload.fluffytail.org/media/21/71/8a/21718a76036c59db5c905bba2b7951e53d3763a445912dc2030d11c1a251f760.png?name=missing-no.png

2026-06-29 19:06:30 CEST

- reply

@nprofile…thtd A path of smaller margins for them.

2026-06-28 14:51:35 CEST

I wish all political mullvad refugees a happy hunting for a VPN that won't log your activity and fuck you over.

2026-06-27 21:03:06 CEST

- reply

@nprofile…qyfx Tor

2026-06-24 09:15:44 CEST

Reddit mod speak
HLbIbh0agAAVnRB.webp
https://upload.fluffytail.org/media/8f/f5/06/8ff50670f0764e36668c613b4fbe925b101d3ce8a407b008d1fde29100ad9ae1.webp?name=HLbIbh0agAAVnRB.webp

2026-06-23 21:34:08 CEST

- reply

@nprofile…5n8n @nprofile…8m52 @nprofile…zr8z @nprofile…9683 @nprofile…tsag Here's a thought, how do you think people would react if the regressions were introduced completely with any AI use at all.

2026-06-23 21:31:04 CEST

- reply

@nprofile…5n8n @nprofile…9683 @nprofile…8m52 @nprofile…tsag @nprofile…zr8z I wouldn't call rsync slop yet. Neither 3.4.3 or 3.4.4 failed me yet and I use rsync extensively.
>Why does being a reasonable person have to be formalized?
Doesn't has to be. His actions to me so far are reasonable. If he decides to stop development of rsync completely, I might be annoyed for some time, but I can't really blame him. What is not reasonable to me on the other hand is the Mastodon™ users having a meltdown over Claude in commit attributions universally all day every day, and seemingly running newest versions of software in production without any testing, as I've seen during the time where the rsync regressions cropped up. Reasonable people would make an issue on Github about it and maybe write a post about the regression along with pasting the link to the issue, but that is not what happened in 99% of the cases.

2026-06-23 21:16:34 CEST

- reply

@nprofile…5n8n @nprofile…9683 @nprofile…8m52 @nprofile…tsag @nprofile…zr8z
>I tried my best bro". Rsync is a mission critical tool for many, many people and businesses (although that whole "the entire world depending on one dude doing it for free" image comes to mind) and that level of carelessness is core to my argument that effort is still required for AI tools.
On the other hand, he owes you absolutely nothing. He wants to do other things than maintaining rsync with basically zero outside help from anyone for years, so you are left with two choices. Get an LLM involved to get at least some level of maintenance and security fixes going, or remain in previous status quo and do almost nothing. These are your only two options you have. Your third choice is to go somewhere else.

2026-06-23 00:05:07 CEST

- reply

@nprofile…9683 @nprofile…8m52 @nprofile…tsag @nprofile…zr8z
>"All these jithub dipshits are putting their hallucinating clankers into my pull requests"
Yes, that is a thing I tend to be somewhat annoyed with. HackerOne sec reports are probably way worse than that, because there's profit to be made there. Daniel Stenberg sadly didn't provide statistics besides "less than 5% were legit" on this when shutting their one out though.
I don't mind reviewing clanker code, but because it tends to be submitted in large chunks and also in large commits, it can make be a rather daunting task which takes longer than usual. One thing I've found after being exposed to clanker reviews though is that I became much more cognizant of the changes being made and am more skeptic/suspicious about them in general, which also translates to the same with non-clanker code. Which I think is a good thing, because it exposed I wasn't as thorough with non-clanker code. Although it also resulted in more nitpicks for ways I would prefer things to be done instead (sorry lain).
> (On the other hand, like, feel free to do opinionated.)
I don't like being opinionated in discussions. Being opinionated to me means being partial to one side, not being objective and replacing facts with emotions and opinions. Opinionated software can be good, but I don't think opinionated takes are beneficial to discussions in most cases. They tend to degrade a discussion into something completely different, usually personal attacks. That's why a good chunk of my effortposts are somewhere around the middle on the fictional axis of opinions, or try to get there.

2026-06-22 23:02:56 CEST

- reply

@nprofile…9683 @nprofile…8m52 @nprofile…tsag @nprofile…zr8z I get where Moon is coming from. If it weren't for you and lanodan in this thread, I wouldn't have replied either, because it had one of the typical setups for the typical "all AI bad, you are an idiot if you think otherwise" threads that me and probably also Moon are tired of seeing on TLs every few days (picrel).
Both of you are the sole reason why I bothered writing my more or less unopinionated and emotion-less reply for the quest of having some discussion in the first place.
ai-llm-arguments.png
https://upload.fluffytail.org/media/28/bb/d4/28bbd428547e32299b623a5bdfcfd840532bc87aba4d6f72d7a6e8bffe75fa40.png?name=ai-llm-arguments.png

2026-06-22 08:56:23 CEST

- reply

@nprofile…8m52 @nprofile…9683 @nprofile…zr8z
>>And 9front's even more elegant:
>3. Submitted code is not computer-generated
Technically that bans also completely mundane stuff like auto-generated headers from things like device trees/spec sheet documentation or m4 macro expansions. Of course it will never be enforced that way.
>and at very least 90% of code is pure crap nobody wants to read (and so nobody reads).
Things got way better in the last year or so with new-ish models. If you take the time to set up a workflow for the new models, the results are still somewhat easy to distinguish from non-clanker contributions, but they aren't bad either and usually require some smaller fixups and removal of some overly defensive checks. If done by someone that knows stuff at least. And even with the "chat in a browser window" way, you can get decent-ish results if you keep things small.
>Plus outside of purely code, since then there's also the absolute spam of completely bogus security reports
On the other hand you also have security reports with really high severity that were found within a day or two of letting a model loose on a codebase. The recent nginx CVEs were all found by LLMs, most of the recent Linux LPEs as well. The way I see it is the same as static analyzers and fuzzers, a useful tool in the hands of someone knowing how to handle it.

2026-06-14 14:04:49 CEST

- reply

@nprofile…nptm
>And one thing that caught my attention is the use of Accept and Reject activities for managing approvals.
This is an ActivityPub issue more than anything else. The "spec" is clear that the side-effects are to be determined by the type of the Object. There is no other indication, you effectively must check the type of the Object referenced. And I think that is correct, the same way Mastodon is correct in its usage of these two Activity types.
If you split the type for other Accept/Reject Activities, like AcceptFeatured or whatever, you are still left with more type determining logic anyway [at least in function non-pattern matching languages :)]. You just moved it somewhere else instead.

2026-06-13 12:34:32 CEST

>gnuware
>Can I also ask you to step down from CoC for violating the CoC here?

2026-06-06 00:00:48 CEST

- reply

@nprofile…tsag People overextended on speculative value. Although these crashes are like a yearly thing at this point.
crypto-investors.png
https://upload.fluffytail.org/media/38/1e/34/381e340a54f9cd8e06b74ab0c85f22ec3d241ef3e51d86c5fc44a64b85b527f0.png?name=crypto-investors.png

2026-06-03 20:42:42 CEST

Funny how suckless software is universally shat on by neverclankers, DeVault orbiters and people generally pushing the linux phones. But somehow if you slap a different name on the same software, add some patches and host it on Sourcehut, it's suddenly kosher.

2026-06-02 17:43:04 CEST

- reply

@nprofile…mrks human eyes can't see more than 30 fps anyway :meowDerpy:

2026-05-29 22:15:50 CEST

- reply

@nprofile…8m52 @nprofile…csn2 @nprofile…euty Reviewing every diff isn't doable I think, unless you actually bring llms into the equation which could go through those. But you also have to manage that as well, not including the various questions about llms which I won't go through. I mean it more in terms of automated testing and checking that upstream didn't do something completely stupid. Netdata suid binary path traversal comes to mind for example, which should have been caught by any distro trying to be serious and packaging it.
On the matter of size, software is getting larger and won't stop. IMO sizes like Linux, LLVM, Firefox, Chrome should be an exception and are a problem in an of itself. Those codebases are all unauditable.
Also on the topic of OP, borgbackup is nice although written in python/cython. Definitely don't use tar ;)

2026-05-29 21:54:04 CEST

- reply

@nprofile…csn2 @nprofile…8m52 @nprofile…euty That ain't happening simply because of scale. Instead distros should start doing what they should have been doing for years, vetting upstream and not taking upstream completely at face value.

2026-05-28 11:44:04 CEST

- reply

@nprofile…wqgn I'm seriously considering developing in a VM, because at this point you can't be certain that someone won't loose their mind over what you are doing with their project or even what you have installed on you system. I don't think that projects being malicious because the maintainer doesn't like a piece of software installed on your system won't become reality in recent times if this is already normalized. And it sort of already is normalized when you look through AGENTS.md/CLAUDE.md of projects maintained by highly opinionated maintainers.

2026-05-28 10:19:23 CEST

You using AI agents is not ethical, but me deleting all your tests because you used an AI agent is 100% ethical.
Dissonance so large that you need scientific notation to describe it.

2026-05-26 19:38:53 CEST

- reply

@nprofile…g80s Tree cum season is coming up for me as well. Usually hits in mid-june.

2026-05-23 15:48:50 CEST

"Our software acts like malware, lmao." - Riot Games
https://xcancel.com/riotgames/status/2057604027941302564
image.png
https://upload.fluffytail.org/media/a7/43/23/a743237ff544a7cbf9ffea1dd5e661395effa984ae840f4f5dce5a87be8d9aaf.png?name=image.png

2026-05-22 17:43:28 CEST

fedi-stalin-gleason-gone.webp
https://upload.fluffytail.org/media/24/aa/09/24aa09286e4de8d623b227212c85ff35f8f71fe8bfd2024eb6bd910f8641826a.webp?name=fedi-stalin-gleason-gone.webp

2026-05-21 21:28:14 CEST

- reply

@nprofile…t3h7 I have the privilege of not being addicted to Youtube, watching a video a day at most and not using any other Google product. I have a Fediverse addiction instead...

2026-05-21 16:37:57 CEST

- reply

I don't see the appeal for the most part. Same with streamers in general. Like what's the appeal of watching someone when there's a 0.5% chance they will actually read your message unless you send donations. It's like watching TV, boring.
The only vtubers I watched were all small indie ones (vidrelated).
arty-kill-me.mp4
https://upload.fluffytail.org/media/01/1a/0c/011a0cd86b2717596dfe609925553c29f9bf337071fb371627bcf3a28cc3b64c.mp4?name=arty-kill-me.mp4

2026-05-20 22:08:50 CEST

#cofe
coffee-life.webp
https://upload.fluffytail.org/media/d5/21/c0/d521c0705265b78e4c31f002087db627d3f05f0239619c798de64a4bde313874.webp?name=coffee-life.webp

2026-05-13 16:21:47 CEST

- reply

@nprofile…y6jc idk, every time I try too look into the codebase, it's incomprehensible to me. The components are one of the few things I understand in it. The moment I look at the JS outside of components, I have a hard time understanding it. Likely from my JS skill issue that I refuse to address and not liking dealing with frontends.
I don't mean it in a bad way, you probably have something similar when looking at the backend.

2026-05-13 16:07:10 CEST

- reply

@nprofile…y6jc I think the FE is already too complicated, no need to make it worse.

2026-05-13 16:03:28 CEST

- reply

@nprofile…y6jc
1. Don't do it
2. Misskey has it for writing plugins and new widgets for the UI. The Misskey games are written in it I think.

2026-05-13 15:55:30 CEST

- reply

@nprofile…y6jc These are the only useful ones anyway: https://misskey-hub.net/en/docs/for-users/features/mfm/
Also PleromaScript when? https://misskey-hub.net/en/docs/for-developers/aiscript/

2026-05-12 09:59:06 CEST

"GET /phnt/pleroma/src/tag/v2.10.0/priv/repo/migrations/20190525071417_add_non_follows_and_non_followers_fields_to_notification_settings.exs?display=rendered HTTP/2.0" 200 33326 "-" "Mozilla/5.0 (Linux; Android 5.0) AppleWebKit/537.36 (KHTML, like Gecko) Mobile Safari/537.36 (compatible; Bytespider; [email protected])"
And TikTok is interested in Pleroma migrations.

2026-05-11 23:19:13 CEST

- reply

@nprofile…jup9 @nprofile…l5m0 It has other problems.
linux-ragebait-shirt-systemd-wayland-flatpak.jpg
https://upload.fluffytail.org/media/41/ff/4f/41ff4f9b66c8e0451e58aa9c69a4b9dc86c52940f21b8e1a1760856392b44902.jpg?name=linux-ragebait-shirt-systemd-wayland-flatpak.jpg

2026-05-10 17:45:44 CEST

>Software being open sores means you can use it indefinitely. If the binaries no longer work, you can just rebuild it.
Alright so in order to build Erlang 22 from 2 years ago (OTP 22 is 7 years old), you need:
period-correct-ish version of GCC from ~7 years ago, because GCC 15 is too pedantic about syntax (oldest GCC Gentoo didn't mask yet is 11 from from 5 years ago)
custom build of OpenSSL 1.1, because Gentoo's openssl-compat package doesn't cut it (doesn't provide headers)Now think what it takes to build software from 20 years ago.
image.png
https://upload.fluffytail.org/media/f0/9e/4c/f09e4ca64a710ef667993b34d3f17a1c3cce068bb536925c85d6e7025264160e.png?name=image.png

2026-05-05 09:34:11 CEST

- reply

@nprofile…y6jc @nprofile…thtd It's deserved. The language sucks and browsers suck even more.

2026-05-05 09:01:06 CEST

- reply

@nprofile…y6jc @nprofile…thtd
js_devs.jpg
javascript-for-kids.png
https://upload.fluffytail.org/media/5c/03/12/5c0312072c059c7a8a1ec6cf5673feb9f18e3ca9f44053a8c398e508f8c86a04.jpg?name=js_devs.jpg
https://upload.fluffytail.org/media/bd/82/fd/bd82fde04a0f6c304d613b98f8d64850640fe1749007362926e19db0536d9387.png?name=javascript-for-kids.png

2026-05-05 08:53:34 CEST

- reply

@nprofile…thtd
zig-rust-c-languages-carmack.webp
https://upload.fluffytail.org/media/65/ec/65/65ec65497d339714797beea904769aaf27c9150b8c87a1ad931c866fdd41c192.webp?name=zig-rust-c-languages-carmack.webp

2026-05-02 15:41:55 CEST

- reply

@nprofile…4kgw Fully trustworthy vendor and non-vulnerable software does not exist. The world you are thinking of isn't real, for many things are you stuck with some vendor and or some other vendor and that's it. You either use that or you can build your own, which usually results in something even worse.
Or are you running an OS along with the whole userspace on our computer that has been mathematically proven do that what it says it does without error?

2026-05-02 15:26:38 CEST

- reply

@nprofile…4kgw
>Let them deal with the consequences of pushing shit code into prod, this isn't your problem.
As I've said on IRC, you only say this because you don't have to deal with the chaos and panic in the aftermath of disclosing high severity issues before a fix is released. You never had to sit down and try to figure out which of your possibly hundreds or thousands of systems is affected and how to fix it.