From: Sam James <sam@gentoo.org>
To: Morten Linderud <foxboron@archlinux.org>
Cc: "Michał Górny" <mgorny@gentoo.org>, distributions@lists.linux.dev
Subject: Re: Looking for advice on how to deal with potential slop packages
Date: Wed, 11 Mar 2026 02:48:58 +0000 [thread overview]
Message-ID: <87ms0fw06d.fsf@gentoo.org> (raw)
In-Reply-To: <aawYbre5_xvhfwKA@framework>
[-- Attachment #1: Type: text/plain, Size: 2324 bytes --]
Morten Linderud <foxboron@archlinux.org> writes:
> A lot of this is probably already a lost cause I think.
>
> Linux accept LLM contributions (look for the `Assisted-by` tags), and there are
> already multiple subsystems that are being developed in-part, or full, by LLM
> agents.
There is a distinction between the chardet case [0] or otherwise
massively-AI driven development introducing instability and bugs, and this.
>
> See the LWN discussion: https://lwn.net/Articles/1026558/
>
> `b4` has also incorperated llm agents for the review workflow.
> https://b4.docs.kernel.org/en/latest/maintainer/review.html#configuration
>
> b4 is also heavily developed by Claude these days.
>
> I don't think we can reasonably argue that Linux is not free software, and I
> don't think we can argue for forking Linux to remove llm generated code.
>
> My take on this is mostly apathy. I don't think we can reasonably challenge the
> use in the FOSS community. The productivity boost of experienced developers
> using these is too appealing when we are looking at overburdened FOSS
> maintainers.
It is of course your right to feel that way and I don't entirely
disagree, but I also think we have some responsibility to our users to
protect them in the way we always have.
I don't think that Linux having some AI-assisted commits means that we
can't discuss or perhaps even have some consensus on handling some packages.
Say, by trying to avoid new chardet and lobbying reverse dependencies to
not require newer versions if they do so, or to port to
charset-normalizer.
I don't think we'd have this response if some software was being made
proprietary. It's the same thing in that it requires some effort to
rebut, just like also if sometihng becomes unmaintained and has a
suitable alternative, or whatever else. We do that all the time.
> We've aleady been repeadetly DDoSed by these companies. Spending hundreds of
> volunteers hours keeping our services running while the companies extract the
> labour to sell back to the FOSS community, using their standing in the Linux
> Foundation to further cement their usage in our communities.
>
> Then the FOSS communities use these models without any care of the ethical considerations.
>
> Is this depressing? Yes.
[0] https://lwn.net/SubscriberLink/1061534/3040c9a3a6271043/
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 418 bytes --]
next prev parent reply other threads:[~2026-03-11 2:49 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-07 10:38 Looking for advice on how to deal with potential slop packages Michał Górny
2026-03-07 12:07 ` Noé Lopez
2026-03-07 12:36 ` Morten Linderud
2026-03-07 15:31 ` Simon Josefsson
2026-03-08 4:00 ` Guillem Jover
2026-03-22 23:53 ` Andreas K. Huettel
2026-03-23 8:14 ` Simon Josefsson
2026-03-11 2:48 ` Sam James [this message]
2026-03-11 2:50 ` Sam James
2026-03-27 8:01 ` Bernhard M. Wiedemann
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87ms0fw06d.fsf@gentoo.org \
--to=sam@gentoo.org \
--cc=distributions@lists.linux.dev \
--cc=foxboron@archlinux.org \
--cc=mgorny@gentoo.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.