@civodul@toot.aquilenet.fr

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4w ago

#Guile now requires contributors to agree to a Developer Certificate of Origin (DCO) and rejects code authored or co-authored by an LLM 👇 https://codeberg.org/guile/guile/src/commit/b52c48a850765ba93d90fa35fbd3eed8356222a2/HACKING#L134-L156 Kudos to Olivier Dion for making it happen!

codeberg.org

Cookie monster!

235

0

130

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 1w ago

Like any good admin, I chose Friday evening to deploy Cuirass 1.3.6 on https://ci.guix.gnu.org and https://pulls.ci.guix.gnu.org 👇 https://codeberg.org/guix/cuirass/src/commit/9c586674283951951580c315512b89d2643e1a08/NEWS#L14-L38 It’s primarily a bug-fix release but it has a couple of noteworthy improvements made by my partners in crime at Inria, @shepherd@fosstodon.org and @vivien_kraus@h4.io . #Guix

ci.guix.gnu.org

Cuirass — Your friendly Guix continuous integration service.

18

2

13

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 1mo ago

Replying to

Under the guise of neutrality, what it actually does is open the gates to genAI-produced contributions. This, is political. The “responsible use” narrative shifts responsibility from the group to the individual. It’s not unlike the tobacco industry’s strategy—or indeed that of the “AI” industry.

111

1

49

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 1mo ago

Replying to

A funny example: the text recognizes that “the legal status of material produced by generative AI systems remains the subject of ongoing discussion in many jurisdictions” (a euphemism) and then states that Debian will “rely on the judgment and responsibility of its individual contributors”—as if individual contributors would know better than lawyers in this domain.

84

1

16

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 1mo ago

Replying to

The message of this resolution is that it’s neither “pro” nor “anti”: it’s neutral, promoting “responsible use”. In fact, it looks like an example of “extreme centrism”: “Debian neither endorses nor prohibits the use of generative AI”, “Contributors are expected to exercise appropriate care”, “[t]his resolution therefore affirms that generative AI is neither exempt from nor subject to special rules”. The narrative seems to be: “AI” is just another tool, nothing changes here—move along.

53

1

13

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 2w ago

I’ve been working on this cute little 🐑 #Shepherd feature: settings. https://codeberg.org/shepherd/shepherd/pulls/153#issuecomment-23343229 A service would have settings, which one can inspect, change, and reset to their default value. This would allow one to fiddle with services at run time lightheartedly.

Codeberg.org

WIP: Implement service settings

This patch set implements *service settings*.

7

0

7

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Well-documented list of free software projects and their use of genAI:
https://codeberg.org/small-hack/open-slopware

It’s already a long list that shows what looks like uncritical adoption, both by high-profile projects (systemd, VLC, etc.) and by niche projects (GNU Mach is a prime example).

codeberg.org

78

24

86

1

Ludovic Courtès @civodul@toot.aquilenet.fr

· 2mo ago

“Protecting our FLOSS commons from LLMs” https://blog.codeberg.org/protecting-our-floss-commons-from-llms.html Good to see #Codeberg take a stance and remind us all of the risks LLMs pose to what we’ve been patiently building.

blog.codeberg.org

Protecting our FLOSS commons from LLMs — Codeberg News

In Brief: Two motions regarding

16

0

9

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

“[LLMs] could demystify the capabilities of a modern Linux workstation and bring them to a much wider audience.”
https://discourse.ubuntu.com/t/the-future-of-ai-in-ubuntu/81130

Demystifying by adding an “everything machine” as a black-box on top of a complex system? I’m skeptical.

Ubuntu Discourse

The future of AI in Ubuntu

As 2026 progresses, LLM-based tools are becoming more and more ubiquitous. Adoption across the tech industry has been mixed, both in terms of which projects are embracing “AI” technologies, and in how companies are structuring their adoption. As a result, I’m frequently asked about what Canonical and Ubuntu will do (or not) to incorporate AI. In this post I’ll detail how AI will play a part in both Canonical and Ubuntu’s future, my framework for classifying AI features in the OS, and how Canoni

40

12

10

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 3mo ago

Version 3 of GCD 008 on genAI use in #Guix is available 👇 https://codeberg.org/guix/guix-consensus-documents/src/commit/a0aca85d5079d740f953794f63722d76611aece9/008-genai.md Discussion 👇 https://codeberg.org/guix/guix-consensus-documents/pulls/13#issuecomment-18619856 📅 Deliberation tentatively scheduled for July 11th.

codeberg.org

10

0

10

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

The mechanism that makes it so that we no longer need to enter our LUKS passphrase twice is pretty cool:
https://guix.gnu.org/manual/devel/en/html_node/Bootloader-Configuration.html#Automatic-LUKS-Master-Key-Passing

#Guix

guix.gnu.org

20

6

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@abucci I’m also of the opinion that, even from a purely free software perspective, putting aside the many ethical concerns (environmental and social), integrating those opaque services as part of the development workflow is a problem. I mean, “we” fought against the use of BitKeeper, and later GitHub, for free software development—among the many things people were critical about. And now all this would go through just fine? @Profpatsch

14

1

5

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Sandboxing & downloading channel files:
https://codeberg.org/guix/guix/pulls/6745

The way to ‘nix run’ style commands but without the ‘curl | sh’ flavor?

#Guix

codeberg.org

13

0

5

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Just watched this nice #LixCon2026 talk on Zilch:
https://app.media.ccc.de/v/lixcon-2026-14-using-nix-without-the-nix-a-retrospective-on-zilch

It’s a Scheme interface to talk to the Nix daemon, it has a thing called “z-expressions” to capture context in staged code, client-side build hooks… sounds familiar, we should talk! :-)

toot.aquilenet.fr

12

2

6

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

#Guix at the annual French #ReproducibleResearch Network conference, May 19–21 👇
https://hpc.guix.info/blog/2026/05/guix-at-the-french-reproducible-research-network-conference/

toot.aquilenet.fr

7

0

9

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

“It may constitute a threat to commit genocide” says Amnesty:
https://www.amnesty.org/en/latest/news/2026/04/iran-president-trumps-apocalyptic-threats-of-large-scale-civilian-devastation-demand-urgent-global-action-to-prevent-atrocity-crimes/

Drop “may”? It’s unambiguous, and so was Gallant’s speech on Oct. 9th, 2023.

amnesty.org

9

0

4

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4mo ago

I’m chewing on this: Lisp is touted as a “programmable programming language”, removing restrictions to what programmers can express.

What does it mean in a world where “programmers don’t write a line of code”, as sold by AI vendors?

More generally, what’s the point of designing a programming language in that world?

#nihilism

toot.aquilenet.fr

4

13

3

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4mo ago

Replying to

@swheritage@mstdn.social: you might like how we put SWHIDs and the archive to good use. 👆

4

0

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

📢 Hear, hear! #Guile birds of a feather session online on Monday, April 6th, 1PM UTC!
https://lists.gnu.org/archive/html/guile-devel/2026-04/msg00002.html

Kudos to Olivier Dion for this great initiative—and for all the energy put into Guile!

toot.aquilenet.fr

6

2

5

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

This latest report by Albanese, obviously well documented, sheds light on aspects we don’t hear much about and puts them in perspective:
https://www.un.org/unispal/document/torture-and-genocide-report-francesca-albanese-a-hrc-61-71/

Well worth a look.

un.org

5

0

4

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Until ‘timerfd’ became a thing, the only way to wait for a real-time clock calendar event (the main use case being cron) was to poll the real-time clock periodically, to account for the fact that it can change at any time:
https://codeberg.org/guile/fibers/pulls/177#issuecomment-13351431

🤦

codeberg.org

4

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

Je vois cette pub EDF dans la rue, blue-blanc-rouge où le blanc est celui du panache de fumée d’une centrale nucléaire, avec le slogan « produire l’électricité dont la France a besoin », et je me dis que tout y est : le nucléaire décomplexé loin de la discrétion suite à Fukushima, la référence nationaliste et l’allusion guerrière.

Quelle belle époque vivons nous.

4

0

3

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

Le ☕ Café #Guix de demain sera animé par @khinsen@scholar.social et comparera #Guix et #CONDA pour des projets scientifiques.

📅 mardi 31 mars, 13h–14h
▶ https://hpc.guix.info/events/2025-2026/caf%C3%A9-guix/

#RechercheReproductible #ScienceOuverte

toot.aquilenet.fr

4

0

4

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

Replying to

@flomaraninchi C’est ça qu’est triste : de plus juste pouvoir ignorer le bouzin et continuer sa vie comme avant. Tout ça parce que le bouzin est en train de bouffer tout ce qui se trouve autour de nous et que trop peu ont su/ont pu le tenir à distance.

3

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@pjotrprins@mastodon.social Well, note that it does so at the user’s request. :-)

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@samuel Same for me: I was expecting some extra work in my config and then realized the thing was already working as advertised. :-)

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@futurile Nice and useful series of posts! Related links: https://guix.gnu.org/manual/devel/en/html_node/Build-Utilities.html#index-substitute_002a https://guix.gnu.org/manual/devel/en/html_node/Build-Utilities.html#index-search_002dinput_002dfile

guix.gnu.org

Build Utilities (GNU Guix Reference Manual)

2

0

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

Replying to

C’est vraiment malin cette idée. Ça rendrait beaucoup plus périlleux l’entraînement sur des données trouvées ça et là ; ça permettrait plus facilement de défendre les licenses copyleft aussi.

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4mo ago

Replying to

@khinsen@scholar.social Yeah, this is one of the option that was discussed. Regarding ‘channel-with-substitutes-available’, WDYT of the alternative option? https://guix.gnu.org/manual/devel/en/html_node/Channels-with-Substitutes.html @zimoun@social.sciences.re

guix.gnu.org

Channels with Substitutes (GNU Guix Reference Manual)

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4mo ago

Replying to

@khinsen@scholar.social @zimoun@social.sciences.re Oh but the question is whether manifest files should be isolated as well—I should read more carefully. :-) I’d lean towards “no”, because like you wrote, manifests can be and often are more advanced than just lists of package names. Conversely, channel files are almost always trivial, with ‘channel-with-substitutes-available’ being perhaps the only attempt to program channel files (but it is little used AFAIK).

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@janneke@todon.nl Nice! What characteristics are you referring to? @wingo@mastodon.social @ekaitz_zarraga@mastodon.social

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@Profpatsch Yeah well, it’s a questionable categorization; I guess their goal is to distinguish between those forbid/allow/boast-about use of LLMs. I dislike the pointing-fingers aspect of it, but I find the links to policies etc. quite valuable.

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@emaste I guess they consider “permissive” anything that doesn’t explicitly forbid genAI-assisted contributions. I don’t see a commit link for FreeBSD, but maybe that’s because you reported it before?

1

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@raito Compromised forges are not a problem: one needs access to a committer’s key to get code in Guix. Coercion? The only way I can think of to mitigate this is canaries. But this is hard to counter by definition, and I’d say beyond our threat model (for now). @zimoun @luj

1

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@raito Supply chain security has been addressed from different angles in Guix, which I tried to explain in https://doi.org/10.22152/programming-journal.org/2023/7/1 Probably not the end of the story (we’re still tackling sometimes ridiculous issues, like avoiding “source” tarballs that contain pre-built artifacts), but a good start. @luj @zimoun

doi.org

1

6

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@zimoun Build reproducibility is monitored per package, per commit, and per architecture, and all that is stored in the Data Service. We can get a glimpse of it at https://qa.guix.gnu.org/branch/master The downside is that there’s currently “only” two build farms to compare. @luj

qa.guix.gnu.org

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Ma baguette a pris 5 centimes depuis hier, « à cause du détroit d’Ormuz ».

Saleté de détroit.

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

Replying to

Seul bémol : la proposition ne parle que de « contenus culturels ». Mais bon, le code, c’est quand même largement du contenu culturel aussi, c’est pas ici qu’on va dire le contraire. :-)

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

Replying to

@wingo@mastodon.social Really? I have a feeling it’d be considered mild in today’s USA.

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4mo ago

Replying to

@zimoun@social.sciences.re Only channel files are isolated, not manifests. @khinsen@scholar.social

0

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@Eeveecraft Yeah, I sympathize with the “not judging” part in pyg’s message, especially given Framasoft’s popular education mission. I still trust them to make informed decisions but I admit the rest of the argument is a little less clear to me. Including the signature acknowledging use of DeepL to write this message.

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4mo ago

Replying to

@jfred@jawns.club It would get a lot more complicated and I’m not sure it’s worth the effort: if you’re going to install packages from a channel, then those packages matter too. Put differently, if you trust the packages provided by a channel, you might just as well trust its package definitions and modules.

0

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4mo ago

Replying to

@jfred@jawns.club True. And that too is missing: running programs in a least-authority environment (you can kinda do that with ‘guix shell -C’ but finding the right set of options is tedious).

0

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 6mo ago

Où l’on parle d’un renard, de cybernétique, de cerveaux électroniques et d’intelligence artificielle (1961).
https://www.ina.fr/ina-eclaire-actu/video/caf88038611/albert-ducrocq-la-cybernetique-1ere-partie

ina.fr

0

2

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@domo @hipsterelectron This is creepy: https://lists.gnu.org/archive/html/bug-hurd/2026-03/msg00045.html To be fair, it’s not run by “the project” but rather by an individual contributor. I would like to see pushback on this practice from other people, while being respectful to this contributor—who is genuinely trying to help with their own time and money. From the maintainer’s viewpoint, I find this situation difficult to deal with and very discouraging. Not the way I envision free software.

lists.gnu.org

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 4mo ago

Replying to

@jfred@jawns.club Actually it’s almost there: https://issues.guix.gnu.org/75595

issues.guix.gnu.org

0

1

0

Ludovic Courtès @civodul@toot.aquilenet.fr

· 5mo ago

Replying to

@hipsterelectron Yeah sorry, that was poorly worded! Rather I guess we can conclude from this that there’s some acceptance of genAI-produced code, but of course with varying degrees and differing policies. (The fact that many projects have policies in place suggests they are, indeed, critical, regardless of the take of their policy.)

0