Ludovic Courtès <[email protected]> writes: >> The document should be as clear as possible, otherwise the consensus >> is of no value if everyone has their own understanding. There was >> another similar, small change in eac798a510 that deleted >> "translations" when translations are still included in "any other >> artifact". How was this clarifying? > > Florian Pelz of the translation team suggesting removing this specific > example due to the current situation of Weblate². > > You are right that translations are still included in “any other > artifacts” though. Florian, thoughts? > > ² https://lists.gnu.org/archive/html/guix-devel/2026-06/msg00282.html
In the commit message of GCD 008 update https://codeberg.org/guix/guix-consensus-documents/pulls/13/commits/eac798a51081c60fab08b0ce024f76048afa79a0 Ludovic writes > The translation situation being somewhat out of control, with > translators submitting what looks like low-quality LLM-generated content > and being mostly disconnected from the project, Florian Pelz of the > translation team suggests not making explicit commitments in this area: > > https://lists.gnu.org/archive/html/guix-devel/2026-07/msg00006.html Because it is unfeasible to fix existing translations and new submissions, “not making explicit commitments in this area” is exactly right as-is, i.e. no explicit promises about Weblate. My suggestion is nothing more. Yes, translations are an “other artifact”, but generated machine translations also are nothing we accept when we notice, because they lack quality. Search “info guix-cookbook.es” for mentions of < it is clear we don’t notice. In the words of https://codeberg.org/small-hack/open-slopware for translations, “human reviews are scarce.” Does this really block consensus? Alternatively delete translations except French, German that get reviewed more. (This would be a worse outcome.) And also, we don’t need a sunset timer; a clash with GNU policy or new legal situations surely will lead to a new GCD or emigration from Guix anyway. Regards, Florian
