Hi Gabriel, I agree with Rutherther. The final state of GCD 007 is ’withdrawn’.
On Mon, 18 May 2026 at 11:59, Gabriel Wicki <[email protected]> wrote: > Regarding GCD007's state (and the question of how to go on from here), > we are at a crossroads. My understanding of the situation: hardly reaching the quorum even after the deadline, getting one unresolved disagreement expressed before the Deliberation Period and tweaking the proposal until the very last minute for the good or the bad, well, IMHO, the lack of quorum under the Deliberation Period and even a very light count after the deadline raises a signal about the state of GCD 007, no? > - be pedantic about timings and set the state of GCD007 to `withdrawn' > and dismiss the document as is. It will be unclear if discussions > will ever be picked up, where from and if we eventually, after > yet-another three month nervewrecking procedure, reach consent in the > form of an `accepted' GCD state. Well, it doesn’t appear to me being pedantic about timings. It’s the rules we collectively decided and agreed on. We cannot selectively tweak the frame aposteriori that GCD 001 installs. Well, I’d like to quote GCD 001: Similarly, withdrawal does not necessarily equate with rejection; it could mean that more discussion and thought is needed before ideas in the GCD are accepted by the community. And as I wrote elsewhere in this thread or as this Codeberg message [1] states, the project needs to clarify “team membership”, at least, and from my point of view, also the role of maintainers. Therefore, this discussion will be resumed later. However… > After authoring and "leading" discussions for this GCD it is obvious to > me that the GCD process needs refinement. The process is relatively > young and much unlike software development social processes can not be > abundantly tested before "deploying in production". I regard the GCD > process as experimental and to be useful for our project it will most > likely need some adjustments. Treating implementation details (like > deadlines) as cornerstones of consent finding seems rather odd to me. > But I guess that's another discussion. …I agree. The process defined by GCD 001 needs to be refined. At the last Guix Days, we discussed a process for “versioning” some modifications about the accepted GCDs; what I’ve sometimes named a GCD for “amending an accepted GCD”. Well, I should come soon with a proposal if no one beats me. Maybe a potential submission after the end of the Discussion Period of GCD 008 in order to avoid an overload. Once we agree on a process that sets how we modify (fully replace or not, version, etc.) an accepted GCD, then this process will be used for modifying GCD 001 based on the identified drawbacks. Yes, it might appear long but what’re some months at the scale of the project time-life. ;-) > And then, there is that third way. After all, who gets to decide about > a GCD's state? There currently is no team or role with the authority to > do so... As for all the things in Guix, the current implicit status quo appears to me: the final word when strong disagreement is from the Maintainer collective. Do we have a strong disagreement here about throwing the deadlines of GCD 001? BTW, from my point of view, the Sponsor is the guard for having a process on the road. Well, the Sponsor acts a timekeeper. - A *sponsor* is a contributor who, during the submission period (see below), informs the author(s) that they would like to support the GCD by participating in discussions, providing constructive comments to help the author(s), soliciting opinions, and acting as timekeepers. As a sponsor, please make sure that all participants have the time and space for expressing their comments. And as sponsor, I’ve probably failed on some aspects details there but I think I’ve clearly set the deadlines. Well, the Discussion Period had been 2 days longer because I was travelling thus mostly offline these two specific days. > If we remain stuck, we could also leave the state as `withdrawn' (or a > bit more adequately set it to `undefined'), and pick up possibly > emerging discussions on the specific improvements from GCD007 in the > form of Pull Requests on Codeberg. As quoted above from GCD 001, ’withdrawn’ does not equate a rejection; it could mean that more discussion and thought is needed before ideas in the GCD are accepted by the community. To me, the final state of GCD 007 is clear: withdrawn. Cheers, simon 1: https://codeberg.org/guix/guix/pulls/8358#issuecomment-14708370
