Hello,
On 9/7/26 21:56, Ludovic Courtès wrote:
Hugo Buddelmeijer <[email protected]> skribis:
Firstly, you're side-stepping the main point of this thread, that it
will be difficult to amend the GCD.
I’m not sidestepping anything, but I’m not sure whether I agree with
this assertion.
One one hand, it’s clear that probably few people are willing and able
to put as much energy and time as I did trying to build consensus on
this one. Building consensus is hard, because you need to make
compromises: in its current state, this GCD is not what I was hoping
for, it has literally zero lines in common with my initial proposal, but
I think it is much better at reflecting where we are.
Yet, I don’t see why amending this GCD would be harder. On the
contrary, I think it can only be easier, because, as I wrote, we’ll have
more insight into all this by the time there’s enough momentum to make a
change.
Amending will be harder because the GCD process favors the status quo.
The current status quo is that "anything goes", and it is clear that
no-one is happy with that.
Everyone wants change, and everyone wants change in the same direction.
But once we have a GCD in place, then there will be several people that
don't want change, or only want change in one direction. I think the
conversation from April that I quoted shows how strong people can feel.
That's why I think it will be even harder to change the GCD than to find
consensus in the first place, because now everyone is looking roughly in
the same direction.
I don’t have a crystal ball though, and time will tell whether I was
right or wrong,
If you are right, and it will be easy to amend the GCD, then there is no
harm in keeping its reach a bit more limited now, because we can always
update it with new information.
If you are wrong, and it will turn out difficult to amend the GCD, then
it would be an even better idea to keep its scope limited, because we
wouldn't be able to correct mistakes. We would need to only include
things that we are absolutely sure will maintain consensual.
> provided we accept GCD 008 to begin with.
It should be possible to get a GCD passed, because we have consensus on
almost everything.
I believe we already have consensus about the topics covered by the
AI-policies/guidelines of other projects. E.g. the one from the software
freedom conservancy, or from godot. As in, we all seem to consider
those an improvement over the status quo.
Based on the discussions, I believe everyone in our community is even
willing to go much further than those examples. E.g. by disallowing
closed source models, by being much more reluctant w.r.t. what we allow
copy-right wise, or by being extremely restrictive on autonomous agents,
etc.
You can be proud of the community if you look at what we actually do
have consensus on. We seem more aligned than any other project.
Yet the GCD goes so much further into territory where we do not seem to
have consensus, to the point that the GCD risks not being accepted at all.
This is a model we rejected early on in Guix (see commit
154f1f0937754fafac0c6288dd458b66b332e6bb) in favor of collaborative
maintenance, now incarnated as teams.
I don't know where to find that commit, it seems not to be in Guix,
could you please give me a pointer?
It’s in Guix.
Thanks. I don't know why I couldn't find it.
Thanks. I'm hoping for something that shows how far the packages are
outdated; whether that gap grows or shrinks.
The gap will not shrink magically by itself; we’ll have to roll up our
sleeves, write a bot, experiment, etc.—just like others did when writing
importers/updaters, ‘guix style’, and the likes.
Yes, and I experience the GCD as extremely limiting our freedom to
experiment.
And if this outdated-gap is widening, I think we owe it to ourselves to
at least explore these genAI options. We would do a disservice to
ourselves if we forbid ourselves from even exploring.
There are plenty of ways genAI can help us that perfectly adhere to the
justification as laid out in the motivation section and the other
objections raised in these discussions. So we should explore.
And yes, we should still do all those other things as well. There is no
need to artificially create a competition between these.
Hugo
