guix-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Discussion on Guix funding // future


From: Ekaitz Zarraga
Subject: Discussion on Guix funding // future
Date: Fri, 25 Oct 2024 00:08:18 +0200

Hi,

Recently I've been discussing with other members of the Guix community about several things we consider we could be improved.

The most important one in my opinion is the funding. I don't know (does anybody know?) how Guix is funded, and it worries me.

I've been funded to work on the bootstrapping part of Guix by NlNet grants. I've been extremely lucky, and I'm very grateful for it. And I tried to spread the money, paying people who deserved it.

Grants are great for specific issues, but we are not going to make Guix survive using only that kind of grants.

First of all, these grants don't pay much, and they are just for a year or so. Many of us have the technical skills to get a job that pays way more than a grant and is way more stable. This makes doing something ethical and good become a punishment, and it's forcing many people to choose. Most of the people don't have the privilege to choose.

Second, grants work kind of well for specific tasks, but what happens with the structural work? Is anybody actually getting paid for it?

Finally, grants push individuals to try to do things, but don't encourage collective action (also the amounts are not high enough for collective action). That's not necessarily bad, but those individual projects also drain energy from those who are structural to Guix. Patches have to be reviewed, and commits need to be merged.


On a side note, I think we are missing reviewers, maintainers and commiters, and I think that view is shared in the community. Let's use my case as an example: I raised my hand to become a commiter, and I don't know how that was lost in the mailboxes and nothing happened. At this moment, I don't care anymore: when I need to make a commit, I know there's people that trust me and I just ping them and they do it for me. Should I bother people try to get commit access again? My life is very comfortable as is... Some questions come again to my mind: Am I really ready for the challenge? Am I going to be a good commiter? Is it fair to continue like I am right now?

This issue and some others could be fixed with money. Simple, huh?

I think we should try to invest more on the people, and that probably means paying them for the work they do. At least to some, so they can invest more time and care in others.

This we can't do with grants with the NlNet flavor. We need other kind of approach.

Sovereign Tech Fund has a very interesting model for maintainers, but still lacks the ability to invest on people freely.

Many people has been thanklessly working for this project, and some will continue to anyway, but not having a proper funding model is probably keeping us in an uncomfortable situation. The lack of people is pushing away new people, and we are in a vicious circle where I think people that are less stubborn than me just go spend time on other projects.

We have had cases of people giving too much for the project for too long. I don't think we acknowledge that enough, and probably we should. We should take care of our people.


I think free software projects use to be precarious and we are too used to that. However, I think we should try to break with that image, and try to push for funding collectively, so we can cover structural costs: people and machines.

I think I'm just somehow sharing my will to help, and also trying to encourage some conversation about the funding and how we could do better. If anyone has ideas, please share.



On a second (and last) side note, I also discussed with some members of the community about the status of Guile. I may send separate email for that, but it would be great if we could use some of the energy we have to give Guile some love. We are too Guile-dependent to just let it rot.

Thanks for all you do,

Ekaitz




reply via email to

[Prev in Thread] Current Thread [Next in Thread]