I feel like IRC is yet another obstacle to newcomers, in addition to email based git flow, debbugs, guile stack traces and zero editor (or very early WIP) integration except for Emacs. This is literally vendor lock-in. I've been contributing for years and now i almost have no trouble, but it was painful and I don't think it is fair to expect everyone to go through all this while with Nix you just need to open a PR.

What is the point of building a completely free system, that does not try to extract value from users, and actually tries to emancipate them by offering a trusted computing ecosystem, if no one gets to enjoy it because you made it so inaccessible that people are not able to use it? I'm exaggerating but I think you get the point. Now with efforts like the survey it looks like a fresh breath of air just entered the project, and the situation with contributions is a little better than a couple of years ago. I really hope we can pull an effort to make the bar for using and contributing Guix a little lower than it currently is, I am convinced that if we make some effort more people could liberate their computing environment with Guix

I feel kind of bad about this but I refuse to join Guix IRC. I use mailing list out of frustration but these communication channels are the proof that Guix is not only a nice, useful and open project but it has born a project for fossbro babyboomers. It is a golden walled garden for themselves , I don't feel ok in that space.

EDIT: I've answered a little emotionally. Let me clarify, I believe most Guix maintainers act of of good will and they want to find some communication platform which is inclusive for everyone (since it is clear also to rocks that IRC is good only for someone born before 1990, so it is good for people aged >= 35) . Some of the maintainers, and some most noisy members of the community make it so bad for everyone else but themselves that, having so much explicit and soft power, the discussion about moving away from communication protocols older than CDs was closed stating "everything is perfect as it is, we reach exactly the right set of people, we do not care of increasing the userbase or making the community more inclusive".

Would be pretty useful, as far as I know there is no way to change /etc/{subuid,subgid} in the system configuration without manually editing.

Well I had to make one :) it is being tracked on https://issues.guix.gnu.org/72337 . You can define subuid and subgid ranges like so:

(use-modules (gnu system shadow)      ;for 'subids-service-type'
                         (gnu system accounts))   ;for 'subid-range'

(operating-system
  (services
    (list
      (simple-service 'alice-bob-subids
                      subids-service-type
                      (subids-extension
                        (subgids
                         (list
                          (subid-range (name "alice"))))
                        (subuids
                         (list
                          (subid-range (name "alice"))
                          (subid-range (name "bob")
                                       (start 100700)))))))))

which would yield

# cat /etc/subgid
root:100000:65536
alice:165536:65536


# cat /etc/subuid
root:100000:700
bob:100700:65536
alice:166236:65536

Another annoyance with podman on guix is making / a shared mount doesn’t work so changes in mounts aren’t propagated.

I think I solved that by using a Shepherd service run on boot calling mount --make-shared / . I didn't do extensive testing of mounts but I'm currently using this on my systems as it's set up in my personal channel. By adding the following to my own system config

(use-modules (small-guix system accounts)
                         (small-guix services containers))

(service iptables-service-type)
(service rootless-podman-service-type
               (rootless-podman-configuration
                (subgids
                 (list (subid-range (name "alice"))))
                (subuids
                 (list (subid-range (name "alice"))))))

I'm able to run the following rootless Podman hello world

$ podman run -it --rm docker.io/alpine cat /etc/*release*
NAME="Alpine Linux"
ID=alpine
VERSION_ID=3.20.2
PRETTY_NAME="Alpine Linux v3.20"
HOME_URL="https://alpinelinux.org/"
BUG_REPORT_URL="https://gitlab.alpinelinux.org/alpine/aports/-/issues"

and with guix shell podman-compose I'm able to run this Podman compose hello world:

$ mkdir data
$ echo hello world > data/index.html
$ podman compose up -d

...

exit code: 0
$ curl localhost:8080
hello world

So some kind of mount appears to work. Thank you for your feedback and feel free to try the service from my own channel if you are interested in providing more or in trying rootless podman on the Guix System.

It is quite basic still and historically had some problems with reboots, but lately it has gained a lot of attention and bug fixes. I have no experience of runnit, compared to systemd is leaner but in my experience there's no big feature lacking

imho having a more accessible contribution workflow would increase the number of people interested in spending time in reviewing.

Regardless of the many problems of web forges today the ability to review only what's changed between the various revisions of a pull request and the comments in a single view is not achievable with a simple email workflow. You end up reimplementing the PR/MR functionality with other tools, exactly as is happening with the Guix QA tools. I love them but we're reimplementing gitlab/gitea/codeberg by parsing patch revisions from emails subjects.

it's not like it's easy contributing to guix :( i really hope they move to something more accessible

If you don't have commit access you don't need an account. I contributed many times without ever registering on Savannah.

Do you have commit access? Otherwise your account is mostly useless also if they don't delete it.

Working Screensharing from first boot lmao

Yes I definitely agree: free software is also a requirement for bootstrappability and reproducibility.

You mean reproducibility and bootstrappability?