Nixpkgs committer proposal: Mentor assignments

[infinisil]

This is just a thought for @Mic92 @jtojnar @NickCao's delegation of Nixpkgs commit access:

It would be really good to assign a mentor to every new committer, who they can ask for any help with committing, or escalate issues to they don't feel qualified to handle.

Here's a concrete idea:

• Create an issue pinging all Nixpkgs committers, asking them whether they'd like to be a mentor or would like to get mentored on commit access.
• Assuming there's enough mentors, assign at most a couple mentees to a single mentor.
• Assign every new committer to a mentor the same way.
• Figure out some process for graduating from a mentee to a mentor, perhaps at the recommendation of the mentor.

How does that sound?

[Mic92]

It could be a bit difficult to find good mentors that also would work well with the new committee. So I am wondering, if the person that proposes someone, should be either the mentor or for self-nomination they maybe should find a person themselves.

[Mic92]

We started to assign mentors. Now we are waiting how this pans out.

[fricklerhandwerk]

Maybe write down the process in Nixpkgs contributor docs? How else would one know how any this works?

[wolfgangwalther]

After mentors had been assigned to new committers in January, the next round of committers didn't get any in #50 (comment).

We're going to hopefully have a new set of committers coming in soon, plenty of applications already.

@NixOS/commit-bit-delegation Is the mentoring proposal still something that is being considered or is it essentially dead?

[pbsds]

How did the mentoring actually go in practice? I think feedback around this could be useful. :)

[mweinelt]

I'm opposed to a repeat without a plan, documentation and a feedback cycle.

[wolfgangwalther]

I'm inclined to agree and just close this issue as "won't do" - I don't think it actually worked out well. But I have very limited insight into the mentorships that happened. I just moved the issue here, because it's not supposed to be in the org repo.

[philiptaron]

The past year has seen a dramatic slowdown in our granting of commit bits, with big delays in granting access to exceptional contributors. If the problem of finding mentors was a part of that slowdown, that's reason enough to jettison it.

I support folks acting as mentors or accepting a mentorship role, but it's more important to get the process working with more alacrity than it is to ensure mentors are assigned.

[winterqt]

If the problem of finding mentors was a part of that slowdown, that's reason enough to jettison it.

Speaking for myself, the problems were multi-faceted:

• the issue was not the best way to get a grasp of pending requests, and maintaining our internal notes in addition was a chore (@infinisil's PR-based process will make it much better, thank you!)
• all of us were busy for a good chunk of the year, 2/3 of us because of the SC
  • I can't speak for @jtojnar, but now that I'm not on the SC, I can dedicate actual proper time to this again.

The ambiguous state of the mentor situation certainly didn't help, but it's definitely not the biggest reason.

For simplicity's sake, I also support retiring this for now.