Stepping Aside as Lead Maintainer

[non-Jedi]

So I've given it some thought, and I think I need to officially step aside as lead maintainer of the python sdk. Practically speaking, this means that unless somebody wants to step up and lead maintenance on the sdk, I'm going to add a disclaimer to the README noting that it's borderline unmaintained.

I don't have time to review PRs of any significance right now, and with grad school + full-time job continuing and baby number 2 due in 6 weeks, I don't see that changing any time in the near or medium future. It's not fair to would-be contributors to not have a disclaimer noting the sdk's borderline unmaintained status.

If anyone is willing to take the lead on maintaining the sdk, let me know in the next few days; otherwise I'll probably add a note to the top of the README this weekend.

I'll continue to review PRs when I can, but it's not fair to my family for me to take time from them to do so, so that will continue to be very slow if at all. This is really just acknowledging a change that has already happened.

[non-Jedi]

As a note of encouragement, the codebase is quite easy to grok, and I'll be continuing to hang out in #matrix-python-sdk:matrix.org and will be willing to answer any questions or give advice as needed. Even unmaintained, the sdk will of course continue to work as well as it currently does; the nice thing about a protocol like matrix is that spec-compliant implementations don't have to change if they don't want to.

[ara4n]

firstly, congratulations on baby #2; hope everything goes well!! :)

secondarily, huge thanks on behalf of the whole ecosystem for having led the maintenance for so long - the number of projects out there using the python-sdk seems to have increased massively, and with E2E support having landed I can only see there being more and more users in the future.

In an ideal world we'd have bandwidth to support the python-sdk from the core team, but in practice we're increasingly bandwidth constrained (at least until we get some sustainable $ in and can fund more folks). So I think maintainership is going to have to come from the community (which is also healthy of course in terms of being independence of the core team).

I wonder if @Zil0 might be interested? or @Cadair? or @MatrixCoffee? or basically anyone on https://github.com/matrix-org/matrix-python-sdk/network/members... (i wish github gave you an @all button)

[Cadair]

I am having to really fight the urge to volunteer. I would really not have enough time to do it justice. I havent even had the time to integrate the changes I have in my forks at stuff. :(

[Zil0]

I don't think I can commit to be the new maintainer either. But if no one has that kind of time or feels like taking on such a responsibility, couldn't some of the main users/contributors of the SDK work together towards maintaining the project? I mean I can review a PR from time to time, but not regularly enough to keep things going smoothly. If @Cadair and/or @Matrixcoffee and/or anyone else can do the same, I think we would quickly reach the point where we can say the project is still maintained.

[Cadair]

I am happy to take on reviewing things related to the API layer as that is the bit I am more familiar with. Maybe we could have a few different people with primary areas they look after?

[ara4n]

This could totally work as a team effort - I guess the main requirement for a lead maintainer is to worry about cross-cutting strategic things like "do we ditch python <3.6" or "do we support micropython" or "do we switch it to run off gevent" etc. Random thought: @hawkowl, would you be interested in helping on this with a core team hat on, being a pythonista an' all?

[ghost]

I'd be happy to pitch in from time to time, but I'm in pretty much the same situation as Cadair.

Additionally, my setup is way too old (Python 3.2) so if I were project lead I'd just hold everyone back.

[joeky888]

Train that baby to become the new maintainer. :)

I made a list of micropython libs here https://github.com/joeky888/awesome-micropython-lib

hope that helps.

[Half-Shot]

I'd consider taking a more active role in PRs in the near future but it depends how bridges pan out in terms of time. There's also the numerous other projects I'm on ..

[xloem]

Does everybody who mentioned opportunity to review PRs have access to merge them?

If github still has that auto-merge feature they tested where users can merge PRs on their own after qualifications like tests or review by X others are passed, it could be helpful to enable this.

If needed, I’ve had good luck in the past tagging everyone who has opened PRs and made forks recently, and asking directly; people usually step up because they’re using and deving the library. But they don’t always read issue threads on their own.