Open Source developers only: how much time per month do you budget to maintain a one-developer project?
Thank you all so much for the replies.
This is a repost of a previous poll I'd made, which really didn't resonate. I asked, How many hours per month should an unpaid Open Source maintainer dedicate to a project?
Although as a maintainer, I felt it was an equivalent question, a *lot* of people got upset that anyone would say what a developer "should" do, especially if they are unpaid.
I found this response really disrespectful. Doing this job well really does require nonzero work, even to keep a project in the same state with the same functionality.
Our programming languages and dependencies and platforms keep changing even if our code doesn't, and we have to adapt or fall behind.
@evan I saw (and voted in) this poll and not the original. But for me, the two phrasings have very different implications. "Should" immediately raises the questions
1. In order to achieve what?
2. According to whom?
And carries with it a much greater sense of obligation
This one just asks about a lived experience
It is possible to leave a package unmaintained. I have dozens of projects on GitHub and Gitlab and Sourceforge (maybe...?) that I haven't touched in years, sometimes decades. There's no shame in that -- especially not in prioritising other things in your life.
Anyways, down to my answer. This is based on my 30 years of making and sharing Open Source software. I am open to lectures from nonpractitioners but please recognise that I speak from experience.
I think a project takes a few hours per month to keep running. That's testing and merging Dependabot PRs, triaging bug reports and hopefully fixing a few, checking on security updates and hopefully fixing some of those, too.
Adding functionality or improving performance or reliability takes additional time.
So, I'd budget about 8 hours per month. For me, that means I can only realistically maintain about 1-2, maaaaybe 3 projects at a time, unless I'm getting paid.
I don't think we do Open Source maintainers any favours by hiding these numbers.
It's OK for volunteers to do a job badly, or not to do the job at all. But that doesn't mean the job doesn't take real, measurable time to do well.
Knowing what that time demand is can help them prioritise relative to other parts of their life, or find other ways to get the work done, or accommodate themselves to not doing the work.
Anyway, thanks for the feedback and the replies. I hope for people who are new to the practice or are wondering how they compare to others, the results are helpful. I've definitely found the experience useful.
@stenhaastrup yes, I understand that it sounds different to you.