Free Software projects should endeavor to make successful contribution approachable for people "scratching an itch" instead of demanding perfection. Otherwise eventually there will be no pool of future contributors. Too many projects (seemingly ones with mostly paid developers) forget this and suffer for it, eventually.
@neal
In the abstract I would like to become a regular contributor to some project, but no single project felt interesting, and I doubt I'd have the energy to contribute long-term.
I think there are many drive-by contributors like me.
Do you think I still in some sense count towards a future contributor pool?
And as a lot of projects have similar kind of workflows and idioms (and flaws…), getting help from one project is likely to help in the next one.
@neal I think there's a big difference between "I've scratched my itch, and now I'm considering putting in effort that would compensate others who deal with my itch", and "I've scratched my itch, and now I want someone else to maintain it for the rest of eternity". the former, I'm interested in; the latter, I don't really care about
@whitequark @neal It was "'my' itch was for a trivial use case to work with macos, which involved a one line change to fix a spec compliance bug". The response was silence for 4 months, followed by false accusations when I promoted a bug attachment proposed patch to a PR, and demands for elaborate synthetic testing on Windows for hypothetical out of scope situations, all with an attitude.
@neal actually they should declare a lot of projects "done". And maybe even start removing a shit ton of features as well.to reduce the unnecessary code that no one fucking asked for.
@citadel9 I disagree with this. Mostly because you can't possibly know that "no one fucking asked for" those features or code. But also "done" in FOSS means "dead" for a variety of reasons, and *someone* needs to keep up with the changing world.
@neal @citadel9 for instance, as a minimum, pretty much any program that accesses the internet cannot ever be "done", because vulnerabilities need to get fixed and any encryption it uses will need to be updated to follow the times
and until someone invents a universal unchanging graphics/UI interface, pretty much any program with a GUI will eventually need updates to not end up depending on a deprecated version of GTK/Qt/...
@neal
I'm a drive-by contributor.
I use some software for years, find a bug that annoys me, investigate it, make a patch, then if I'm not particularly tired, send it upstream. If it gets merged, I'll continue to happily contnue use that software for years without sending more patches.
I don't think any maintaner has any reason to invest their time guidng me through the contribution process, even though I appreciate when they do.
1/