Far too many people seem to be using the issue tracker as a support forum. Perhaps you should make a policy that all support requests should go on stack overflow or a mailing list of something.. That way only reproducible issues are allowed in the issue tracker. Otherwise it makes it hard to see what the actual bugs are :) Just a thought...
We already have such a policy, though currently we're not strictly enforcing:
Should we turn down issues requesting support automatically?
It surely reduces maintenance overhead.
How do you think of this?
I agree. We need a ban feature on Github.
Do what I do with RABL, reply quickly and sometimes brutally. If it's not a bug, I immediately close it. I reply with a (possible) solution and sometimes I suggest taking a different approach in their code, using StackOverflow, etc. But I don't let their application or skill problems show up as an Issue that's specific to RABL.
As an open source user, I check out Issue lists of projects I plan to use to see if they're active in resolving issues and looking at this list of Issues - my initial gut reaction would be that the owners are not actively participating.
You simply have to be consistent and have multiple people who are "on it". I don't even write a lot for RABL, I spend 80% of my time replying to Issues, confirming bugs, etc. That gives the author more time to work his magic.
I also suggest going through any issues older than a few months and closing them. If it's still a bug, have them reopen the issue with more details after trying latest. It never hurts for someone to reopen an issue.
You guys also don't use labels much. You should come up with a system that allows you to help triage issues whenever possible. If you don't know if it's an issue but think it could be, flag it "Investigate" or something. That way other core admins can dig in.
I managed to close a dozen or so issues tonight without working too hard, mostly "needs info" tags that had sat inactive for months. If someone can't be bothered to provide the info needed then we shouldn't feel bad about closing their issue.
I think we should also close enhancement issues that have sat for more than a few months. Maybe they're good ideas, but if nobody cares enough to work on them then they're not going to happen.
I think that the best way to honor this issue is to close it ;)
Nice. I'd recommend setting up a google group or something to discuss features/provide general support - that might help by keeping the open issues to a minimum.
Already done, and referenced in the README: https://github.com/sferik/rails_admin#support