On 03/04/13 19:37, Quim Gil wrote:
http://www.mediawiki.org/wiki/Requests_for_comment/Wikitech_contributors#Fir... is supposed to be completed in 3 months, and even there you have some easier tasks that could be implemented pretty fast, namely forms & templates for
- User profiles.
User profiles are not that helpful. They may serve to bring some folks together and it'd probably be nice for those folks to have them, but adding forms and templates and further complexity for that is not something most of us need or want. Using something like SocialProfile might be a good way to get it set up quickly if you really want it, but anything more would entail resources better spent elsewhere.
- Projects
- Tasks.
Specified projects and tasks add bureaucracy and make it harder for folks to jump in and just fiddle with things. You see the same ideas rejected time and again on wikipedia and other projects for precisely that reason. We do not need more confusing barriers, we need less - we need to be able to jump in and do stuff, or people will be less inclined to bother.
If you feel that more design resources / budget should be allocated on a nicer UI then I can also ask for it, but in a context of limited resources the priorities would be clear.
There are plenty of things that definitely do need a nicer UI. Gerrit comes to mind, but UI is only part of it - and it is indeed probably the biggest barrier we have that new contributors and old alike face. I would seriously suggest addressing that /before/ getting into the details of what brings folks into the fold in general.
And yes, I know what you said about gerrit being upstream and having its own community and issues and all that, but when the entire platform is flawed from the ground up, perhaps there are other things to consider. If the only way to resolve the problems it places on our community is to replace it, then we should seriously consider replacing it. And if there is nothing with which to replace it, then perhaps we should be using resources otherwise spent puttering around initial details to do so, because those initial details don't mean a whole lot even if they do help bring in more contributors... if said contributors still wind up crashing into the sludgy brick wall that is gerrit.
Obviously I'm not terribly fond of gerrit, but it's not for lack of reason. Another active thread on this list demonstrates a piece of it.