On Sat, Jan 25, 2014 at 06:22:14PM -0500, Rahul Sundaram wrote:
I've said before, I blame the software, not the people (and that includes not blaming the askbot authors -- it's just a really hard act to follow and SE is constantly improving and tweaking to get the experience just right).
There is no particular reason Fedora as a project couldn't do it too.
We could certainly put more into it, although if we switched every paid person working on Fedora to working on it and were able to also convince a huge number of community contributors that it is the top priority, we would still be significantly behind the resources that Stack Exchange has in putting things even further forward. I don't think that _necessarily_ means we should give up, but just to keep what we _can_ do in the right scale.
Other web applications hosted by Fedora typically has Red Hat paying developers to work on them - pkgdb, bodhi, tagger etc. Unfortunately I haven't been able to get Red Hat to allocate some engineering time or just engage the upstream developers via a contract which they are open to do. I did try via Spot earlier but that didn't work out.
There are a lot of development projects which would help Fedora. I don't have any insight into Spot's exact decision making-process, but we do need to prioritize, and there are more big-impact todo items than time. Right now, as I've said in other messages, I think HyperKitty and Taskotron are really our top priorities. Adam W. is lobbying for improvements to Bodhi, too. I know one can't necessarily scale up a project by throwing more people at it, but generally a team of software developers can et a lot more done by being focused than by having a dozen projects which all get pushed forward a little bit.
We do get community volunteers to help out by pushing patches upstream whenever we can but naturally that is a slower process. In the last few days, I have talked to atleast three different people who have signed up to complete a few features we need (marking questions as duplicate, pre-filled templates and disabling flags FYI). If we care about the user experience, we need to invest in it.
Agreed. My points are a) it needs a big investment and will likely continue to need it in order to keep up in usability and b) there is an opportunity cost in chosing to put effort in any one thing. I think having community QA is important, but in my personal judgment some of these other things are more pressing.