On Monday, 13 May 2013, James Cameron wrote: > On Mon, May 13, 2013 at 12:28:58AM +0200, Daniel Narvaez wrote: > > I'd like to propose to make it a requirement, enforced by code > > reviews, to provide good test coverage when submitting new code. It > > will raise the bar for contributions but it's essential if we want > > to improve quality (and I think we have to). > > Why? How has the quality tracked? >
Unfortunately we don't have very objective means to measure quality. Very little testing is done, even less bug triaging, as William pointed out in the bug tracking thread. So I'm afraid I can only report about my personal experience. I've tried a couple of times to use Sugar as my desktop and I've noticed pretty trivial bugs that you would expect to be fixed the day after they are introduced. I'd say subjectively quality was obviously very low. Multiple sections of the settings panel has been completely broken since they was ported to gtk3, until Walter fixed them a few days ago. The frequency of backtraces in our logs is so high that it makes development difficult, not quite a proof of poor quality, but certainly not a good symptom. The collaboration framework, which is an essential part of the user experience, is completely unreliable. Activities still randomly fails to start, seen both in automated testing and in use. Fortunately I don't think I need to prove our quality is pretty bad to assert automated tests are a good development practice. And the fact that very little human testing is done, is a good reason to increase automated one. > Or is it that you have too many contributions and you need to raise > the bar to slow them down? > I cant tell if you are being sarcastic here or not. If you are I think that's a bit ingenerous, I've been working a lot to lower the barriers and with some success. Anyway, no, I want the barriers to be as low as possible, but not at expense of code quality and good development practices. -- Daniel Narvaez
_______________________________________________ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel