Cool system, Brion!
On Fri, Oct 3, 2008 at 11:57 AM, Brion Vibber <brion(a)wikimedia.org> wrote:
Aryeh Gregor wrote:
Bureaucrats can also add people. I'm
assuming that people who have
had commit access for a reasonably long time and have shown they know
how to use it should be marked coders, but I'm still really uncertain
given that nobody's actually said what the various statuses are
supposed to actually mean. Are people going to actually scap on the
basis of nothing other than the fact that every commit is marked
ok/resolved? If so, it's probably a bad idea for people other than
Tim or Brion to add those markings, at least on a regular basis,
unless we really want that.
The current theory is we'd like it to be easy to mark things as needing
*more* review, but hard to mark things as needing *less* review.
So that probably means a split-level permissions model, perhaps with
distinct pre-review and super-review (to use the Mozilla term -- patch
reviews get "super-reviewed" by a core committer, or some such crazy thing).
It is, of course, an evolving system. :)
I was thinking somewhat along those lines. It's nifty to be able to
have people pitch into the review process - it might take some load
off Tim and Brion for people to be able to point out problems with
code before it hits the final review - that way, mostly good stuff is
left for Tim and Brion to look at.
We could also have a preliminary tag for "looks okay, but needs to be
looked at by somebody else before it's put live", so that less
thorough reviewers (e.g. myself) would be able to look over code
without actually marking it as "definitely OK", which should be
reserved for thorough, trusted reviewers like Tim and Brion.
It also might be a good idea to add a 'tested' tag - for code which
has been verified to at least *work* (ideally, of course, this will be
done BEFORE committing, but perhaps testing by somebody else would aid
in the review process).
Just a few thoughts...