On Fri, Jul 22, 2005 at 03:34:09AM +0200, Martin MOKREJ? wrote:
I think the discussion going on here in another thread about lack
of positive information on how many testers successfully tested certain
kernel version can be easily solved with real solution.
How about opening separate "project" in bugzilla.kernel.org named
kernel-testers or whatever, where whenever cvs/svn/bk gatekeepers
would release some kernel patch, would open an empty "bugreport"
for that version, say for 2.6.13-rc3-git4.
Anybody willing to join the crew who cared to download the patch
and tested the kernel would post just a single comment/follow-up
to _that_ "bugreport" with either "positive" rating or URL
of his own bugreport with some new bug. When the bug get's closed
it would be immediately obvious in the 2.6.13-rc3-git4 bug ticket
as that bug will be striked-through as closed.
Then, we could easily just browse through and see that 2.6.13-rc2
was tested by 33 fellows while 3 of them found a problem and 2 such
problems were closed since then.
most likely, only a small minory of the people downloading a patch would register at such a "project".
The important part of the work, the bug reports, can already today go to lnux-kernel and/or the Bugzilla.
You'd spend efforts for such a "project" that would only produce some numbers of questionable value.