On Sun, Oct 22, 2000 at 10:04:27AM +0200, Klaus Naumann wrote:
> Actually writing the bug tracker wasn't an act of boredom
> (if you meant it that way). I had the strong feeling that we have
No - I didnt think of it that way ...
> > The one sending these reports has to give possible additional
> > hints on what he did/patch/change to get it to work. So probabably
> > we will than be able to keep track on which kernel versions
> > worked on which machines.
>
> This is indeed a interesting idea. Writing it wouldn't be a big deal.
> But the question is if it's worth writing it. First we would need
> some ppl which actally send the info somewhere - I'm pretty
> much in doubt, that we will find enough so that it rents.
> Second question is if ppl are actually interested in such a thing.
> I'll not again waste time writing something where I don't get at least a
> small ammount of feedback and which noone is using.
Right - I just thought as a feedback for the kernel cvs commits the other
way round that you implemented - Sometimes (most of the time ?) bugs
keep not found because noone is using some specific feature (Like the
Timeing stuff in the decstations) and though a lot of people simply
think its their fault instead of complaining - But with something like
that and possibly a simple shell script we yould collect not negative
but positive information on WHAT is actually working. Nevertheless this
only works for a small subset of the mips stuff (kernel) and though might
be pretty useless as most of the kernel stuff works and userspace is
getting the problem.
Flo
--
Florian Lohoff flo@rfc822.org +49-5201-669912
"Write only memory - Oops. Time for my medication again ..."
|