Reporting bugs

[Jim Gettys]

There is a balance here: I may not be hitting it the balance right.

Any report of bugs is goodness; but if trac's signal to noise ratio goes
bad, we can't see the forest for the trees.
.
A reproducible bug is much more valuable to us than ones that are not; a
bug against current bits are much more valuable than old ones, exactly
because we have a better chance to go fix it. Having developers spend
time wading through bugs that there is no way to reproduce, either
because the recipe for doing so, or insufficient information on what
versions were being used.

So if a bug clearly needs more information to be able to be chased, it
seemed reasonable to me to request information from the reporter, wait
for a few days, and then close it if the reporter does not bother to
even acknowledge the request for more information.

Is this reasonable, or not?
                                   - Jim


On Tue, 2007-10-30 at 03:17 -0400, Albert Cahalan wrote:
> Bert Freudenberg writes:
> 
> > I've had comments like this on some of my own reports. These are
> > usually not reproducable, strange things. Having to follow through
> > when this is far from my own area of expertise simply takes time I
> > can't effort. There indeed is "lack of interest on the part of the
> > reporter", no denial here. I'm just saying that because of that, I
> > stopped reporting those one-off bugs. Seeing comments like yours
> > reinforces that decision.
> 
> This is sad, but understandable.
> 
> What is a bug reporter supposed to do about something that is
> very difficult to reproduce? Just not report it?
> 
> I've hit a few bugs that I can't seem to reproduce. This does
> not mean the bugs are non-existant or that they won't someday
> destroy something like a child's journal content.
-- 
Jim Gettys
One Laptop Per Child