[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: The new installer



> On Tue, Feb 26, 2002 at 02:31:43PM -0500, Brian Mays wrote:

> > Indeed.  Personally, I would rank "faster feedback on changes in bug
> > status" as the most important trade off and a significant advantage
> > of the new system.  Why should others be working to fix a bug that
> > has already been fixed in an uploaded package that has not yet been
> > processed?  They should not, and the problem should be marked as
> > resolved by closing the report.

Adam McKenna replied:

> That's why someone else suggested that the bugs be marked "fixed".

A valid suggestion, but in my opinion, it needlessly over-complicates the 
whole process.  I prefer to have the bug closed and the issue concluded.

> If you close bugs immediately, before new versions of the package are
> available, then you run the risk of people submitting duplicate bug
> reports. ...

That risk is not affected by closing the bug report immediately, since old 
"resolved" bug reports stick around long after the new package has been 
installed into Debian on all architectures.  If someone is submitting a 
duplicate bug report, it means that he has not looked at the existing list 
of bugs for the package at all, an all too common occurrence these days.  
I seriously doubt that the number of duplicate bug reports will be 
increased in any significant way by closing reports as soon as the new 
package has been accepted.

> ... Granted, that's not as much of a waste of time as if two people were
> working on the same bug, but given that people generally don't work
> on bugs "out of the blue", (unless they are NMU'ing), but rather with
> close communication with the package maintainer, I don't think that's
> as much of a concern.

True, there is not much work being done out there to clear up existing bug 
reports by anyone other than the maintainer of the package.  Nevertheless, 
I have, in the past, stumbled across a bug that has caused me trouble and 
have taken the time to fix it and submit the fix to the maintainer of the 
package.  If I had seen that the bug had been fixed already in the next 
release (whether or not it had been installed in the archive), I surely 
wouldn't have bothered.  Still, such a situation is rare, which was your 
point.

- Brian




Reply to: