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

Re: xqf removed from potato because of 1 bug...

On Thu, Feb 03, 2000 at 06:43:09AM -0500, Michael Stone wrote:
> The decisions about which packages "we can live without" are not
> immediately obvious. There are a large number of pacages on the RCB list
> that haven't been fixed for months and which I wouldn't loose sleep
> over. Other packages are targeted immediately after they get a bug
> report. If there were a harder line, and more packages were targeted, I
> think it would be more fair. As it stands, the process seems very
> arbitrary.

The main reason it seems arbitrary is that I stopped assigning [REMOVE]
tags when the freeze began; I've been far too busy for that since then.

The point of those tags was that I couldn't remove packages before the
freeze (since there was no "unstable" to leave them in), and I wanted
a way to see how many bugs would be left after the initial round of

Even back then, my time was too limited to evaluate all the RC bugs
being reported.  What you're seeing is the subset that I have managed
to evaluate.

Look at the math... 5 minutes per RC bug, with 270 bugs, is 22 hours
of work.  That would take up my Debian time for a whole week, and 5
minutes is not all that much.

That is why I _must_ rely on the package maintainers, or other interested
parties, to evaluate the release-critical bugs on their packages.  If
they don't, then there is a possibility that a package is removed when
it needn't be.  That's unfortunate, but I'm not going to let it cripple
the release process.  I'd rather release 95% of potato in a few weeks
than 100% in a few months.

The reason I haven't removed more packages yet is that doing so is
a lot of work.  Each package doesn't take much time, but it adds up
(see math above).  I'll do another round soon, though.

Richard Braakman

Reply to: