Free DVDs!

[Kevin Kubasik]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

The issue I more have with this subset of bugs (something I have also
tried to shrink several times before) is the simple and overwhelming
lack of bug quality this category tends to entail. Things like 'All
programs close' with no stacktrace are almost worthless, and many time
receive no response on clarification requests. Granted these are
generally what should be closed, but the issue that I find is I might
make a pass and ask for more info on 50 bugs, get 1 response, and I'll
follow up on that bug, but won't get around to going back and rejecting
the other bugs.

Just a thought, maybe I missed some policy on the wiki, but if we could
come up with a nice way in malone to mark bugs privately that might
help. Like personal bug folders or the like. So I could go back through
all bugs in my 'Requested Info' folder and mark them after a certain
time. Status and searches/tags can do this to some extent but I'm
generally tenative about such things since they are public and attached
to the bug, not to my profile.

Just my $0.02
Kevin Kubasik

Carthik Sharma wrote:
> On 12/12/06, Christian Robottom Reis <kiko at async.com.br> wrote:
>> To ensure the message is across: we want both confirmation (or
>> invalidation) of bugs, and indication of what source package(s) they are
>> related to. You get extra good vibes if you add upstream bug watches for
>> the upstream bugs, but the thing I'm really looking for is confirmation
>> and source package indication.
> 
> I have tried my hand at tackling this exact subset of bugs before. The
> biggest problem is that a majority of the bugs cannot be assigned a
> package -- "Ubuntu" -- best describes where the bug belongs. A bug
> entitled "(K)Ubuntu crashes frequently" is one example. Others deal
> with specific laptops/hardware and so it may not be evident which
> package the bug belongs to. My rough estimate is that about 100-200
> bugs of the lot will be of this kind - where a package isn't obvious.
> 
>> On Tue, Dec 12, 2006 at 02:17:34PM +0100, Henrik Nilsen Omma wrote:
>>> I think zero is a difficult task though because two new ones will always
>>> sneak in at the end. But you can of course determine that by looking at
>>> the dates.
>> You're right -- but I want to see that number get down to one digit, and
>> stay there. I know there is a lot of up-front work that needs to be
>> done, which is why I'm putting up this bounty (and also because movies
>> are always good karma!), but I want us to keep the number down once
>> we're there! I'll be okay with is stipulating an error margin of 20
>> bugs, but I'll be even happier with a hard zero!
> 
> Nice bounty though! Let me take a crack at getting there :)
> 
> Carthik.
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
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=c71G
-----END PGP SIGNATURE-----