Re: Attempt to Triage Bug 329386

[Elijah Newren <newren gmail com>]

On 2/11/06, Lex Hider <lexhider internode on net> wrote:
> That's interesting. How are we moving to "such a point", and what things
> has bkor added to "make this better"?
> Also, what is the "needed infrastructure" of which you speak.

The rules for whether a created bug should be unconfirmed or new used
to be really bad.  I don't remember the details anymore, but I
remember it being messed up.  That has been fixed.  bkor has added
some code so that e.g. bugs filed by developers are automatically in
the new state.  A lot of his triaging stuff is enabled for unconfirmed
bugs only (though I personally would prefer most/all of it being
available for other bugs too).  I think he's been looking at other
stuff too and may have done other things I've forgotten.  He'll have
to speak up here.

> I do think we should make a decision here. If it's not something that is
> of any use to someone, remove or change things so it is useful.

We're trying to change things so they're useful.  But fixing bugzilla
takes time and few people work on it.  (In fact, only bkor is doing so
right now; I drop in every once in a while and start doing stuff, but
he's doing 95% of the work.)  I consider the new vs. unconfirmed
suckiness that currently exists to be a temporary situation.  One that
has lasted many years, granted, but one that will be killed sooner or
later.  And then my long diatribe on not confirming bugs will be
totally obsolete and wrong.  At long last.  :-)

> I think the idea of developers having only to look at confirmed bugs is
> definitely attractive, and could perhaps become a goal of the bugsquad.
> I do realize that this would be a goal and not something that could
> actually be acheived overnight.
>
> It is also reliant on the bugsquad being knowledgeable enough to
> distinguish between "real" bugs where the bugs are highly technical.

It also depends on the ability to have developers not be emailed for
unconfirmed bugs, an ability we don't yet have.  I think that's the
biggest obstacle.  I think we have sufficient manpower to push bugs
from unconfirmed to new in such a case where developers don't see the
unconfirmed bugs.