Re: Debbugs bug tracker for Octave

[Thomas Weber]

Am Donnerstag, den 29.05.2008, 18:48 +0200 schrieb David Bateman:
> Thomas Weber wrote:
> > Hi, 
> > 
> > Don Armstrong has agreed to us using the Emacs test[1] tracker. I attach
> > a not too short introduction as text, an HTML version is available
> > online at http://tw-math.de/~weber/using-debbugs.html.
> > 
> > "Using" here means testing. If we decide to continue with debbugs, we'll
> > need to host an instance of it somewhere else.
> > 
> > [1] I don't know how much of a test tracker this still is or if the
> > Emacs guys have already settled on a decision.
> > 
> >     Thomas
> > 
> 
> These things work by the desire of people to actually use it. I think we
> have to try and make an effort to place all new bugs in the system so
> that this works, 

Well, quite a lot of bugs are actually reported by the developers, so I
thought they could start with their own bugs ;)

I don't want to put too much real data (especially from occasional
reporters) into the bug tracker yet. This is a test tracker and I see
good reasons for _not_ choosing it.[1]

So, maintainers should play with it to the point where they at least
have some feeling about it being good or not. But hey, that's just my
opionion, the tracker is open.


> perhaps redirecting address@hidden to debbugs.

John proposed this already, but as I said, I think people should start
by using it for either bogus test-reports or trivial things, before
committing real stuff there.

> One concern I have though is wiki spam. See for example
> 
> http://emacsbugs.donarmstrong.com/cgi-bin/bugreport.cgi?bug=304
> 
> If we use address@hidden as the gateway address it will be filtered,
> which seems to me to be a good thing.

As we want an open tracker, spam will be an issue no matter what. An
e-mail based tracker has the advantage that there are several
off-the-shelf filter solutions available.

[1] Actually, I also see very good reasons _for_ using it, but well,
that's what tests are for. 

        Thomas