Re: Bug tracking

Hi Joe,

We're looking at elements of the Tigris architecture for the site
that I referenced in an earlier email (so Don, we're not reinventing
the wheel).  Scarab has apparently been in development for awhile
now - the Tigris project is currently using Bugzilla until Scarab
itself is ready for prime time, then they intend to switch it over.

We've also spent some time looking at an app called BugRat (even
ported it to Postgres from MySQL)... it seems to offer a pretty good
mix of features as well.

The greatbridge.org site will likely get started with either BugRat
or Bugzilla, then transitition to Scarab when/if it's ready.  BTW,
the goal of the site (which I didn't really explain too well in my
earlier message) will be to provide a hosting infrastructure for
some related projects (like interfaces, or the apps and tools we're
listing at http://www.greatbridge.com/tools/toollist.php)... also,
all the software we develop internally will go up on that site.

So please don't interpret our interest as getting in the middle of
what the main Postgres project uses for bug tracking - although the
requirements outlined in the earlier messages was helpful just for
our own development of the greatbridge.org site.  If Ben and the
OpenACS gang are willing to put the time into a bug tracker for the
project, I think that couldn't help but be a good thing.

Regards,
Ned



Joe Brenner wrote:

> Well, just to throw another piece of information into the
> mix, there is a new bug-tracking system under development by
> the folks at collab.net.  They call it "scarab", and last I
> talked to them that they thought it would be ready for
> production use Real Soon Now:
>
>    http://scarab.tigris.org/
>
> I guess they regard this as a replacement for bugzilla.
>
> Some personal opinions:
>
> (1) I would actually like to see bugzilla fixed rather than
> replaced.  The collab.net guys are into java servlets
> because they're CS geeks who are down on perl.  Me, I'm a
> perl loyalist who thinks that Larry Wall is onto something
> -- mathematical elegance may not be the right standard to
> judge a computer language.
>
> (2) And maybe it'd be nice if the "religous wars" could be
> dropped in favor of pure objective technical decision
> making, but i don't think they can be: the social and the
> technical don't neatly split into two little piles.  (A case
> in point: the argument that using a mailing list for bug
> control is somehow "warmer" or "more human" than a bug
> database.
>