tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jon Stevens <>
Subject Re: Any progress on the bug tracking system?
Date Tue, 18 Jul 2000 03:52:17 GMT
Ug, I love this redundant bantering.

on 7/17/2000 8:28 PM, "Nick Bauman" <> wrote:

> Sorry, but Buzilla=="another mess"???

Uh, have you looked at the code? Not only has it been found in the past to
be rife with security holes, but it is also done as perl cgi's. Gag! 1991

Summary: The source code is a hacked up mess.

> And a
> replacement is _months_ off?

Yes. It is not easy to do a issue tracking system. This is simple to deduce
by the fact that no one in the OSS world has gotten it right yet. I'm trying
to change that.

> Why not stick with "the
> simplest thing that could possibly work [today]" and
> not try to do a Taylorist approach to everything?

Because bugzilla didn't work. While it wasn't the complete reason, it was
partially responsible for the break in. Sorry, but a product only
gets one break in chance on

> Do I
> smell a whiff of "not invented here"?

Give me a break. We are all intelligent people here. Don't make useless
statements like that.

You might smell a bit of Perl vs. Java Servlets, but I don't think that is a
bad thing (especially to the people on this list). Having multiple
implementations of something in different languages isn't necessarily a bad
thing. In fact, I hope that Scarab eventually has multiple front end
implementations to a really nice schema on the back end.

> I mean, whats' wrong with Buzilla?

The code is an absolute mess, the schema is not normalized and the HTML UI
is difficult to use. Note that that is all expected from a project that was
never intended to see the light of day. It was an internal Netscape project
that only became open because became open and they needed
something that at least kind of worked for them. Early versions of Bugzilla
even had a tcl requirement.

Did I mention that Bugzilla currently only works on MySQL? RHAT did an
"oracle port", but I think that is a pretty bad way to deal with things.
Scarab will out of the box support all major SQL databases through the use
of Turbine's OR tool.

> Maybe I can help?

<>. We are currently in database design phase as
well as the fact that I have Cameron Barrett <>
working on the UI. I'm meeting him in the SF city tomorrow to discuss the UI
further. I can send you a URL of the most recent schema although it will be
changing a bit more soon since Brian Goetz is also helping me on that.

The reason why it is taking so long is that I'm the only paid person working
on it and I have been recently pulled off to work on doing a transparent
http proxy called Noodle (which will also become OSS) for the last week.

Noodle was done so that we could write a HTML wrapper around bugzilla to put
it in our website look and feel because bugzilla was never written to be
customized UI wise (yet another reason why bugzilla sucks). Scarab will have
that feature as a core portion of it.



View raw message