Re: [pgsql-www] Collaboration Tool Proposal - Mailing list pgsql-hackers
From | Andrew Dunstan |
---|---|
Subject | Re: [pgsql-www] Collaboration Tool Proposal |
Date | |
Msg-id | 403FE522.20502@dunslane.net Whole thread Raw |
In response to | Re: [pgsql-www] Collaboration Tool Proposal ("Bort, Paul" <pbort@tmwsystems.com>) |
List | pgsql-hackers |
Bort, Paul wrote: >>My apologies, then! I was operating off of the statements >>of others, and the >>fact that the only RT impelementations I've used were running >>on MySQL. So, >>questions: >> >>1) can you compare/contrast RT vs. BZ vs. Simplified >>bug-tracking, like >>GForge? >> >> > >I've used Bugzilla for searching for FOP issues, and a couple other places, >and I find the RT search much more obvious. I can get what I want out of >Bugzilla, but usually by creating a really broad search and sifting entries >one at a time for likely candidates. The fact that an RT search is iterative >is much more obvious, because the bottom of the search page lists all of the >current criteria, and the box for adding new ones. Add or remove, and re-run >the search. I like the trick that it does with mutually exclusive >conditions: It assumes an 'or' between them. (eg, all tickets that are in >state 'open' or 'closed'.) > >OTOH, Bugzilla tracks a whole pile more fields by default. I've taken to >putting version numbers in the ticket subject in RT because for the small >project here, it's easier than learning how to add a version field. (I >haven't tried adding my own fields.) > >Both handle attachments and comments sanely. I don't know if BZ has an >e-mail interface, but the one in RT has filled the basic needs here. (We >haven't pushed the limits of the e-mail part.) > >I have never tried to install BZ. RT's install (RedHat 8.0, PostgreSQL 7.2.4 >from RPMs) was straightforward once all the Perl modules were up to date. >(All of the needed modules were available from CPAN.) > >I don't recall using any simplified bug tracking on-line, except maybe at >ImageMagick.com, which seems to be more a forum or mailing list search, with >no real tracking fields. > > > >>2) What help, if any, would we be able to get in supporting >>RT from the RT >>community? >> >> > >I'm afraid I have no idea what or where the larger RT community is. I know >there's commercial support available from the author (whom I have no contact >with), and I found the answers to my (self-created) problems during setup >using Google. I found RT because of a (don't ban me, please ;-) discussion >on SlashDot. (http://slashdot.org/article.pl?sid=03/10/06/1854211) There >were a large number of proponents of RT there; their posts claimed years of >use at many sites. > [some stuff about RT] FWIW there's a good directory of bug tracking systems on google: http://directory.google.com/Top/Computers/Software/Configuration_Management/Bug_Tracking/Free/ I have looked at RT briefly today, and its technology appears to be sound. (Just the fact that it will run under both mod_perl and FastCGI means that a lot of common insanity is missing - these environments are very good at blowing up with badly written software. They are also far more scalable than standard CGI web environments, so we would be less likely to have performance issues.) > >I would be happy to lend my meager talents to setting it up for a trial, if >that's where the group decides to go. > ditto > >But Josh made a good point off-list: are we trying to solve the problem of >ticket/bug tracking, or community/collaboration in general? > The discussion arose in the context of an alternative to the rather simple bug tracking system that comes with GForge. The biggest issue with any replacement would be to plug it in successfully. If too much glue is needed, we haven't really made an advance on the current GBorg code base. cheers andrew
pgsql-hackers by date: