[Trac] VERY COOL!!!

[Hendy Irawan]

Excellent!

I got Trac 0.8.1 working! (with Svn 1.1.4, so I did download them
all... even though I had to wait one complete full day... oh that's
the past anyway)
Thanks for the help... although I could should have read the Windows
Installation Guide... ;-) [I could've sworn I tried to read the
installation documentation first, I guess I have the frontpage
dilemma, if it ain't in the frontpage, I ain't gonna read it: gotta
remove this habit]

And it works FINE, SWELL, EXCELLENT!!! I've tried Mantis and
phpBugTracker in the past... Tried to install Bugzilla but never got
it working (remember, Windows here). Trac is *beautiful*, visually,
functionaly, and everything! It's SVN integration is bliss...
Everything works without a glitch... smooth as Hikaru Koto's skin...
pure diamond :-)

Aside from the hassle of installation (which isn't all that bad, once
you know what packages you need, how install each of them, and some
hefty bandwidth), everything is great. One little downside is that I
don't particularly like the Wiki, I like MediaWiki's style more
(CamelCase? come on guys this is 2005!!). But that doesn't really
matter, as they're only used for tickets anyway. I maintain my own
MediaWiki for the projects stuff.

Anyways, I'd want to ask: Do we really have to create one subversion
repository for each project? I currently have the tendency to cram
things up in one repository, then split them if I deem it necessary
(internal repository copy/moves are cheap and fast, free of hassle,
this isn't true with external repositories). Trac seems to treat each
repository as one whole project, so I end up treating the projects as
components, although this isn't factually true. I may be tempted to
split the repositories in the end due to how Trac works... But I need
to hear other opinions about this, the ups and downs etc.

Thanks a lot for making a cool project!!

P.S.: Microsoft Indonesia had just started a community project (open
source?) in .NET for bug tracking system. I suggested Trac as
inspiration. I hope they don't get mad at that (they're evangelists
anyway). I'm also currently skeptical about the success of the
project: I don't think it's easy to start a community project using a
top-down approach (the team members are geographically distributed,
working, etc.). My observations report to me that usually this goes by
one man (or two) piloting the project to somehow Alpha state, and then
the community will dive in progressively after enjoying the initial
versions... Anyways, I would also be interested if someone has
reports/case studies about open source/community projects with a
top-down approach and also geographically distributed.

-- 
Hendy Irawan
http://www.gauldong.net
http://dev.gauldong.net