Re: [Tracker] Is tracker development dead?

[Jeff <web kiddo free fr>]

> Why not check out the latest from svn? I just built it yesterday.
I think that was the point. You don't get something as important as
Tracker tested efficiently if you don't release early and often.

Clinging to your code in SVN entitles you to 5% of the population
willing to test it. Releasing it has psychological benefits for you and
your users:
- you feel like you achieved milestones instead of endlessly committing
to SVN without seeing some hard, tangible achievement (I mean, having
releases early and often really helps motivation)
- your users trust you to make said releases instead of spending time
flaming Beagle without providing them means to see for themselves.
That's what I would call vapourware. I love the idea behind Tracker, but
this "any day now" has been going on for months
- your tarballs will most likely end up packaged for various distros.
Don't you seek world domination? You won't achieve it with 0.5.4, but I
know that 0.6 is a lot better and might help you with the conquering
plans.
- I don't want this project to look like the gaim 2.0 series that took
forever to come out. And now that those folks changed the name to
Pidgin, interestingly enough, they seem on a frenzy to make a small
release every three weeks. This is good.

> I'd like to thank the folks working on Tracker for providing such a
> nice alternative to beagle.
Don't misunderstand me: I love this project and I want it to grow tall
and fast. This is why I'm siding with the original poster here.

Providing "such a nice alternative to Beagle" currently means "providing
an SVN trunk version that is really good, and outdated releases that
make all the project look like crap".