Re: svn 1.0 in 45 days

> In my opinion, subversion is practically in a freeze.

Is it? I've been mostly going by the dev list.

Aside from the discussion, I keep reading about new code going in or
expected to go in before 1.0.

So -- it doesn't look like a freeze to me.

I also read about self-hosting success, some 3rd party project and 3rd
party experiment success. It's already in a nearly stable,
demonstrably useful state.

It's also a large, complicated, multifaceted project. The importance
of this is that I think it is _too_ large to "perfect" for a 1.0
release (unless you want to play the Emacs joke and state at 0.xx
releases for the next 20 years).

It's also particularly weak in having client-side features and
easy-enough admin that it will, just on the face of its brilliance,
take over the world upon reaching 1.0. It's a great architecture.
It's a paradigm shifting tool because it enables applications built in
ways people have never used before. It has applications _within_ a
rev ctl. killer app. But it far more than 45 days away from being a
killer app itself in the fairly narrow and crowded source code
revision control space.

So I'm suggesting (admittedly, partly for selfish reasons having to do
with my own projects) is bottling what svn has already got, by the
shortest path possible. Document in release notes the features that
aren't done or the cases known to perform oddly; add in the admin faq.
Just fix bugs and port for a fixed time period and call that 1.0 --
then put together some app demos that show off its existing strengths.

> Um. Have you browsed the subversion issue tracker?

No. Can you please hand-hold me a little bit and give me a specific
URL?

-t

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sun Oct 13 17:53:31 2002