Hi,

Raphaël Quinet writes:

I suggest to create that branch immediately after the 1.3.21 release. What do you think?

We create a branch immidiately after 2.0 is released. We don't have the resources to handle more branches. If you branch now (or after 1.3.21), we will never get a 2.0 release out.

Sven

On Wed, Oct 01, 2003 at 07:39:58PM +0200, Sven Neumann wrote:

Hi,

Raphaël Quinet writes:

I suggest to create that branch immediately after the 1.3.21 release. What do you think?

We create a branch immidiately after 2.0 is released. We don't have the resources to handle more branches. If you branch now (or after 1.3.21), we will never get a 2.0 release out.

I second this. If we create a 2.0 branch too soo, I'm afraid people will think "ok, this branch is for bugfixes only, let's play with HEAD instead". This way coders will be much more interested in adding new features in HEAD than in finding bugs into the 2.0 branch. So please delay the 2.0 branch until 2.0 is out.

Regards,

DindinX

Hi,

Sven Neumann wrote:

Raphaël Quinet writes:

I suggest to create that branch immediately after the 1.3.21 release.

We create a branch immidiately after 2.0 is released. We don't have the resources to handle more branches. If you branch now (or after 1.3.21), we will never get a 2.0 release out.

Agreed. Pre-release branches might work on other projects with more developer resources. The reality is we want everyone who has time to work on bug-fixes to have a 2.0 before Christmas (that's still the goal, and we're still on target for that).

Cheers, Dave.

Raphaël Quinet writes:

According to the roadmap proposed a few weeks ago, we should be in feature freeze and we should have released 2.0rc1 by now. Also, the old web site was supposed to be replaced by the new one in order to be prepared to support the new release.

None of this has happened: new features are still being implemented (e.g., the nice tool presets added yesterday),

This is not a new feature, persistant tool options have been there for a long time. The functionality of the "save" and "load" buttons just needed to do something actually useful, that's what I did, finishing an existing feature to avoid shipping loose end.

This will definitely happen some more times before 2.0 is out.

we are planning for
1.3.21 soon (not ready for 2.0rc1 yet)

We cannot make a 2.0 release candidate until libgimp is API frozen. Apart from that, current CVS is more stable and more consistent than it was at the time we released 1.2, so actually we could release 2.0 at any time.

and the web site hasn't been
moved.

You know my opinion about this.

This is not a real problem because this is usual in Free Software projects and there are be good reasons for these delays: a 1.3.21 release would be really useful to test the latest bug fixes.

I totally agree and still wonder why we still have that old website. While we have good reasons to delay 2.0 release candidates, I fail to see a single reason for further web site moving delays.

But if we still want to keep the goal of releasing 2.0 before next year,

As I said above, I don't see a reason why we should not be able to do this.

then it would be useful to revive the idea of the feature freeze. One way to do that would be to create a branch for 2.0 (like what was done for gimp-1-2) and say that only bug fixes can be done on that branch. Being "forced" to move away from CVS HEAD would probably change the way we consider the code for 2.0, and discourage the addition of new features which may be very nice and useful, but go against the feature freeze.

Nothing went against the freeze. See above.

I suggest to create that branch immediately after the 1.3.21 release. What do you think?

No.

ciao,
--mitch

Michael Natterer wrote:

I totally agree and still wonder why we still have that old website. While we have good reasons to delay 2.0 release candidates, I fail to see a single reason for further web site moving delays.

I never use gimp-1.2 anymore. Consider the lack of good reasons to move it, perhaps.

carol