I would suggest the following approach which I believe is practical / realistic:

1. Keep master-1.2.x going with fixes and small enhancements.
2. Release 1.3.x out of master, once this happens, retire master-1.2.x.  Use http://www.mantisbt.org/bugs/view.php?id=14088 to identify list of blocking issues.
3. Use master as the basis for future innovation.  Note that this will be easy since merge/cherry-picking between master/master-1.3.x should be easy.
4. Take inventory of improvements that we have on candidate next branches, and then port such features to master.  If we focus on one big improvement at a time, then we can port, stabilize and release.  Only port to the real master once it is complete and stable.

As usual, I don't like boiling the ocean, I never seen it work before.

So let's get v1.3 out and make our lives easier.

On Sat, Oct 27, 2012 at 7:58 PM, Damien Regad <damien.regad@merckgroup.com> wrote:
On 23.10.2012 08:05, Victor Boctor wrote:
> I'm confused about master-2.x and next was a long term branch.

"2.0.x" is Paul's own "next" branch. He's been working on that for many months
now, and I believe he would like this to be the new "master". At some point I
even thought he was going to fork (i.e. the mantisbt2 org on github).

I have tried on several occasions to get him to speak with dhx and agree on some
common ground and merging / rebasing changes so that we have only one "next"
branch (see IRC logs).

While there are interesting improvements in Paul's work (particularly the new DB
layer), it is a rather radical approach and I personally have several issues
with the branch as a whole
- it is "back-end" changes mostly, which do not bring any major user benefits
- it breaks plugin compatibility
- Paul has very strong opinions on what should be part of "2.0" or not, and has
therefore intentionally not included some commits to master, not to mention
dumped the soap api entirely
- it was forked from a very old master commit, and not everything that came
after has been cherry-picked (see above point)

Based on the above, and to directly answer Roland's points, I think:
- you did not miss anything that I'm aware of
- the mentioned tracker issues should definitely be reopened (again)
- it's way too early to have a '2.0.x' version created
- 'next' and 'master-2.0.x' should be merged

Like Robert, I focus mostly on fixing and sometimes improving 1.2.x because
that's what I use, but I'd be glad if we could release 1.3.0 soon (as I'm
getting tired of having to port commits all the time, especially considering
that some of them are not simple cherry-picks).

> master is close to being consumable with a bit work.

What do you mean exactly ?

There is quite a lot of open issues on the Roadmap, and a much shorter list in
http://www.mantisbt.org/bugs/view.php?id=14088

As things are, I think we need to have an agreement as a team on what comes next
(if I dare use this word), and then commitment on making it happen, otherwise
1.3.0 will remain "floating" indefinitely.

Damien



------------------------------------------------------------------------------
WINDOWS 8 is here.
Millions of people.  Your app in 30 days.
Visit The Windows 8 Center at Sourceforge for all your go to resources.
http://windows8center.sourceforge.net/
join-generation-app-and-make-money-coding-fast/
_______________________________________________
mantisbt-dev mailing list
mantisbt-dev@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mantisbt-dev