We now consider 02-Jan-15, so far only placeholder in task #7804, now target
for S2_6 branching.
The idea is the same as we had with early S2_5 branching. It's not necessary
to stop all feature development for 2.6 as soon as it branches, but
the difference
between early S2_6 and trunk is only that more random features will go
to trunk only.
Any feature already under development can, and should, be finished in
the branch too,
and even new features can be added if there's a good reason to do so. With these
guidelines branching is not necessarily blocked even if 2.5.0 has not
been released
yet that date. S2_5 was branched several months before 2.4.0 was released.

All that said, I hope people already consider postponing start of
major (in terms of
amount of work) new features to 3.0 in December. We don't want to
spend years in stable
branch finishing unnecessary features that just happened to be started
at the time of
the branching.
There's still over four months to branching, so you have time to do big changes
before that, but you shouldn't postpone start for a long.

I've been reading through commit logs since S2_5 was branched to
create first draft of
2.6 feature list. I'm still about 1500 commits (in terms of revision
number, so that count
includes commits to other branches) in the past:


As for the features I'm myself working on (replies to this email from
others telling where
their features are going would be welcome), here's short summary:

- Extras: Mostly ready, bugfixing and some missing details still to
do. I *don't* currently
          have plans to adjust UI to support all kinds of rulesets in 2.6
          Whether both bases and roads remain as separate ruleset
items, or get completely merged
          to main extra item, is still a bit undecided in 2.6 timeframe.

- Sdl2-client: Needs a lot of work, but I hope we can release it as
supported client in 2.6 already,
               and in 3.0 to drop old sdl-client completely.

- freeciv-ruledit: I don't yet know what I expect to achieve before
2.6 release. Maybe we
                   keep it experimental in 2.6, maybe we can get it
usable for some special
                   use-case. It will be very limited in 2.6 in any case.

- lua-5.3: I'm following development of lua-5.3 itself, trying to keep
us well prepared to take
           it into use when it's out.

- Styles refactoring: (National/City/Music) I'm not aware of any more
changes needed

- Music system: Minor adjustments needed, but the feature is mostly ready

- gtk3-client quality: Improving in all branches. Mostly I've been
fixing bugs as they have
                       been causing trouble to me. I'm using
gtk3-client as my primary client.

- Threaded AI: Despite there still being four months before planned
branching, I probably
               leave changes other than those related to autosettler
functionality to 3.0.
               For 2.6 I hope to make autosettlers to take advantage
of the freedom to use more
               CPU cycles by analyzing potential tile improvements in
more detail

- Animals: I've added Lions to custom variant2 ruleset, and from there
we hopefully get
           experience guiding us in further development of the feature.
           Nothing planned at the moment.

- Achievements: Simple additions of new achievement types ongoing.
This could use the
                "customizable freeciv engine strings" work I've been
planning for years, so
                I might still start that for 2.6.

- Culture: Main part missing is lua scripting changes

- Alien ruleset: Fixes are made when ever there is an test game. I'm
now quite set to the
                 idea that alien ruleset won't be part of 2.6
distribution, but will debut
                 in 3.0.

 - ML

Freeciv-dev mailing list

Reply via email to