On 13 February 2011 20:26, Jacob Nevins
> I guess that the move to Cairo is probably the biggest deal for us, but
> I haven't looked into it.
Yeah. And now I remembered again how one of the last things we ever
heard from mbook before her disappearance was something like "I'm
doing final testing of new cairo based drawing engine. Will submit it
in a couple of days."
> Even having looked through the referenced article, I'm not sure if we
> can get away with compile-time conditionalisation or whether we'll need
> to fork the client :/
Even if we could do with conditional compilation, it might make sense
to have them as separate clients. For one, it would allow gtk3 based
client to be take advantage of new features in gtk3 without worrying
about how well it fits common gtk-client code. I'm actually quite sure
that having them as separate clients (that one can even compile at the
same time with '--enable-client=gtk3,gtk2') is correct way to go. Of
course that means that any bugfix to their common code must be done
for two separate clients. gtk3-client should start out as copy of
gtk2-client. Initially it would not even compile. For the sake of the
maintenance effort of separate clients we should make the initial
copying in sensible time, not before someone is actually going to work
on gtk3-client development.
>> What about schedule? Do we need to change something for 2.3, or can
>> these things wait until 2.4?
> IMO, given our new sleek release cycle :) I doubt we need to worry about
> 2.3, but we should probably start thinking about it for 2.4.
Yeah, I think that it's definitely too late to do anything for 2.3,
and even with our current release cycle (seems to be about 1.5 years
between 2.2 and 2.3) we need this before 2.5 (which would be more than
3 years away). Doesn't leave many versions to choose from in between
We are already understaffed in clients department. I consider
gtk3-client much higher priority than Qt-client, so personally I will
cut down Qt-client development. This doesn't mean someone else
couldn't take active role in its development.
I have already compiled gtk3 (and obviously all its dependencies) for
myself so I can get my hands dirty already.
Freeciv-dev mailing list