Re: [Adonthell-general] Adonthell French Meetic Report - GUI, SDL, Clanlib

[Alexandre Courbot]

> ClanLib is not bad, but it is not really the one true thing that makes
> you happy till the end of all times. It has its flaws like any other
> library and is known for not having the most stable API on this
> planet. ClanLib-0.7 should be quite a bit better in many aspects, but
> I havn't really used that one that much. There is also the problem of
> lacking X11 support in ClanLib-0.7, it only has the OpenGL stuff up
> and running at the moment. So in the end, using ClanLib might be ok,
> but I wouldn't try it again without an abstraction layer inbetween.

Clan 0.7 is still a development version, I suppose it is going to
support other backends soon - at least I hope so! :) I guess it must
have its own drawbacks, but I wonder what we are more doing in
Adonthell: the game, or the abstraction layers. Hence my doubts about
the current direction. Having a library that takes care of everything
is kinda utopic, however.

> About the GUI, I don't think that clanGUI could be of that much help,
> its a nice piece of code, but after all its build for
> mouse-interaction and I don't think that it is really useable with a
> gamepad or keyboard. And as long as a GUI shouldn't use a zillion
> different widges its not that difficult to develop and own, which
> might serve the purpose better than a generic one.

Problem is, the GUI is also supposed to be used for map-making tools and
others. So it has to provide a minimum of stuff. But as you pointed,
something controllable via keyboard is nearly mandatory - at least for
the game itself, it doesn't matter much if the tools require a mouse
after all, and in terms of intuitiveness it is probably better.

Actually, having a GUI that we could map on top of our current layer
would be great. But unfortunately, it seems not to be an option. And
hacking an already-existing GUI that is too much dependant of a library 
is not necessarily better than writing one from scratch, I fear.

The issue is, I think it'd be better to concentrate the efforts more on
actual game development and less on the alternative aspects - for this
we can use what already exist. Kinda utopic I know, but I'd like 0.4 to
be released, say, before 2005! ;)

Alex.
-- 
http://www.gnurou.org