[Freeciv-Dev] AI code organization

["Brandon J. Van Every" <vanevery@xxxxxxxxxxxxxxxxxxx>]

From: Mark Metson [mailto:markm@xxxxxxxxxxxx]
>
> Apparently we are now much further along toard a
> code base that will permit some actual improvement in the AI.

Actually at first glance I believe that.  The small amount of code I've
looked at so far looks well-organized and not ponderous.  Easily
Pythonized.  (Uuuuuh, can't remember if it was *AI* code. :-)  But I
haven't looked at the Big Enchilada, and really now there's no point.
Next step is to rip the lid off of the client / server main()s and just
start Pythonizing it.  Then I'll find out exactly what's hairy and what
isn't.

If I live through the experience, there will be a new project based on
Freeciv starter code.  But one project goal will be to discourage the
writing of new C code, to instead use Python as much as possible.  C
should only be for what has to be fast, and for what's legacy + stable +
not worth changing.

I don't know what I can and can't retain, like clients and platforms and
all.  I'm on Windows and will not be doing anything on Linux, so a
Linuxer would have to join me if that's going to work.  Nor do I know
what I can / will track in Freeciv's development.  As I said, the goal
will be to wean the project of C code.  In 2003, C is a bad habit.

> What might be a really worthwhile challenge for your awesome
> Python-A.I.
> skills might be to code a python CLIENT with Client-Side AI,
> something
> that could be run as a daemon one copy per A.I. opponent.

Yes, agreed... but my first priorities are on getting the project
running, getting it managed properly, and changing the game design.  Civ
II cloning has to go.  I won't support clones of any game in my project.
I have no problem with games that are basically similar to commercial
games, so long as they are different games as per copyright law.  For
instance, Call To Power II is basically similar to the Civ games, it's a
knockoff or "me too" title.  But legally speaking, it's certainly not a
clone.  And it does a few things here and there that are better than
Civ.  For instance, it had larger increasing city radii before Civ III
did.

And that, people, is the point of the project.  Innovation in game
design.  Improvement, not stagnation.  Prototyping and trying things
out.  Not talking / Gatekeeping about whether people should be allowed
to try them out.

> But actually there are also a bunch of other little things
> that have been
> happening, such as various support toward Lord of the Rings types of
> scenarios. Such as "game loss units". Also, even if some seem
> to you to
> indicate a game design engine isn't what is being built,
> there is more and
> more flxibility as time goes on, so for many purposes it
> looks like it
> should be at least a scenario design engine.

I guess I haven't really seen it.  I've seen different flags for the
Mordors and the Dunedains, but that's all.


Cheers,                         www.indiegamedesign.com
Brandon Van Every               Seattle, WA

20% of the world is real.
80% is gobbledygook we make up inside our own heads.