DESIGN

This document contains some misc. thoughts about Freeciv.

The language:
-------------
Freeciv is written in C as opposed to C++ or Java. We settled for C,
as we wanted the game to be easy to compile/run on as many unix systems
as possible. Unless we had sticked to pure ARM C++(and even then), we
couldn't have archived the same degree of portability. We could have 
limited us to g++, but decided against it in favor of more platform
support. We decided against Java, mostly because of speed considerations.
Today that still seems to be the right decision, as Java+JIT is still
quite rare on unix platforms. Also the AWT in JDK 1.0 doesn't look ready
for real developing yet(as JDK 1.1 - still in beta - is a proof of). 
As a sidenote, as the project grew in size, we very often wished we
could use OO techniques, as this game really could benefit from an 
OO design.

The widget set:
---------------
We decided to use the athena widget set, as opposed to Motif or one of
the free available widget sets. Motif was not an option as there's no
free implementation yet(Lesstif is getting close though), and none of
the free ones seems to be sufficient established/widespread. That said
Xaw starts showing it's age(didn't it always?) and is truely a pain to 
work with. On top of that it doesn't even look good.

Civ as a multiplayer game:
--------------------------
Although civilization has been available as a multiplayer game for
some time now(civnet, openciv), one can still question how fit the
game really is for multiplaying. At least one of the freeciv developers 
tends to become quite bored late in the game, when the turn length 
never goes below 3 minutes, because some players keeps fiddeling with
their cities. Ofcause they refuse to decrease the timeout. One issue
concerning the gameplay is what to do when one player is loosing. Should
she keep playing for another 3 hours, knowing the game is lost, only to
amuse the other players? Most of us would in the same position just quit
the game in a single player game against the computer AI. Anyway all
this is just a matter of taste, and this section does not represent the
official attitude of the freeciv team ;)

Simultaneous moving:
--------------------
Currently Freeciv only supports simultaneous moving. One nice effect of
this is that the overall game length is much smaller. However some times
it results in ridiculous action situation at the start of a turn, when
for example two attack units is standing close. We do intend to fix this
in a later release. The best idea we got so far is to allow 
players to issue goto orders for units with 0 movepoints(they can't get
focus as of v1.0). When a new turn begins, the server will randomly 
select units with a goto order, and move the unit. In the case of the
two attack units, this would give them both equal probability for
striking first. Another solution would be to temporary play with
synchronous moving when such situations occurs. We'd rather avoid that
solution, as it increases the game length. In case you have another 
solution then please let us know.

The future:
-----------
No question that freeciv 1.0 still contains a range of bugs and missing
features. Some problems are rooted rather deep in the design of the game,
and fixing them is not trivial. With the release of 1.0, we get the time
and chance to fix some of these and clean up the code. We'll also implement 
more report dialogs, and rewrite some the the older ones.
As for AI we already got a (rather) experimental version running, and we'll
keep improving on it. As soon as it is even remotely interesting for the
players, we'll make a beta release.

best regards
  the freeciv team