Re: What to do with Xconq

[Stan Shebs <shebs at shebs dot cnchost dot com>]

"A. Rick Anderson" wrote:

> a) XConq used to be on the standard linux distribution.  It's no longer
> there.  Years ago, I bought the Yddrasyl (sp?) Linux distribution just to
> get xconq (it wasn't there.  potential copyright issues with Apple caused
> them to drop it from the distribution.)

In retrospect, that was unduly cautious.  While the FSF gets copyright
assignments for GNU tool bits, Linux proper is a real hodge-podge of
ownerships, and nobody seems to worry about it much.

There isn't actually a standard Linux distribution anymore.  Red Hat trims
theirs to fit on one CD, so it has almost no games, while Debian includes
Xconq in its large game collection.

> b) There should be a game-players-only turnkey install with the best of the
> game-moduels available immediately upon startup.

Amen.

> I played XConq back in the 5x days and on a networked AIX box, it was great
> fun.  Conceptually, I love the open-source nature of xconq.  There have
> been any number of times that I downloaded everything.  I get all set up to
> work on the code, but I just can't bring myself to get excited about
> leaping backward in time and going back to hacking around with 'C'
> /TCL/pseudo-Lisp code.  I've seen that movie and it just holds absolutely
> no appeal for me.  It beats Cobal and BASIC, but not by much.
>
> I was very excited a few months back when you seemed to be considering
> doing a re-write in Java.  I can't speak for everyone, but I know that I
> would be more then willing to spend a lot of time on developing my own
> enhancements/refinements if it was written in a language that I am
> currently interested in.

For me, the primary purpose of the project is to build a working game.
It's true that I've introduced ideas for the sake of experimentation
(isometric displays for instance), but I've never introduced a technology
just because it seemed interesting.  The Lisp syntax was introduced to get
around limitations of the Forth syntax, which was introduced as a simple
way to get run-time customization.  Tcl/tk was a way to improve interface
building over Xaw, while enabling a Windows port; I looked at a bunch of
alternatives before settling on it.

So every technology idea needs to be evaluated in terms of what it does
for making a good strategy game.  A change of implementation language,
or a reimplementation in a new language, is a huge step and a lot of
work to make happen, so it needs a correspondingly high payoff.

I've looked at Java twice now.  While the language has a lot of good
features - it might even be called "Lisp done right" - I don't see that
the payoff is high enough to make it worthwhile to convert.  For instance,
Xconq has performance-critical areas; not for the standard game on a small
map, but when scaling up to larger games.  How much slower will a Java
version be?  One way to test this is to build small proofs-of-concept;
I did this for tcl/tk, there's one (not in cvs) for gtk.

Will being in Java attract lots of developers?  I don't know.  I do know
that the most active open-source game projects - FreeCiv, NetHack, etc
are all in C, and I don't know of any large Java game projects.  What
I *have* found for Java is a couple of dead strategy game projects that
seemed to start out well, but then broke down - not clear whether it was
the language or the developer, but not encouraging in any case.

So as I said to you in our previous discussion, what I think we need here
is a prototype testbed.  Make a program that does nothing but put up a
hex map and scroll it around; steal algorithms from kernel/ui.c, images
from the library.  It should take about a weekend for somebody who knows
Java, if it takes longer than a week, that's a danger sign.  If the
result runs great, then it's worth expanding on; if it's slow, it's small
enough to be tuned until it does run fast enough.  The prototype can also
be carried around to other machines, to check that performance is sufficient
on all platforms.

> A lot of open-source folks are late-night college hackers.  A Java
> implementation is the Y2K equivalent of 'C' back in the eighties.  To get
> the best job, you need experience.  Compare the resume-enhancing effect and
> marketable skills that a student gets from contributing to a major Java
> module of a large-scale, distributed, networked project, versus being able
> to say that they added a "game module" written in a proprietary lisp or
> that they cleaned up the TCL-based interface.

To be brutally honest, I don't see that any of the major contributors to the
big open-source projects work on them to enhance their resumes - they do it
because they're interested in the project.

> I'd suggest a survey.  Ask folks:
> o - How many would rather continue work on C/TCL/Lisp  ... or how many
> would be willing to help if a different technology, such as Java was used
> instead.
> o - How much time they might be willing to spend on it.
> o - Who would be willing to develop game levels
> o - Who would be willing to develop graphical "level-editors"
> etc.

I'd certainly like to see the results!

> XConq is very, very cool!!!  Stan, don't let it die just because us
> old-farts are going through a mid-life crisis :-)

:-)  As long as nobody else has an open-source game with a Peloponnesian
scenario, I'm going to be working on Xconq...

Stan