Re: Gnometris

[Alan Horkan <horkana maths tcd ie>]

On Thu, 30 Nov 2006, keith preston wrote:

> Date: Thu, 30 Nov 2006 15:17:25 -0600
> From: keith preston <keithpre gmail com>
> To: games-list gnome org
> Subject: Gnometris
>
> So I, like everyone else that e-mails list, want to add a game to
> gnome-games.   Well I guess I really don't care, but I would like advice on
> what would be the guidelines for games in gnome games?  Here are some
> thoughts
>
> Simple to Play and Fun

Sometimes the difficulty is what makes a game fun but it is important to
make sure things are only difficult when it improves gameplay rather than
harming usability.  One example is forcing players to play levels in a
specific order, something I find very annoying.

> no additional dependencies (stick to GTK/glib)

That is certainly preferable but many Gnome applications carry around less
portable crud.  The trick is usually to make things like for example
gnome-vfs a compile time option and avoid hard requirements on anything
beyond Gtk/glib

> C/C++/ Python? / Mono (ok that may be a little far)

I dont think any of the current games are C++ but I dont think it would be
a big problem if they were.  So long as games are not dependant on a
single developer and there are at least two people empowered as
co-maintainers it is a lot easier to make sure patches get applied and
applications dont stagnate.  It is a poor sign when an application is part
of Gnome and doesn't have more than one maintainer.

Aisleriot use Guile (Scheme/Lisp) so we already have one interpreted
language and the prospect of adding more and having them become a
maintaince problem later weighs heavily on our minds.  Scheme seemed like
a good idea at the time.

Gnome at large has adopted Python in a big way, so it is near the top of
the list of choices.  If I recall correctly python is used by Gnome Soduku
one the new games added to Gnome Games.

> Good Graphics and Sound

When others say good graphics they mean eye-candy.  For Gnome Games it is
important to have an accessible high contrast theme for all games.
Scalable themes help too.

I'm not sure if there are high expectations in terms of sounds, would be
nice though.

> Network abilities.

I'm do not think there are high expectations in terms of networking
either.  The name (some acronym) escapes me but there have been efforts to
start using a standard networking framework.

> Documented and maintained

It is mildly annoying at times but I've a lot of respect for Callum
because he made sure documentation was provided before adding new any new
games to Aisleriot.  The well organised documentation made this relatively
easy.

Gnome games should unequivocally not be used as a dumping ground for games
no longer being developed.

> I say this because I'm am a student and want to develop a GTK game.   I've
> fallen in love with Super Puzzle Fighter X (puyo puyo) in college because it
> is a greatly designed game.

KDE includes a puyo puyo game, and there may already be GTK versions of
this game out there you might be able to improve and add the advanced
features from Super Puzzle Fighter rather than starting from scratch.


Myself I've always loved Puzzle Bobble and the Open Source game Frozen
Bubble.  Last year I spent a lot of time playing a game called Zuma,
similar to an older game known as Puzz Loop and various others and do hope
to find an Open Source equivalent.
http://en.wikipedia.org/wiki/Zuma_%28computer_game%29

> http://www.sirlin.net/Features/feature_slipperyslope.htm   However my friend
> is moving away that I play it with, so it is our duty to continue puzzle
> fighting by obviously developing a network version of it for linux.   Yes I
> know there is already "Angry Drunken Dwarfs" but it doesn't do multi
> player.  I also want to do this as more of a "student" project / learning
> experiment and I've become a Gnome kind of guy lately so I want to do
> something of Gnome Caliber and that's why this e-mail is being sent.  What
> is a Gnome Caliber Game?

Those two things sound in conflict.  "student project" doesn't give much
confidence for long term maintaince which is the really hard part of Gnome
games.  If you have read the mythical man month or any other good software
engineering book they will say ongoing maintaince is 90% of the work (and
I tend to believe those who say it is closer to 99%).

Each of the problem areas you oultined above are actually huge and take
years to get done properly.  Doing things in all the languages Gnome
supports means you simply cannot do it alone, so you need to plan to share
out work as much as possible from the beginning.  The job of a maintainer
is subtley different from being a developer and a lot of the time requires
more management than programming.

> http://mail.gnome.org/archives/games-list/2006-August/msg00033.html

> > There is a problem with removing this: It is one of the most popular
> > games and tetris is far from "embarassing simple". However, if the
> > code is unmaintainable, we have an issue.
>
> Our implementation of Tetris is embarassingly simple. If there is a
> replacement version of Tetris available, we could consider it for inclusion.

The simplicity of the game is not so important as the complexity of the
code and lack of a maintainer for that game specifically.

Jason has stepped up as Gnome Games maintainer and although I would
encourage him to be very slow to remove things but as maintainer it is his
decision.

> So how can I take two birds with one stone?   Well in my beginning design
> phases  I have been trying to design the game in C++ class glory and realize
> that it is similar in ways to a  Tetris Game.   I've been designing to
> abstract a "falling block" game and then eventually build a tetris variant
> and a puzzle fighter variant on top of it.  Since you have discussed looking
> for a replacement for gnometris, what features are you looking for in a
> replacement?

The best thing is to start on your project, and see how it goes.  By all
means ask for feedback and encouragement and the occasional review to keep
you on the right track.  As for being included in Gnome Games that is more
likely to happen for a game that is already mostly working.  If you reread
the discussions about adding games over the past few months you should get
a better idea of the kidn of quality we all hope a game would be at before
being included in Gnome games.


Sincerely

Alan Horkan

http://advogato.org/person/AlanHorkan/
http://alanhorkan.livejournal.com/