Re: Policy for accepting new games in gnome-games

[Alan Horkan <horkana maths tcd ie>]

On Fri, 25 Aug 2006, Jason D. Clinton wrote:

> Date: Fri, 25 Aug 2006 17:19:06 -0500
> From: Jason D. Clinton <me jasonclinton com>
> To: "[ISO-8859-1] Andreas Røsdal" <andrearo pvv ntnu no>
> Cc: games-list gnome org
> Subject: Re: Policy for accepting new games in gnome-games
>
> Andreas Røsdal wrote:
> > I therefore don't think that we should begin accepting new games, but
> > rather focus on improving the current ones already in gnome-games. We
> > don't have an objective way of selecting new games, or deciding which
> > games to replace.

> I think that there are certainly some games which could be removed and
> indeed even some push from distros to do so (see Callum's email from
> 2006-05-30). And given that those games are removed, we could then
> consider adding a strong candidate for inclusion.

"I never had the guts to cull the list of games. Although I was pushed by
many distributors to do so. "
http://mail.gnome.org/archives/games-list/2006-May/msg00068.html

I don't know why I didn't respond to that at the time but during his
tenure Callum did replace Same Gnome, and remove a game or two over legal
issues (which I cannot for the life of me find reference to in the archive
and I've scanned back over 18 months already).

Also the work which made it easier to build only parts of gnome-games was
probably appreciated by distributions.

> This will not be popular unless the games replacing them have a wide
> appeal, low learning curve and high quality.

If an application is unmaintainable there is only really one way (maybe
one and a half ways) to argue against it being removed and that is to
volunteer to maintain it (or argue it doesn't actually need much
maintaince).  Adding any application which did not have *at least* one
active maintainer or ridiculously high quality code only risks
history repeating itself.  Accepting new applications could work but it
will require a lot of care and attention and it is but one possibility
among many which could draw more attention to Gnome Games.  [1]

Gnome Games has always struck me as a remarkable example of how important
longterm maintaince is and what a large part of the development cycle it
takes up.  It is hard to believe books like Mythical Man month which say
maintaince is more than 90% of the work but looking at Gnome Games it
looks more like 99% and the truth is crystal clear.

The ongoing work to document and games into all the languages Gnome
supports is a massive task (and the occasional cultural issue like red
light, green light not making sense to everyone).  Porting to Gtk 2 and
updating to newer standards and improving coding is a lot of work, not to
mention the efforts required to make games fully scalable.  Recently we
are starting to see more Networked games.  There are other questions of
keyboard only accessibility, not to mention the more complicated issues of
real accessibility I only begin to understand.  I rarely turn on sounds
but multimedia is another question we could delve into.

I'm interested by usability and when done right it is something you should
hardly notice, so all the little tweaks and adjustments that have gone
into gnome games to make it nice and comfortable are hardly noticable.
perhaps more noticable is how clean and compact the applications all are
and you could nearly run them on a smalls screen portable device without
modification[2].

Aisleriot is a massive project by itself and is a good example of the
difficulty in maintaining more languages and platforms.  Aisleriot has
less than a hundred games compared to other Solitaire games claiming
(somewhat dubiously) to have hundreds of variations.  I'm impressed how
piece by piece something quite as huge as Gnome games has come together
and held the interest of enough talented developers to keep it alive and
well.

Point is there are loads of big ideas and mabye talking about them more
and putting together something resembling a Roadmap could encourage others
to get interested again.

> I suspect that Jeff Waugh's recent "swimming upstream" has something to

Speaking of Jeff Waugh, dare I mention 10x10 and questions of expanding
market share?  Porting to windows (or React OS for the dedicated) or any
other platforms certainly opens up a larger audience if it is a goal
people are interested in following.

> So for now, I think that following games are strong candidates for removal:
>
> * Gnometris: almost unmaintainable; somewhat embarrassingly simple

Again maintainability is a very strong reason to remove things and exactly
the reason why adding any new game comes with a lot of risk.  Removing it
later sounds simple but it results in more discussions like these.  The
longer any game is in Gnome Games the more the wider community puts into
it, in the many ways I outlined above.

> * Robots: low quality graphics; doesn't scale to screen size; not really
> our target audience, really targets the hacker/TTY folks who really
> probably aren't all that interested in playing a GTK+ version anyways.

Similarly some of the games falling into the "arcade" category (usually
games using SDL) have the least to benefit from a GTK versions and dont
fit so well with the current collection of more card and puzzle oriented
games.

> Games which are probably popular (will need to take survey to know for
> sure):

I do hope this message is received as it was intended, a lot of
encouragement for all the possibilities of Gnome Games.

Sincerely

Alan Horkan

Inkscape http://inkscape.org
Abiword http://www.abisource.com
Open Clip Art http://OpenClipArt.org

Alan's Diary http://advogato.org/person/AlanHorkan/


[1] I often wonder if I should write about these things in my journal/blog
too and perhaps a passing developer will be encourage to take a look at
gnome-games.

[2] Handheld gnome games
http://mail.gnome.org/archives/games-list/2005-April/msg00022.html

[3] There is no 3 but I realise I was reaching for similarities in various
places.  ;)