Re: Anyone on this list?

[Steve Baker <sjbaker1@airmail.net>]

Xarvh Admin wrote:
> Jan Ekholm:

> I'm working on a strategy game, and i found no one seriously interested in 
> helping me, or even just telling me what's wrong with it...

It's hard to get *ANY* new project started from cold.  Sometimes projects
start from a grass-roots need - most times one person just starts something.

* In the first case, the peril is that the whole project never gets going
   because people just talk about it and write design documents without
   actually getting any code together.

* In the second case, it's very hard to get other people to sign on to your
   personal vision of how things should be.

In both cases the answer is to write code more or less as a solo activity
and get the project to a stage where it's actually useful for something.

In the case of projects with lots of people already on board, this focusses
discussion and activity around a concrete task - in the case of solo projects,
you stand a much better chance of recruiting more people if you have something
more or less working so that they can see what they are getting into.

Either way, the bottom line is that you have to be able to write most of the
code yourself.

> Even worse, all the people wich swore to be interested in it left me just 
> once i relied on them... quite the same thing happened to Arianne....

Don't be discouraged.

OpenSource projects in general have a spectacularly high failure rate.

I've started/run eight web-based projects over the past few years.

* Toobular was a project that I lost enthusiasm for before anything useful
   was ever completed.  It looked good and was working as
   designed - but the gameplay was crap - I gave it up without
   wasting too much time on it.

* PrettyPoly generated a ton of great code and masses of enthusiasm
   then just fizzled out.  I still don't understand why - the
   project was doing well - I got too busy at work to contribute - the
   other half dozen developers lost interest at about the same time and
   now it's dead.

* freeglut is a working product (all written by one person) but there has
   been very little enthusiasm for doing anything more with it.  This
   project isn't a failure - it just doesn't *NEED* a lot of work and
   very few people use it.

* LodeStone generated a ton of enthusiasm - but nothing but talk ever
   came out of it - so it fizzled.  If just one person had contributed
   working code, I think it would have worked out OK...but I was (again)
   too busy to write code for it.

* AgTools died after I passed the baton to another maintainer - because
   it was based around the failed Agenda palmtop hardware.  That wasn't
   our fault - but you can argue the project should never have been started
   in the first place.

* Both TuxKart and TuxAQFH are what I'd call successful - but still solo
   projects.  They are good enough to get 5 star ratings and to make it
   into many Linux distro's - both have had download counts in the hundreds
   of thousands.  The only help I ever got for either one was a contribution
   of some original music to TuxKart - and some minimal help in getting the
   games to run on other OS's.

* PLIB is wildly successful, has a large developer and user community and
   a fairly active mailing list - it's the most successful project I ever
   worked on.

I'd rate this at maybe a 30% success rate...but if you count the projects
with large communities around them - maybe I have a 15% success rate.

I get the impression that this kind of track record is par for the course.

> LGDC is quite dead (no update since June)... there's no community around 
> linux gaming!

I don't think that's true - it's just not happening at LGDC.

> The only way i could play under linux was StarCraft + Winex...
> I'd tried gladly Pingus (i'll beta-test it) and lots of other games, but how 
> could i know about it?

You need to read HappyPenguin (aka Linux Game Tome) twice a week.  It used to
be just an archive of game titles - but it's grown of late to include a slashdot-like
community.  There is a fair amount of traffic there these days...mostly from
games *PLAYERS* - not from *DEVELOPERS*.

> There is no community, i feel quite alone and i feel that what i'm doing 
> interests none... anyone else feeling this?

Well, sometimes.

We are unlikely to want to talk specifically about your game here - but problems
of a general nature would certainly engender interest and lots of advice.

HOWEVER:

Games *are* personal visions.  I can't tell you the number of times I've seen
wildly enthusiastic people with ideas for games who seem to think that
I'll just *obviously* want to jump in and help them.  Well, I'm sorry but I
have some great ideas of my own.

Ideas for games come very, very, cheap.  Code for games is harder work. Artwork
and Music is harder still.

It follows that people will not join your project just because it's a good idea.
They *may* join if they see a great code-base emerging - but to REALLY attract
developers, they need to see the whole thing basically working - playable, etc.

Take a look on Sourceforge's Game's Foundry.

      Board Games (307 projects)
      First Person Shooters (420 projects)
      Multi-User Dungeons (MUD) (630 projects)
      Puzzle Games (217 projects)
      Real Time Strategy (423 projects)
      Role-Playing (1049 projects)
      Side-Scrolling/Arcade Games (280 projects)
      Simulation (605 projects)
      Turn Based Strategy (535 projects)

That's 4500 games projects...and those are only the ones on SourceForge.

What percentage of those turn into actual playable games?

Well, SuSE Linux (which happens to be the distro I use) packs in as many
OpenSource projects as it can.  I think a reasonable measure of success for
a game is "Did it make it onto the SuSE distro".  You may have another criteria...
but this is one I can measure.

There are just 112 games released with SuSE.

You could consider the games registered with HappyPenguin.  There are
926 games in the archive...so right away, only ~20% of games are ever
finished to the point where it's worth announcing them on HappyPenguin.

But of those 926 games, 400 are 'unrated' - ie so bad that people didn't
even bother to rate them. only 218 got 4 or 5 stars on the rating system.

So, you have a 3% chance of "success" - according to this 'SuSE-acceptance'
criterion and a 5% chance on the '4/5 star rating' test.

> I'll continue developing Xarvh myself, but the overall situation seems quite 
> sad...

It's a hard call.  Do you carry on with a project as a solo activity in the
hope that you'll get fame and glory when it's finished - or do you abandon
it because nobody else thought it was a great thing and start something new?

It's a difficult decision.

I've decided that there is little to be gained by writing GAMES as a community
project - all of my new games are developed by me and my son without external
input.

HOWEVER, there is a lot to be gained by writing SUPPORT TOOLS for games writers
in a community setting.  I put as much of the code from my solo game projects
into my shared library code project (PLIB) - so that the ratio of game-specific
code to shared-library-code is as small as possible.

My first Linux game (TuxAQFH) was about 60% game-specific code and 40% library
code.  My second game (TuxKart) is about 10% game-specific and 90% library.  My
next game (to be announced) will be about the same ratio.

This means that I get lots of help in writing, debugging and maintaining the
vast majority of the code (and that's mostly the boring parts) - and 100% control
over the places where the creativity lives.

---------------------------- Steve Baker -------------------------
HomeEmail: <sjbaker1@airmail.net>    WorkEmail: <sjbaker@link.com>
HomePage : http://web2.airmail.net/sjbaker1
Projects : http://plib.sf.net    http://tuxaqfh.sf.net
            http://tuxkart.sf.net http://prettypoly.sf.net