Game Development

[AaronSDG]

So how would one write a game for your platform? Is it open in that
way?

[Wesley Tanaka]

Hi Aaron,

I would like to reorganize and document to make this possible, but it
is not currently. Does SDG allow people to write games for it? How
many people take advantage of that feature?

[AaronSDG]

It is possible, but not trivial. SDG wasn't initially designed to
allow it. If one day I ever rewrite it, I will pay closer attention
to APIs. To date, out of the 113 games currently available, only 6
were written by (3) other people. I know there would have been a
handful more if the API was programming-language agnostic.

[Wesley Tanaka]

I'm curious, do you think that the time designing and documenting an
API is more or less than the time it takes to implement those handful
of games?

[AaronSDG]

In my case the API was not designed. There were a few people with
games they wanted on the site, and I didn't have time to do it
myself. I ended up describing how I plugged new games in to my
system. Designed from the ground up, I think it would be much
simpler. (SDG was a very organic thing that just kinda happened.
There was not much forethought.)

The time to describe the interface wasn't too bad. It took about as
long as it would take to code a straightforward game (a couple of
hours). Designing, if done from the beginning, I don't think would
take any more time. The big thing I ran in to was my API was not
language agnostic. It was Perl or nothing. There were a couple of
other people that would have contributed if this was not the case. My
planned rewrite for SDG will include making the game modules
completely separate from the rest of the system and therefore language
agnostic. Public APIs of course bring in the security concerns as
well though.

To answer your question more directly, I would not rewrite SDG just to
fix the API. I'm not convinced there would be enough demand to
warrant the many hours it will take to rewrite all that code. For me
the issue is speed. The system is currently a memory hog and very
slow. That I need to somehow fix to remain viable. If I could find a
way that didn't include recoding, I would just let it stand.

[Wesley Tanaka]

It sounds like I am in a similar position. I can describe how to
create a game (in python) if you or anyone else is interested in
adding one. But I'm also not convinced there is that much demand for
that at the moment.