This initiative aims to improve the status quo of developing games and other multimedia applications using Haskell. We will create a set of modular libraries with the intent that they be good enough to become the de facto standards for starting new games in Haskell.
This project is currently in its early stages. This is the best time to get involved if you have any grand visions or objections to what you know about it so far. If you are interested in contributing in any way, feel free to create tickets in the bug tracker and/or drop into the #haskell-game channel on Freenode.
The functionality of our libraries is intended to be in competition with libraries like pygame. As a bare minimum, we will support common game-relevant data types, windowing, event handling, sprite-based graphics, audio, and various game-relevant file formats.
The interfaces must be simple, high level, and consistent. The process of creating a game should be straightforward. The dependencies and language extensions we use will be minimized. Even if we believe the interfaces are quite discoverable, there must also be a ridiculous amount of accessible documentation and tutorials. We should keep in mind that, if we meet the goal of becoming the de facto standard for games in Haskell, our efforts will set precedence for any work built on top; it is important to get this right as early as possible.
At the same time, we must take care to make sure that any higher level interfaces we create must not sabotage efforts by game developers to write custom, lower-level things, either. For example, while we will provide a fairly nice 2D graphics API, it should be reasonably possible to augment it with custom 3D graphics (using raw OpenGL or some other library built on OpenGL which also plays nicely with others).
The libraries we create will be fully supported on at least the most major desktop operating systems: Windows, Mac OS X, and some common Linux distributions (the other distributions should easily benefit from this support, even if only second class). Support for other operating systems or any web browsers are not primary goals, but we would be willing to try to make the job easy for a contributor, especially if you get involved early. This end may already be somewhat approachable due to the fact that we are minimizing dependencies and language extensions.
Any barrier to entry should be as low as possible. While getting this into the Haskell Platform is not a primary goal, we desire the same criteria that would be necessary, such as limiting our dependencies and avoiding conflicts with other Haskell Platform packages.
We intend for these to be practical libraries, not research projects. We have nothing against functional reactive programming or shader-generating domain specific languages, but our intiative does not implement or directly support such technologies. However, a user should be free to mix our libraries with more experimental ones.