I'd like to say this project look's quite the opposite to promising.
1. If you're smart enough to program for the gba, then you're smart enough to know gm wasn't designed to make applications
2. gm hides most of its complexities by allowing people to operate on a much higher level without having to worry about the smaller, lower-level things, it can do this because computers are powerful enough to deal with these, the gba is not
3. he practically released a demo for a product that does "nothing"
4. ^^ if he had knowledge of any of the above he wouldn't be attempting this in the first place