I'm not against the idea of allowing "other" games, but they're probably not going to get many votes...
Part of the reason behind providing an implementation instead of just source code is because not everyone has access to the same build environment. If you gave me a GameMaker file, I wouldn't be able to do anything since I don't have a copy of GameMaker to load it with. If you give me a C file, I might be able to build your game, though I'm out of luck if it was coded against the Windows API. While an implementation doesn't
guarantee I can play your game (there are still considerations like Windows vs. Linux, 32-bit vs 64-bit, X86 vs. ARM...) it does significantly widen the audience.
For board games, its certainly possible to upload "assets" (depiction of the pieces) and "source code" (rules of the game). The only thing you can't upload is a "build environment" (though you can describe it!). Once again, this limits the audience that can play the game. Sure, there may be a few enterprising individuals who will grab their scissors, rulers, and sharpies, but most of us are just going to read the description and PDFs.
That's not to say I'm against making games that aren't likely to be played. For the past few years I've been trying (and finally just recently succeeded) to
write a 50,000 word novel in 30 days. Not because I want people to read my work, or even because I want to be a writer. I've been doing it simply because its a challenge. If writing a game in 48 hours which PEEKs and POKEs a Commodore 64 is what you really want to do, far be it from me to say "no" and prevent you from uploading your efforts. Just don't be surprised if nobody plays it
JigPu