Pandora, But For Games
I've been listening to lots and lots of Pandora. Pandora works for a bunch of reasons, but one of them is the quality of the underlying data contained in the Music Genome Project database.
In summary, the Music Genome Project empirically characterizes music on a song-by-song basis. Does a given song have syncopated rhythms? A dynamic vocalist? An electric guitar solo? This analysis is performed by knowledgeable professionals, so even somewhat subjective characterizations (does a given song have "demanding instrumentation?") is generally accurate. Importantly, the database does not record extraneous-to-the-music-itself factors as commercial popularity, marketing hype, or the publisher's estimation of its musical genre.
I'd like to see a database of information like this about tabletop games. Does a given game use cards? Feature simultaneous play? Require collectible components? Scale well for different player quantities? Have professional production values?
It can be very difficult to get a sense of a game's gameplay from its box or back cover text, and being able to explore a database of games based solely on objective gameplay characteristics would be a true marvel, especially if the characteristics were culled from hype and marketing by analysts knowing whereof they spoke.
Sadly, we'll never see it. The big problems of assembling such a database are twofold:
First, the process of exploring a Games Genome Project database wouldn't be a rewarding activity in and of itself, the way listening to music on Pandora is. With Pandora, you're experiencing the product in final form as you explore. With a Games Genome Project analog, you'd simply be delving the information, not playing the games.
Second, there's no money in it. Selling music is gigantic business, but selling tabletop games? Not so much. There's no way you could attract quality analysis, and without sparkling data, the whole thing falls down.
Comments
Seems like Boardgamegeek has a lot of that data already.
Have you seen Design Patterns of Successful Role-Playing Games?
One could put together something like a Games Genome Project if there were some way to mine key words out of reviews at RPG.net and other on-line review sites. My impression, however, is that there are no standards when it comes to how to describe games there, so this might be considerably more trouble than it is worth.
Seems like Boardgamegeek has a lot of that data already.
It occurred to me while I was writing that post that they might have something like that, but I didn't check. I presume you refer to their "Category" and "Mechanic" characterizations? That system is a good start, though the searching interface isn't half as useful as an analog to Pandora might be.
Have you seen Design Patterns of Successful Role-Playing Games?
Wow, that is a truly marvelous thing. Thanks for the tip.
No comments on a Game Genome project, but WOW - Pandora is GREAT! I've finally found a good way to explore new music and track what I like - plus a good option for music at work. Thanks for the rec.
-T