This application will maintain a database that will hold a large number of lyrical stanzas collected from several music APIs across the Internet.  The service will allow users to play a game in which they tag song lyrics with emotive keywords.  Then, that data can be used for interesting things, including suggesting songs to users.  As data on tagging is collected, users that answer with the most popular tags will be rewarded in some way.  Initially, the system will be fed a number of seed artists, albums, songs, and lyrics from Last.fm and song lyrics services.  As users browse through the catalog of artists and their music or play the game, items that are incomplete in the database (for example, an artist having no songs in the database based on intentional, temporary omission) will be grabbed from the proper APIs and loaded into the database on the spot.  Dynamic loading of this sort will prevent the need to have a complete catalog of all music, yet make it possible to reach all music that exists in music datasets that are integrated with Lyrics Commander.  The database will be intentionally kept smaller by this method, until additional data is necessary for the service to function properly.