The root problem is fundamental: you're trying to recreate a physical experience in a virtual world.

More rigorous would be deconstructing the inherent constraints of the problem space as well as deconstructing the goals the system should have: is it to discover interesting books? Is it to facilitate research? Is it to enable seeing connections across genres or languages, etc. Those things, and the tradeoffs they bring, should drive your information architecture, navigation, and representation.

Great user experience is more about interaction design (that's effective for the problem to be solve) than visual design. The interaction should be molded to the problem space, rather than the other way round.

More bluntly: the 3D walk around the library is cute, but doesn't serve a purpose (as far as I can tell).

Thanks for that..I suppose the fundamental goal is to allow for serendipity, stumbling across something you'd never search for but grabs your interest when you see it