Weird question, but let's give it a shot. Off the top of my head: Standards, semantics, simplicity of implementation + time-to-implement, cost at runtime...

Weird question, but let's give it a shot. Off the top of my head: Standards, semantics, simplicity of implementation + time-to-implement, cost at runtime... What's the impetus for re-implementing image maps with JS instead of just using the browser-native implementation?