I like unusual approaches, or perhaps challenges, to generic conventions.
Sure, if they work, if they are good. In the same way that a score obviously is not good merely for following convention, I don’t think a score is good merely because it is unusual or challenges convention, it has to proof itself with the result.