> It's very rare to see any kind of sensible research when it comes to the science part of CS.

Have you discovered the work of Victor Basili?