[PRL] legacy code and AOP

Shriram Krishnamurthi sk at cs.brown.edu
Wed Feb 9 15:39:15 EST 2005


Karl,

If I understand the example correctly, yes, I would call this a
refinement, in that it extends the set of behaviors of the program.
It might, for instance, be possible to show (under the right
conditions) that one program simulates or is bisimular to the other.

Shriram



More information about the PRL mailing list