For another example Arizona State U. also copied many of the Rochester problems, but John Jones has since replaced most of those problems with "fake" problems that simply point to the original problems using the includePGproblem() macro. That process hasn't been carried out yet with maCalc.
We could simply drop the problems in maCalc (or at least the ones that are exact duplicates of other problems) but that might break some existing set definition files which can be a pain if it happens too often.
Having said that, maintaining backward compatibility, while a good thing in the short run, can cause a lot of extra work and often instability in the long run.
People are gearing up for a way to "clean up" the NationalProblemLibrary, winnowing out duplicates and substandard problems (if we can figure out how to identify what is substandard :-) ). It's not a process we can or should go through very often but after 10 years of accumulating questions in the NPL it seems reasonable that it is time to do some housecleaning.