[Harrison Pratt]

varM_real (as a variant of varM_integer) seems like something that would be commonly used and would be trivial to implement. I use a construct like that frequently in reports for totally & subtotalling.

Of course, both inc() and dec() make little sense for reals.

[Thomas Linder Puls]

Ok.

[Harrison Pratt]

The predicate inc() would be useful when calculating running means, so I guess it does make sense to keep the interface for varM_real the same as that of varM_integer.

[Thomas Linder Puls]

Ok .