In programming, "single character" variable names are generally considered bad practice, since they make it more difficult to understand subsequent uses of the variable (one must refer back to the definition of each symbol in an expression, which is tedious and makes the expression more confusing). Instead, variable names should in some sense "summarize" their definition; i.e., for a remainder in a division algorithm, remainder or rem would be preferred over r, which is standard in number theory texts. That's not even the worst case, since at least r can be seen as an abbreviation of "remainder": often, the letters have nothing to do with their meanings!
Is there any reason, except for historical reasons, that we can't do the same in mathematics? Many theorems I've come across would be much easier to understand if some of their symbols were replaced by programmer-style identifiers. Has anyone tried this before?
