From F77 to F95/2003

by Yi Zhang

No, this is not a metaphor you found on tutorial texts of F95 or F2003. This is literally the real thing I am working on now. In a middle of a project trying to coupling two solvers, I am working on one of them, whose origin gives it considerable amount of unsecured variables due to F77 tradition.  What I am trying to do is to modulize it and pack & tag those variables, functions, subroutines. And, during this process, indeed some naming pitfalls showed up, apparently not caught during past coding or running. Even though somehow they do not contribute any malfunction of the solver, it indeed shows me why people make those rules of new versions of Fortran. However, believe me, it’s not the best job in the world….

FYI, what I am doing is absolutely more than this:

via Terece Tao’s GBuzz