> > Comparing the code > > records isn't > > going to tell you anything useful. > Sometimes it is enough to know that one object file is the > same or different from another. That is complicated by > compilers that put the compilation date and time, or some > other data that changes between compilations of the same source. > Most likely they should be compared member by member, though, > and not a compare of the entire TXTLIB.
Yes, that's what I meant. All you're going to get is "X is different from Y" when you compare members bit-for-bit. It's not going to tell you anything about usage or why/what is different or anything useful to evaluate whether X is a substitute for Y.
