----- Original Message ----- > I hold little love for UML, and as it never really "caught-on" in > Computer Science (for various reasons), it's clear I'm not alone.
What I'd say has failed is to use UML as yet another way of making applications without writing any code. I'd say that the positive point of UML is that it offers a standard vocabulary for diagrams > I understand the Qt docs are created from a Doxygen-like > "derivative", but as projects get larger, I find UML-like graphs to > take up too much space and provide insufficient information (they > aren't very useful) in production-level infrastructures. We have UML > graphs available (for free) in our code bases, but we don't really > find them useful (they are too busy, take up too much visual space, > and provide the wrong kind of detail). YMMV. I think that should be the other way round. Doxygen was written to emulate the generated Qt docs (and the generator was not freely available). I agree that for all but small classes/structs, the collaboration diagrams quickly degenerate into a spaghetti of boxes and dashed purple lines that you'd need a wall sized monitor to see all at once. I don't generate UML as that is even more cluttered. A+ Paul _______________________________________________ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest
