>You may have individually proven every algorithm, but interaction between algorithms and the event driven OS introduces more possibilities than can be reasonably proven.

Well said, and precisely my point. An example I use in my testing presentation (and CoDe article, IIRC) was created by Genford Myers (well respected testing author). The program had a few pieces of branching code, a loop, and that’s about it. Could be written in most languages in about 20 lines of code. Myers calculated that it had 100 trillion different code paths. A fast tester could test that in about a billion years. :-)