Hi Glenn, Patrick and others :)

I've just caught up with my email and scanned the "test data" thread ....
I like the way Patrick came up with an "opposite" to almost all my comments.
I felt like I was being beat at my own game. And I agree 100% with his
"closing arguments" ... but that is when I noticed he had stopped ending his
sentences with smiles :-) and was now being serious. I'm afraid I was not :)
I felt Glens original post was 'light' and that it merely made a point about
comparing number minds to word minds. After all it was "Off topic". So me
thought it best to reply in similar tone and I tried to give an "amusing"
programmers view of the test data and how a programmer would defend his or
her use of it (the opposite game I mentioned and got beat at by Patrick
... )

*Of course* you can't enter 3 or 4 lines of similar text into an app and
proclaim it tested !! but the serious side of test data has been covered
already ... :)))

As for progressively making test data harder ... I don't know. It reminds me
of a supervisor I once had. I'd proudly announce program X is ready and he'd
say Ok let's have a look. He would then proceed to bash the keyboard hitting
any and all keys quite randomly. Function keys where a favourite. Give it
something here, add this, delete that, all very quickly and with a minimum
of fuss. This would last no more than a minute. To my surpass and
embarrassment he could catch me out and get the program to do something it
should not. Not a very scientific approach but it was quick and produced
results. Not the errors as such, but the fact that the uncontrollable and
relentless one minute app bashing would scare me the programmer into being
more careful next time ...

Costas Anastassiades,
Athens-Greece