It'd be great if they could release the source code and its history
Then we could see how many times there were things like
if intvar > maxint then ..
// if InputOutOfRange() then Reject() ; // removing this made the routine 0.001% faster!
in the code.
I always wondered if the authors were responsible for Microsoft's Stacker clone, which wrote to the real disk and then just assumed the write had been successful because that saved a microsecond over actually checking that it had.