Subj : Re: experience with indians? To : comp.programming From : Richard Heathfield Date : Tue Sep 06 2005 03:10 pm Phlip said: > Richard Heathfield wrote: > >> Eight people, six months, broken program which "worked" more by luck than >> judgement. Impossible to maintain. Would have given the comp.lang.c++ >> guys the screaming heebie-jeebies. > > At the 1 month mark, did they submit a working program with the most > important 1/6th of the functionality? :-) I don't actually know the full background. All I got was the maintainer asking "Rich, I have to mod this C++ code and I haven't a clue how to go about it". Not his fault, poor lamb. I doubt whether Mel himself (cf the Jargon File) could have figured it out just from the code. Fortunately, the maintainer had a thorough grasp of the original spec, and of his required mod, from which it was a simple matter to rewrite the program quickly and efficiently. You know the really scary thing? The code was about 450 lines long (ish). And it took ***four man-years*** to write it! I have to keep reminding myself that this story is true; I don't really expect you to believe it, but I do know that at least one Usenetter will, since at the time he worked at the next desk, and I think it was he who directed the maintainer to me. (Is that right, Bry? Mists of time and all that.) It consisted of two functions - main(), and a routine to display a clock at location 10, 10 of the display (bang in the middle of your last DIR statement), whose purpose - apparently - was to prove that the program hadn't crashed yet! (This in turn was because it took a good ten minutes for the program to process a measly few hundred KB of input data.) Having finished the program, I went and had a chat with the person most likely to need full knowledge of precisely how that team had spent its last six months. And that knowledge, once acquired, was acted upon. Nuff said. -- Richard Heathfield "Usenet is a strange place" - dmr 29/7/2005 http://www.cpax.org.uk email: rjh at above domain .