1229 REP & 47F0

the comments in the buffer overrun thread about patches take me back in time. around 1970, the programmers where i worked got Sunday evening hands-on time (during the week, testing was 'remote' - we packaged up our card decks on the dumb-waiter for our two test shots per day plus one overnight. even though we usually had 2 or 3 programs on the go, this still left lots of time to work on cryptic crosswords).

between the batch tests and the hands-on time that involved line-ups at the 2540, nobody wanted to risk their precious test shots with a failed buttembly so we always left empty space in our programs. most people did it with DS CL256 so as not to use up to much of their base register space but i can remember using as much as a quarter of a base register's addressability. i guess about half of the space ended up being full of unconditional branches.

the senior programmers were routinely permitted to promote their programs into production in object form with lots of patch cards in them but the punks like me needed permission to do that. you'd get your shot, then line up at the 029 keypunch. after a few months of this you knew all the opcodes and formats by heart and could compose object code on the keyboard without a coding form...up to a point.

each senior had one or two juniors under their wing. a few hours into one very unproductive Sunday evening, i threw up my hands and just pulled out all my REP cards, re-buttembled and got a clean buttembly. my 'senior' was askance. he had been patching for so long that he had sort of forgotten this was possible. i still remember him shaking his head.

now that i think about it there was actually a certain machismo at play - this 'senior' was held in some awe for his 4-inch object decks, nearly two inches of which were usually patch cards.

we also had a user who when meetings became too technical was notorious for saying "can't you just patch the program?".

i have no statistics to offer on whether the advent of 'change control' reduced production abends but i do remember spending two weeks finding a cross-platform error after our 'quality control' department was moved to another country and found it easier to fiddle C-language "DEFINE's" than to make a phone call.

pc