Graham Jones (Surrey, British Columbia, IT Pro)
Graham Jones continues his series on the past 40 years he has spent in the industry.
--------------------------------------------------In part 2 I described my initial experiences of trying to do software development with Fortran 66 under pretty much impossible circumstances. Inevitably things had to change. The people “making the money” (ie. the operating parts of the company) wanted improved operation/profitability and were solidly behind our work. So they were butting heads with the people who had traditionally managed/owned the computing facilities, ie. the “commercial” end of the business (accounting, order processing, logistics, etc.). Something had to give and in the end it went to very high levels in the company. Suddenly we were permitted to go and work at the computer centre in the evenings (not during the day – that was still to be the “commercial” domain). The computer centre was very much like you see on the older movies, a huge air-conditioned room with lots of tape drives and cabinets (mainly hard drive enclosures). The machine itself was quite large since it used soft-iron core memory, a total of 512K if I remember correctly, split into several operating partitions.
Anyway, the new situation was much better than nothing and I could get as much done in one evening as I had previously in about 2 weeks. Now things were moving along. Mind you we were still limited to our 100K of 32 bit words which was really beginning to become a problem. In our attempts to manage with this we were pushing the use of “overlays” (see part 2 for a description) to its limits, and beyond. The net result was that on a number of occasions we crashed the OS shutting down the computing centre which took several hours to fix. Such things really do a lot for your popularity – not!
It was one thing to be unpopular with the computer centre management it was quite another to upset the computer operators! We had worked hard to “get to know them”, ie. get our jobs bumped up in priority to maximize the return for our time at the computer centre. Next thing we knew there were threats to ban us from the computer centre – back to square one. More high level discussions ensued and wonder of wonders, we would now be permitted 180K of 32 bit words together with much swearing, gnashing of teeth and promises of bloodletting. The “commercial” guys were slowly having to concede ground and they definitely didn’t like it!
As an aside I would like to tell you about a colleague (not on my team) who was particularly brilliant but, let’s say, a little “odd”. I am told that genius and lunacy aren’t separated by much. I wouldn’t know personally since I am too far away from genius to have to worry about it :). His name was George Wray (wow, I can actually remember his name after all of this time – he must have left an impression). If you went to his office you couldn’t actually see him at his desk because he was buried behind piles of printout up to a 2’ high arranged in very neat piles (a neat freak and definitely obsessive/compulsive). His actual working space was about 2’ square. This was George’s filing system! A request for a particular printout was usually met with a “grunt” followed by a short pause followed by a “dive” into one of the printout piles followed by holding it up for you to take. It was quite extraordinary, he never missed. He would do all of this without actually looking up and simply carried on as if you never existed. Even more extraordinary was when it was returned. He would replace it exactly where it came from, again without even looking up.
There were those who insisted that the only way to actually “communicate” with George was to feed punched paper tape into his ear and he would chatter out the reply in code using his teeth. And you thought computer geeks were a new phenomenon! A further example of his supreme “geekiness” was when he was at the computer centre. He would suddenly declare that his job had started sometimes followed by “%&$$*..., it failed”. I once asked him how he knew. “It’s obvious was his reply. Look at the PSW (64 bit Program Status Word on the front of the IBM 360/370 that showed machine activity). I can tell by the pattern of the lights. Look my job is running now and its going Inner, Temps, Vaps (subroutines in his program). It failed in Inner again. Rats!”. At this point he would rush over to the line printer to watch his printout appear. “Ok, George and what are the aliens telling you now?”. His final “glory” came when he suddenly decided that he wanted to become a Computer Centre Manager. Computer Centre Manager and George in the same sentence does not “compute”. It involves “real” people. Within 2 days of getting the job all of the female punched card operators walked out :).
Ok back to my story. Although things were progressing my distillation program was now up to approx. 10K punched cards (5 boxes) which had to be loaded into the reader every single time and weighed a “ton” to carry around. I forget the actual stats but card readers were far from perfect and there was a good chance of a misread or a jam resulting in a “chewed up” card $&(@!@%*^. “There had to be a better way”. Disk space was at a premium in those days (very expensive swappable disk packs – big and quite heavy). Clearly we had to be permitted to store our source code “on line” and only submit changes to the program. The next “battle” ensued which again we won. The computer centre people were now really “….off” to the extent that “sabotage” was suspected on a number of occasions. For example, card decks would mysteriously disappear and our “on line” storage would suddenly get erased only to be restored later. Although it would be quite some time before we got our own machines the idea was now floating around at higher levels. The returns for such an investment were beginning to be understood. The “holy of holy’s had been breached but it was never going to quite surrender”.
I had intended to discuss some of the technical challenges of using Fortran 66 and how we solved them in this part but that will have to wait for part 4 since I want to try and keep each part at a reasonable size (actually it’s an instruction from my wife so that she can get to speak to me occasionally to discuss such mundane things as “life outside computing”!).