Tony Wachs TW stories wanted 1285

Michael Wojcik ...

around 1977 a friend wanted help setting up a service bureau. 370's had been out for a few years and 360's were cheap then, so he bought a model 30 and some disks for $30,000 (IIRC) and talked a small market research company into paying for half and housing it in their industrial unit. couldn't afford a 360 I-O set (eg. 1403, 2540 and controller - i forget the controller model number) which even used cost $50,000 at the time so he found a 4K 1401 with two 729's, printer and card read-punch and and convinced me to go work there. plus a 7-track drive for the mod 30 (that stuff was another $5000 or $8000 or so). they put in a transformer and then another guy and i, knowing practically nothing about such things ran the big power cables to all the boxes - we just laid them on top of the false ceiling and cut holes in the asbestos tiles above each unit. the ceiling didn't droop too badly but the floor was sturdy, not raised, just concrete on top of bedrock. he found a former CE who preferred making canoes to give us advice. took me a while to figure out how to a 'DOS' sysgen with no IO set, just fiddling the device dials on the console to use the 1052 at x'00f', IIRC and then using the console to type in the dump-restore 'cards'. by then i was starting to wonder if he really was my friend.

to me, that 'computer room' was magnificent. 3-inch cables snaking all over the floor and six or more ashtrays (which the cleaning lady emptied once a week), one at the console, one on the 1402 reader, one on each tape bank and so forth. nobody ventured near without my permission.

we used to get ratty old tapes for free and we went through a lot of them. i'd screw around with BAL, RPG and PL-I D tabulating programs all day and before i went home, load up a tape with a job stream that could potentially run all night. they never did but i didn't care as we only needed a few cpu hours per month to produce all our reports, so when they failed, usually because of a tape error of some sort, i'd just re-start at the previous step the next night and cross my fingers again. every tape must have had a dozen half-torn off labels on it. soon i had run out of labels and just took a magic marker to any free space i could find.

it was all quite logical in its way. in summer, the A-C wasn't sufficient to operate during day-time but in winter, all those boxes kept the room warm. regardless, it took half an hour to IPL because you had to warm up the cores first.

money was tight there too - most of the H-W money went to paying for two big Xerox colour copiers - the colour hand-drawn charts gave our reports some flash, so it wasn't long before Raytheon refused to satisfy the maintenance contract without money in return. after that we went on what IBM called "best efforts". the canoe man had showed me how to shunt contacts on various devices and if something mechanical broke like a tape drive belt, i'd just drive across town to the IBM depot and pay cash for a new one. eventually, i had removed most of the doors and panels and left them off to save time and, so i thought, to help dissipate heat.

all the programs and jobs were variations on 'extract', 'sort', 'report'. to do a national (Canada) roll-up took a while because you had to get through the Maritimes, Quebec, Ontario, Prairies and BC first. the mod30 had 64K and i used to dream at night of 256K - if i had that much memory i would have been able to do the whole country in one go.

the owner of the research company was quite a character but he did manage to compete quite brilliantly with the big guys - AC Neilsen and Time Inc who were then trying to muscle into Canada. among other things, he had originated industry 'pooling' by convincing food manufacturers that their compebreastion wasn't themselves but rather the government and the consumer. he had to be tough and smart and he was, but anything to do with computers was my domain and i had lots of fun. it was the first time anybody had ever put me in charge of anything. he had been using a manual technique since the 1950's, with 2 foot grids on large sheets of paper that he called 'spreadsheets'. this was all aided by Ella, a grand and towering Scotswoman who had gone to comptometer school as a teenager - getting electric comptometers service is another story. Morrey was always pestering me to write a computer program to do spreadsheets but i would say to him "oh Morrey, that would never be any good". not much later, Visicalc came out and it started to dawn on me that he might have right.

the 1401 was there mainly to get cards into the mod30 and print lines back so i didn't program it much. for my purposes, i didn't even need AutoCoder, part of the charm of that machine was that i could design my own little coding sheet on which i'd write object code about half-way across and follow it with comments! i thought the word-mark was a great idea. who needs an buttembler when you can comment your object?

"best efforts service" wasn't so much fun. if something broke that i couldn't handle, i'd put in a call to IBM. sometime in the next week or so (IIRC it never took longer than about 10 days), i could expect a CE to arrive. but usually not until 2nd or 3rd shift, so sometimes i was there until 3 am. the IBM CE's were all trained to be nice to customers even ones like us who were more of a nuisance than anything else, so i'd make the coffee and we had some nice chats about the future of computers. around 1980 one such CE had just transferred from India where there were still quite a few workaday 1401's. as soon as he came in, he exclaimed, "I left India to get away from the 1401's".

once, i had been staying late for more than a week, deadlines were approaching and still no CE in sight. so Morrey made some calls. to the president of IBM Canada. told him that we had 6 IBM machines and couldn't get decent service. this was sortof true - we also had 4 1950's-vintage Selectric typewriters. the next morning i arrived at my usual sometime-before-noon starting time and all hell had broken loose. the place was full of suits and very fine ones too. all the way down from the regional VP and service managers galore - what a hierarchy they had. after the initial surprise, the VP departed but the rest all enjoyed Ella's excellent coffee, had a good laugh and one of them who had been out of the field for years rolled up his $100 shirt sleeves and got me going. i forget his name but i saw a lot of him after that. i think he enjoyed getting out of the office. he would bring trainee CE's with him and while holding his personal coffee cup that we had designated for him would inspect the quality of their soldering.

there was a 2321 (data-cell drive, AKA "spaghetti-picker") too. i was drooling over its 400 MB capacity compared to the 4X7MB 2311's and i thought 500 msec access time was pretty good compared to how long it took to mount tapes, let alone keep the capstans clean. came from Manitoba across the northern Ontario wilderness in the back of friend's brother's pick-up truck. it had around 1000 hours on the clock which seemed almost new to me but everybody wanted thousands per month to maintain it. i decided to get the neat little rotary compressor out of it to run a model steam engine i'd received as a wedding present. it looked like beautiful stuff - motors with labels that said "lubricate every 10 years". NASA quality, i thought. IIRC, purchase price was something like a quarter of a million in the mid-sixties. there were also warning labels everywhere about high-pressure nitrogen and tubes and pressure fittings everywhere. after some months i tracked down a former specialist and told him what i wanted to do. the first thing he said was "have you got it in your machine room?" told me to get it out of there because if you disconnected the tubing in the wrong order, the contents of the big hydraulic oil reservoir would blow all over the place. so one fine summer day, i rolled it out to the parking lot and dismantled it in one of the parking slots. neighbours thought i was some kind of rocket scientist. ended up leaving less oil in the space than my 390 LTD normally deposited in a day.

years later when fashion dictated that mainframes were to be secreted in out-of-the-way hiding places, a brilliant young programmer asked me what a disk drive looked like. i told him that they were about so big but came in an even bigger box so that the CE could get his head inside. i was jealous of him and it made me wish i'd been born later, because in actuality so much of my so-called programming career was more akin to manual labour.

well, the editor (friend's dog) is telling me enough already and just as well since i never did get around to adding my extended version of TRT using those little clear plastic cards that sat in a tray inside the mod30 with a little vacuum cleaner permanently attached so as to keep them from sticking to each other. somebody told me you could make your own instructions and the box came with some spares and all you had to do was scrape off some of the default gold etching! not long after i convinced Morrey not to let his son spray-paint the boxes in the company colours of orange and black we dismantled it all and son got the mod30 console for his bedroom. scrap-dealer took most of it away, never paid me and i never did hook up the second mod30 which had only 32K in my basement as i couldn't afford a transformer. i don't think i could have gotten it down the stairs anyway. found myself an MVS job and gave away the 360 core as it was too small for my poor eyes to see but i still have a slice of the nice big cores from the 1401's print buffer (7 bit bytes).

wishing i had a picture, pc