Work . At uni , having written my first program in 64, I’d been bitten by the programming bug (no not that kind) so I went into programming. I was mainly on the techy side rather than the commercial side – drivers , simulation , middleware and so on.
OK – let’s dive in to machines , languages and applications. I can’t really remember them all but here goes. Machines: Starting with the Elliott 803 , on to Emidec/ICL 1101 and ICL 1900 XDS Sigmas of various flavours and OS , a bit of IBM 360 , HP , Prime, then years of PDP-11 , Vax , and of course PCs . Languages and OS: Machine code , Assemblers (including Plan , Macro-11 , Macro-32 , TI-9900) , RSX-11M , VMS , Unix , Windows , Android. Lisp , Prolog , Matlab , Fortran , C , Java , Cobol , Perl , SQL , Scripting , DCL . Application areas: Program development , Simulation modelling , graphics package development , numeric control , reverse engineering , text to speech , test suites , internationalisation , dynamic web design , configuration management , client-server collaborative music performance.
Starting off
Year 1967 – with a science degree and as long as you could identify the sharp end of the pencil and spell your name roughly correctly the door to the world of computers was wide open.
I started at Smiths Industries on the I.C.T/Emidec 1101 which had 1K words of memory , supplemented with a 3K swap drum , and plunged into commercial and technical programming . I only really remember one job. I wasn’t told who the client was but the job was the calculation of the transmission of sounds between water , through layers of various materials , and air – Hmm, who could it have been.
I learned two important work lessons from my two bosses , now unfortunately no longer with us. The valuable advice from Ted Manzocchi was “never wash a frying pan”** and from Roy Coveny was ” always keep a healthy amateur attitude to the whole business”. I have tried to keep to those two pieces of advice in my many jobs. I was writing machine code ( not even assembler ) and autocode at the time and when they switched to Cobol I said “no thanks” and departed with some redundancy money which kept me in poverty while I gigged around for a while.
** Some practices are not only not necessary but actually counter-productive and harmful.
Then back to programming to I.C.T. – yes the U.K. had a computer manufacturer then – International Computers and Tabulators. I was still with long hair and flowery jeans and shirts and often brought my guitar into work if I had a gig so as a result I was assigned to write the first copyright and royalty suite for the P.R.S. or whatever it was called at the time. I was writing in Plan for the 1900 series but after a year or so – well blow me down with a feather – they too switched to Cobol so once again I said “no thanks” and left with some redundancy money and so back to poverty and gigging again.
Then the call of the wild life of programming drew me back again to a French company Logabax. I/we actually worked the machines and fed the cards/tape in and sat at the consoles and I was the hooked. I worked on a program that was actually sewn in to the machine. They had a machine that was little more than a glorified programmable calculator. The program was hard wired in ones and zeros in ferrite core memory So you would write the program on a grid with ones and zeros and give it to the girls ( yes they were all girls ) who would sew copper wire ‘thread’ around the cores either clockwise or anti-clockwise for the ones and zeros – and that was the program sewn in.
The onto a period of writing simulation models at the UK branch of On-Line Decisions International Inc. It was owned by Col. Jackson Gouraud , seventh generation military , Purple Star and a real gent – and a very early proponent of solar energy. We wrote models that were really advanced for the time – from shop floor up to the financial statements. I wrote a scheduling simulation for TV company ATV. Lew Grade would come back from America and decide to make a six episode sitcom or a spectacular or whatever and the planners would have to fit it in to the existing schedule and my model helped them to do that – and the whole thing was written and run on a 10 character per second Teletype machine on 110bps dial-up. For my office at ATV I shared the Julie Andrews suite (she insisted on a new suite when she appeared there) with Danny La Rue , me during the day and him at night.
At On-Line there were never more than 3 or 4 of us ‘consultants’ and at the prestigious office in Mayfair (David Frost occupied the office below us) we would entertain and put on presentations to the top brass of some major companies. But the back office was sub-let to one Bob Hirschman who managed bands and there was a constant coming and going of said bands. If they played their new releases to Bob – at fill volume – and we had business visitors we had to pretend it was the ‘programmers’. The ‘programmers’ included Mott the Hoople , Genesis and Queen.
The Joy of contracting
The best job move I ever made was to leave On-line Decisions and to start contracting. Originally contracting was just to be while I looked for another job but by the time I retired 35 years later I hadn’t found one. And the best contracting jobs I had were with the first company I contracted to : Digital Equipment Corporation -DEC.
Technically Digital were the Google of the time and were producing the best software of the time and I worked there for the best part of 15 years. I worked on the PDP-11 range, mostly under RSX-11M and the early Vaxes. They had eventually tempted me to join full time with the offer of sponsoring a PhD (I hoped with John Darlington at I.C. who was building a functional program machine) but they then drew their horns in and decided to reduce their presence in the U.K. and the job and the PhD did’t materialise. But the work was fascinating. I wrote device drivers and middleware and even part of the VMS run-time system (a small part – to do with internationalisation).
The group I was part of wrote the drivers and software for special one-off or small run pieces of hardware – like mimic displays – the kind of thing you used to see in disaster movies which showed huge displays of the pipework of a nuclear power plant about to blow up – well we made and programmed that kind of display.
We also worked on internationalising Dectalk. Dectalk was/is a text to speech system originally written by Denis Klatt at M.I.T – it was the basis for Stephen Hawking’s voice machine. Everything we did was leading edge – for the time.
DECtalk was a speech synthesizer and text-to-speech technology,
developed by Digital Equipment Corporation in 1984.
It was based largely on the work of Dennis Klatt at MIT.
His source filter algorithm was first known as KlattTalk
I did a job for Shell which was to design and implement a system to integrate scheduling their own bulk oil transport trains with the British Rail schedule , since they used the B.R. lines – their previous in-house attempt had run aground. And I continued on and off with various Shell companies over the next 12 years
I developed an early computer graphics computer aided design tool for a small innovative UK company. The software also ran various numerically control laser cutters to cut out the designs and I ported the system to other companies in Canada, the US and France.
I then went on to software configuration management ; 10 years for Cellnet/O2 plus contracts at Vodaphone , Elisa ( Finnish telecomms ) , the UK Rural Payments Agency, and a company in Gibraltar (a place I loved – old fashioned UK and a 10 minute walk to Spain)
Early on I had done a bit of reviewing and one nice job in 1980 was for Practical Computing magazine which meant getting my hands on a pre-production Acorn Atom – the precursor to the BBC micro.