SADIE, SUSIE and the Molecular 18

 
I joined BCL in about 1967 or 1968 in the Tottenham Court Road Office as a programmer working on SADIEs. Programming was hardwired and SADIEs were normally 4K or 8K! We used oscilloscopes to read off what was happening in the processor and although I don't understand too much about the engineering, the machines used a delay line to do all the calculations. Basically they gave us two lines of four 'cells' each - alpha and beta, 1 - 4, but each cell could be split into two. All the programming, in octal, consisted of allowing entries of figures and moving them around the delay line and adding, subtracting, multiplying, dividing them then printing out the results. I can still remember some of it for some bizarre reason:

01 - alpha shunt (move a1 - 3 to the right, bring a4 to the original a1 position)
02 - beta shunt
03 - crossover (swapping alpha 1 and beta 1)
04 - add (b1 to a1)
05 - add (b2 to a1)
06 - subtract (b1 from q1)
07 - left shift (multiply x 10)
10 - right shift (divide x 10)
11 - multiply (b1 x b2 answer added to a1)
12 - negative multiply (b1 x b2, answer subtracted from a1)
13 - divide (b1 / b2, answer added to a1)

Can't remember the rest, but they included carriage return, enter b1, enter b2, tab, switch off, print a1, enter description, print description, etc, all the interface between the operator and the processor. Then there were all the program jumps - if operator enters A do this, if B, do that, or go to this step in the program, etc. Also if a1 is negative do this, if positive do that. And that was more or less it.

You probably know the operator interfaced via an IBM golfball typewriter and the processor of the SADIE was in the left hand side of the metal desk. We churned out hundreds of simple invoicing, statements, payroll programs. SUSIE was a bit more sophisticated, larger and not hardwired, but used the same octal language, but was capable of more complex problems and had soft storage capacity - I particularly remember a program for working out the optimal ingredients for animal feed, taking into account nutrition and costs - that was a ****** to write!

We had an interesting tool to help with the programming, loads of lego bricks that had been printed with the different octal codes, and rather than write the programs we built them with the bricks in trays that would hold about 4ks worth of code. Then when we checked them over and found errors we could easily pull apart the bricks and add, subtract, modify the code. Remember we had to get it exactly right before the machine was hard wired, hell to pay from the engineers if they had to rewire software errors.

Now, as to other branches, they certainly were quite a few of them around the country, but I can't remember how many of them were set up when the head office was at Tottenham Court Road, and how many of them came along after the move to Portslade which was in about 1968 or 1969. We were certainly in Portslade on Decimalisation Day 15th February 1971, which will forever be engraved upon my heart!

The other sites were all sales offices as far as I remember, those that I visited were Birmingham, Newcastle, Manchester, Liverpool, Bristol, Glasgow, but I think they were all while I was based in Portslade. We even sold in Brussels and Amsterdam. The sales people (all men except for one female, Monica something) would make the initial pitch and then bring in the Systems Analyst/Programmer to clinch the deal. Once the program was written and tested we would write the instruction manual and then go to the company to check all was working ok and train the staff with the manual. The machines may have been pretty basic, but I remember excitement from some companies when we were able to replace a whole room of card sorting machinery and accounting machines with just one SUSIE. By this time the SUSIE could have a drum fitted I think ( or was that MOLLY), about the size of a dustbin lid. SUSIE certainly had some storage capacity, as well as the processor.

I can only remember one of the Branch Managers, a Ken Sutcliffe at Manchester. The Programming Manager at 180 - 182 Tottenham Court Road and at Portslade was Derek Hough. The Technical Designer was the wonderful Mick Moorhead, a brilliant man who, as far as I could make out, invented most of the processors and the software although he had no formal training - he had been a typewriter mechanic. I think he started it all up but that his 'baby' grew too big for him and he had lost control of it all by now - no business man, but a brilliant mind.

I remember in Tottenham Court Road there were both programmers and engineers, both based in the basement, sales offices on the ground floor, but whether or not the machines were actually made there, or they were just a maintenance crew I can't remember.

Another name that come to mind is Bernard O'Leary, one of the directors, based in London and Lew Harris, another director, based in Portslade. (He got out, or was pushed, and ended up with an upmarket charcuterie in Hove). By now (1972 ish) The London office had moved , I think to Gillingham Row near Victoria Station. Certainly Tottenham Court Road would have been far too small as the company had expanded very quickly. There were certainly programmers in Victoria and Portslade, but I can't remember about the other offices, they probably remained mainly sales offices.

All the hardware and software development for MOLLY was done in Portslade by BCL. I was on the development team by then and spend time working in binary, writing the bootstrap to load the hexadecimal programming language, and working with the engineers to try to 'break' the machines, programs that constantly tested calculations, etc:

Rumours began to fly about in the early to mid 70s that the company was a bit shaky and I was headhunted by Allied Business Systems who were then based in Charing Cross Road, close to CentrePoint. ABS were part of the Trafalgar House Group then. I worked for ABS for about a year or two before leaving the country. At that time ABS were competitors of BCL. I had no idea that there was a working connection between them later.

Before I joined BCL, they had produced a machine called BETSIE which was a specialist computer for the betting industry, my memory is that that was their first product, but I may be wrong. All these names were acronyms, I don't remember what they stood for, except the final E stood for electronically, so SADIE would have been something like Salaries, Accounting, D??????, invoicing Electronically.