Interview with John Steele

John Steele was born in 1943 in Timperley, Cheshire. His mother was a housewife and stayed at home to look after John and later worked in the family shop. His father worked initially as a bookkeeper and later as a buyer at Manchester Oil refinery, before leaving to open a shop.

 John says his father was his most important influence in early life, adding: “I have absolutely no idea what gave me an interest in technology. My father was a manager and an expert amateur piano player; no interest in sciences at all.  From the age of about 9 or 10, I told him how to do DIY jobs especially in terms of electrical things. I had an interest in electronics, I have no idea where that came from.”

John went to the local primary school. Having sat and passed his eleven plus, he then attended Manchester Grammar School, which he describes as ‘the elite school in the North of England’.

Having chosen computing as his main interest, John looked around for computing courses at university. His options were Cambridge or Manchester and he chose Manchester. He was just 17½ when he started there.

He says his interest in electronics lead to his interest in computers: “I built my first radio set, a crystal set, when I was 10. I could listen to radio Luxembourg on my headphones.  I was dabbling around with electronics, I learned about electronics, from magazines like Wireless World and Practical Wireless which were around in the ‘50s and learned about early computing in there. That us what gave me in interest in computing.”

It was on a school trip to Manchester University in 1958 that John decided he wanted to go there to study computing. He says: “That visit helped me decide that I wanted to do computing and choosing Manchester to go to university because it was the only undergraduate course available, and then, it was only one course in the third year.  So, I did my degree in Electrical Engineering and finished up with the computing course and getting a job in computing.”

John joined Ferranti, in Wythenshawe, Manchester, in 1963 straight out of university. The first job he was given was designing the logic for the Argus 400 computer.

He says: “I joined expecting to just do circuit design, I knew nothing about logic design, which is where you’re managing just noughts and ones. I was very naïve in terms of what computing would be about. They sent me on a site visit to West Thurrock Power Station to assist with the acceptance testing and get some experience of what a site looks like, and some experience about what computer logic was. They then told me about the Argus 100 computer that they were using for processing control. That was a germanium transistor computer based on the Argus computer which was the launch computer for the Bloodhound Missile.  Ferranti were interested in making integrated circuit computers. They had a paper design for the integrated circuits and needed someone to convert their Argus100 logic design to use the new  integrated circuits.  So, I was given the freedom to actually do the logic design, never having done logic design at all, to design the Argus 400. That is what it became.

“I discovered that the integrated circuits didn’t match the functionality of the germanium transistor circuits, in particular the bistables, we call them JK Flip Flops, had very different characteristics.  So, I effectively had to redesign about two-thirds of the computer, and that was my first job.  Ferranti then had to work out how to make the integrated circuits, how to lay them out on multi-layer printed circuit cards because Ferranti were only doing single-sided printed circuit cards then.  The drawing office and the production people then spent two years to get it into a manufacturing state.  The computer design also iterated a little bit from that time, so, from design to production was about two years. That was my first job was to design the logic for the Argus 400 computer and I discovered I’d got a flair for logic design, much to my surprise.”

At the time that John joined them in the automation systems division, Ferranti was evolving from military computers to commercial computers. John says: “They were moving from being military-based to do a commercial process control system and they already had a prototype system running at ICI Fleetwood. My perspective came from commercial data processing, this was however industrial process control where you’re using a computer to control an industrial plant. Ferranti’s, when I joined in 1963, had the computer division at West Gorton and at one at Bracknell. They were making commercial computers like the Mercury computer, they had just started to produce the Atlas computer, and they had the Sirius computers, they were all commercial computer intended for commercial work.

“A few months after I joined Ferranti’s commercial computer division was sold off or merged with another company to form ICT, and West Gorton went as part of that division, including the Sirius and the other commercial computers.  Some of the people I worked with had actually worked with the people at West Gorton but that was an area which I didn’t work in at all until I joined Computer Technology many years later. It was all process control at Wythenshawe and all my experience initially was designing the computer and the equipment to go with process control, which had requirements, of course, for reasonably high availability.  If a computer goes down you don’t want an ICI production line to stop, for example.  So, it did have some critical reliability issues to address at those stages, At Ferranti, having come from a military background, you design for reliability, you take all the components at worst-case design, and if you check it works on paper at that point, then you’ve got a reliable design; and our equipment was very reliable.”

John left Ferranti at the end of 1969 to join Computer Technology as a hardware designer, based in Hemel Hempstead. He says he made the move after the offer of a 40% salary increase. He adds: “When I joined Ferranti, I got the best salary as an undergraduate from all the friends I worked with, but the salaries were falling behind the industry and I got a 40% salary increase to move to Hemel Hempstead to Computer Technology.”

John says of the differences between Ferranti and Computer Technology: “There was a different working environment, in some cases, it was better, in some cases, it was not quite so good. I worked with Ian Barron, who was founder, I knew him fairly well while we were there.  Ferranti’s didn’t have anything like that, it was a much bigger company and you weren’t quite so close to the top people. I had quite a lot of responsibility, I joined as a hardware designer but the my first project (Magnetic Tape Controller) had a heavy software element. They were writing a new operating system, these days it would be comparable with the early versions of Unix.  CTL were in commercial data processing, not industrial process control, so, it was a different emphasis.

“They were trying to produce an operating system which was more suitable for commercial work.  The first job I did was design a tape controller.  It was a low-volume product and it was by far cheaper to use a whole computer to manage it than it was to spend a man-year or so designing hardware to do the job. I had previously designed a tape controller for Ferranti. The CTL version was a heavily software-based project, I did all the hardware design, we had a team of two programmers doing the software, which was both writing the operating system drivers and also the code for the computer that was being used to directly control the tape deck.  I worked quite closely with the programmers and we delivered the tape controller and that was quite successful.

“Just as I’d finished the Magnetic Tape tape controller they were writing an operating system, E4, the aim of which was to be something roughly comparable with what Unix is; we knew nothing about Unix at the time, but it was that sort of ilk, so you could actually use it for multi-user, multi-tasking on the CTL computers around at that time.  They wanted me to join the project team to implement disk drives, which didn’t exist initially in the new Operating System E4, so, I had to add disk drives to their operating system. I was the lead having worked with the software people doing the tape system.”

John finished up leading the E4 Operating System team of five people writing the assembly code for the computer. He adds: “We were aiming at multi-user, multi-tasking and the computer’s maximum memory at the time was 56k words i.e. 112 kilobytes of memory.  We did double it later to 224 kilobytes memory.  When I was working on that team, we actually got up to 70 concurrent users on that operating system, which was quite interesting.  I worked on that for several years.”

In 1979, John was poached to join Data Logic by Jerry Gross the technical director. Jerry and John had previously worked together on a project at Computer Technology. Jerry was bought in as a consultant to advise on the design of an advanced Index Sequential Filing System which John implemented.

He says of the move: “The terms and conditions were such that it sounded a good move to make at the time. So, I joined Data Logic with the intention of them forming a microprocessor division, and Jerry wanted me to be the lead engineer.  So, I joined Data Logic expecting to be part of a company to be called Micro Logic. It was actually created as a division  We did some interesting special projects and then we moved into dealing rooms.”

After the successful launch of the Dealing room business he was with Data Logic, John then became a technical consultant advising banks on how the Data Logic solution would help their business. With the company wanting to expand into the Far East, and with his experience of dealing with very senior people, John was asked to go out to Japan as technical manager in 1988. He stayed there for two years.

He says: “We lived there for two years but Data Logic was over ambitious in its expectations in the dealing room business. We went from nothing to being, one of the leading suppliers in London, but they were too ambitious, and they were getting into problems.  The financial world was not investing as much so, by 1990, they had already made about half of the team in London redundant and I was made redundant while I was living in Japan in 1990.”

John continued to work with a Japanese partner for a couple of months after his redundancy before deciding to return to the UK, despite several job offers from Japanese companies.

He says of his time there: “One of the early jobs I did was to work out with the Japanese, how we could put Japanese symbols onto the keyboards because in financial markets each of the financial suppliers has a different keyboard for all of their services.  This was my first experience of spreadsheets, I used an Excel spreadsheet to create a printout of the keyboard layout with the Japanese characters overlaid on each of the keys, so, it would actually show the dealers what the keys would look like on the first job.”

John’s layout for the keyboard was put into action when they delivered a dealing room, he continues: “The Japanese had got a contract from a company called Asahi Life which was a company they were partnered with to actually put a dealing room in, so, we did actually deliver one dealing room. I had to get the Japanese Kanji characters to the British engraver in such a form that they could engrave the tops with three characters on the top and three characters on the front of the keys so that the Japanese keyboard layouts could all be mapped.  I’ve always been interested in computers, and the languages that are being used on computers, long before the days when they standardised character sets and that was quite an interesting challenge. The Japanese people complimented the British engraving company on the keyboard layout. That’s one lesson I learned, was about the language. The other lesson I learned is it’s a damned difficult language for foreigners to actually cope with. … Japanese is a language where the words you use depend on who you’re talking to. It is a very complicated language to actually learn to speak properly and I never did learn much Japanese, I think we learned about 100 words.

“I got on extremely well with the Japanese, they’re great fun to be with, they’re not staid and standoffish.  They’re cautious about presenting things which they think you might think offensive, so, they’re very cautious until they get to know you. They’re also reluctant to ask questions in meetings.  … I knew I’d succeeded when they started asking me questions, I was then part of the team, and that happened after about five months and before I went to live in Japan during the seven trips I did in the six months before I went to live there in Tokyo, in Harajuku, a well-known area of Japan.”

Following his redundancy while living in Japan and return to the UK in 1990, John established Soroban Systems Limited, the name was based on a Japanese abacus. He explains: “Banks in Japan were still using sorobans in the back offices in many cases for doing calculations.  It’s a modification of the original abacus, the Japanese one has got two sets of beads, so it is a combination of base 2 and base 5.  The Japanese can do arithmetic on them extremely quickly, quicker than you could put the numbers into a calculator.”

John worked freelance through his company on and off between being employed having roles with other companies.

John was approached by BT Syntegra in 1992 to work with them following a project he had done with them previously while contracting as Soroban Systems Ltd. It was his ex-colleague Jerry Gross who approached him to work on a project for a demonstrator as part of a proposal using an IBM PC; the project was based in Brighton.

 John knew he could draw on some software he had previously created in Japan. He explains: “After being made redundant in Japan, I stayed on for two months to create some software for C. Itoh (now Itochu Corporation) to provide the interface from the dealing room to a Wang computer. It was basically taking data from a real-time feed from Reuters, formatting it and sending it to a Wang computer in formatted form. I did that using DOS and I wrote some pseudo-multitasking software for DOS which allowed you to generate data patterns on the screen so you can very easily customise what data it scraped from the Reuters feed and send it to the Wang computer. With their permission, I took the right for that software with me and I had a support contract for that software.

“When I was asked to do this job in Brighton, it became apparent that something that they wanted to do was real-time data scraping from a different source, so, I offered them the software that I’d written in Japan free of charge and we used that for the demonstrator. That project was successful. …  I left the project after it came to an end and then got a phone call from Jerry saying, they liked what I did and asking if I would consider a full-time job with Syntegra.

“Basically I got the projects which were where they were complicated requirements comprising software, hardware, networking, I got the jobs that nobody knew how to do and I usually got an solution for them.  I did quite a lot of interesting jobs while I was working for BT Syntegra at that stage.

“For example, one was to provide a better conference system for dealers.  BT Syntegra had a dealing room subsidiary and they wanted to put a TV camera on top of the PC monitor to be able to instantaneously set up four-way video conferences between other dealers and counterparties sing ISDN (Integrated Digital Services Network). The ISDN camera cost £2000 in 1992. I wrote the interface software for doing that and it had to communicate over ISDN. I also wrote all the software for doing the video displays.  We put a demonstrator together so I had a video camera on top of my PC for about a year while we were doing all of this software development.  This was using integrated services digital network, ISDN as the bearer for communications and they needed one ISDN connection for each of the parties that were involved in the call. How to manage the audio was the challenge, because audio is quite difficult to manage, it’s one of the things that Zoom has got right. I designed an audio bridge to bridge the audio, it involved four resisters, and nobody believed it would work, but it did. Having my electronics background as well as a software background meant I could do all of that design.”

John designed the software so that it would split the window into four segments for up to four people on one call at any time.  The success of that first project led to John being known as the problem solver, he adds: “If anyone had a problem that they didn’t know how to solve, it was John that they came to.”

While at BT Syntegra, John worked on a project for the Land Systems Reference Centre, at Blandford. BT bid against Plessey for the project to introduce a certification system for manufacturers of army communications equipment which until that point did not exist. John says: “This was a project which took a good many years of my life.  They wanted a way of manufacturers who were offering equipment to be ready for service, to have a proper certification scheme for bringing equipment into service.  So, this is equipment that’s got to be mounted in things like tanks and armoured vehicles, land rovers, all sorts of things.  They wanted a way of interconnecting all of these devices together in a reference system so that new products coming into service or upgrades to existing products could be tested and properly certified.

“They wanted to also use it for trials of new methods of working, and to to be able to work through what they called remote detachment.  For example, if they sent somebody over to the Falklands, they wanted to be able to join those into the tests and trials as well.  So, we had to cope with communications links to far sites, all of the army’s protocols, which was quite a large number, and to be able to plug and play into a reference centre so they could actually do all the testing.  That was the challenge.”

John joined the team initially to write the UNIX code for the testing process before switching to design all the hardware infrastructure for the new building which included ensuring it was TEMPEST proof: he explains: “TEMPEST is the method of protecting your signals from being picked up by an enemy at a distance by electronic emanations that’s leaking from the signals and I had to design the whole of this building so that it was  TEMPEST proof and I was responsible for all of that side.”

John explains how he introduced the idea of using CAT 5 cabling to increase performance and reduce cost. He says: “The army was using Ptarmigan cables designed in the late 60s, so, it was very old technology.   They were very expensive, the connectors could be £200 to £1,000 each.  So, if you want to put a bay of 20 vehicles to plug into these things you’re talking about a huge amount of money just for the connectors, let alone the cables.  I had the wild idea of, we can’t do much about the connectors in the hard standing, as we called it, where you park the lorries, but I can do something about the cabling that’s coming into the building.” John asked BT’s research centre at Martlesham Heath if CAT 5 and Ptarmigan cables would be compatible. He continues: “I had a gut feeling that they were similar, so, I asked my people at Martlesham to check and they said, “Hmm, I think it’ll work.”  So, I took the risk on BT’s behalf. The design authority at DERA was very sceptical so we took a reel of Cat 5 cable and a reel of HF Quad  Ptarmigan cable, and we joined the two together and passed a signal down them.

“Anybody who is interested in electronics knows that if you get a mismatch on the join, it’ll reflect a signal back; however, it matched! So, instead of using expensive military cable, I cabled up the whole of the Land Systems Reference Centre with Cat 5 cable.”

The Cat 5 cables also passed the  TEMPEST tests for security.

Of his proudest career achievements, John highlights his work on security systems where the accreditors of the systems he built complimented his work. He says: “I took some very interesting decisions for the Land Systems Reference Centre, I don’t know whether the company knew what risks that they were taking on or not but I got every one of them right.

The secure project for the government was another where I was responsible for all the security and we got all of those right, there were no problems there with security. I did some other jobs to do with secure networks when I was working for BT and one of the proudest moments was from a security creditor who was looking at the system saying, “We’ve found three tiny problems, but nothing significant whatsoever, it’s all fit to go,” this was the very first job I had a full accreditation. I had another compliment when I was working at another site on a secure system a few years later, where I designed a system for them to use to allow them to access sensitive data across the internet.  Their accreditors said “I’ve been working in this for 20 years and this is the first time I’ve never been able to find anything to comment on, you’ve got everything right.” I’m quite-quite proud of that particular moment. Those were the two moments where I realised that actually did understand security.”

Asked about the challenges for the technology industry for the next ten years, John points to quantum computing, adding: “It is something I know very little about, it’s going to come along at some stage in the future.” He also highlights that technology is ever evolving and at pace and as a society, “we don’t really understand what’s going to happen yet.”

 Asked about challenges in defence and security, he adds: “It’s a question of trying to keep ahead of both the criminals who are trying to attack and also quite a lot of things come from foreign agents, foreign intelligence services, which we always have had to be cautious about.  I’ve got a reasonable amount of respect for our UK security authorities, and to an extent, the American authorities, having worked on the side of the fence where we’ve had communications with both.  It’s not that I don’t trust their security knowledge, it’s just that their security knowledge is more focused on protecting American interests rather than the rest of the world.

“For commercial work, there’s enough protected measures that commercial companies can take if they choose. I’m still very concerned that some of the advice that gets given out tends to be for ease of use rather than security. Computer manufacturers don’t provide proper security advice on new computers. I give the computer club I am a member of proper advice and there are papers which explain how to set up two windows accounts, for example, which is a basic method of security and yet nobody tells you to do it, so, that area worries me.”

Asked if there are any things he would do differently, John says: “I’ve been taking opportunities as they arose, so, it was a question of the opportunity arises, do I take that opportunity or not?  I don’t think I really regret any of the decisions that I’ve actually taken.  I’ve had disagreements with people and put ultimatums in and changed my roles because of that.”

John left the Land Systems Reference Centre following a change of management and highlights a project with BT where he did not believe the outcome would be beneficial. He adds: “There was one area where I was asked to work on a particular project to do with the National Health Service data communications when I worked with BT.  I said, “I don’t believe that will work, I’m not working on that”, and I didn’t work on it and eventually, the project was abandoned.  I stuck up for my guns and decided I wasn’t going to do something I was asked to do but did not believe would work.  I have been cautious about the things I’ve taken on, I’ve been very fortunate that I’ve been able to do things which I’ve found interesting.  It’s perhaps because I’ve been reasonably good at doing what I did, that I’ve been able to get away with it.”

Since 1992, John has been involved with the Gerrards Cross Computer Club which was initially founded in the 1980s. He explains: “It’s a self-help group mixing people who are amateurs and people who are professionals.  We’ve got about 30 members of which about 5 or 6 of us are, or have been, professional people.  It started off initially for the BBC Micro. By the time I joined, the IBM PC was starting to get just about affordable in 1990. I bought my first IBM PC in 1992, I paid £1500 for it, quite a lot of money in those days, it was a fairly simple machine with just 4 kilobytes of memory. I then heard about the Gerrards Cross Computer Club, which by then was using IBM PC as the main computer. I went along, I quite liked what they were talking about, so I decided to join and coincidentally, the lecturer for the following meeting announced at the meeting that he was unable to attend, he had to do something else.  I found out what the topic was C-programming so I volunteered to give his presentation for him. It was impromptu and I’ve been doing probably half of the presentations every year since then.”

The group covers a wide range of topics and more information can be found at GXCC.org.uk.

Interviewed by Elisabetta Mori

Transcribed by TP Transcription

Abstracted by Lynda Feeley