Introduction

Post-mortem tributes to Arthur can be found at the page Arthur Young - Tributes.

Below is a semi-autobiographical text.

Submission for proposed centenary History

 This account was written at the solicitation of Sir Robert Telford, as one of several contributions by radar staff requested by him in 1992 to guide the content of a new Marconi History proposed to be published for the centenary celebrations in 1997. Unfortunately this never happened – another casualty of the machinations within the company managements prior to that date. However, several of the accounts have survived in the archives at Sandford Mill, and with the permission of the authors, where possible, they are being included, with minor editing, as personal reminiscences in our History. They should be read bearing in mind the context for which they were intended and the period in which they were produced.

My Time with Marconi - by Arthur Young

I was among the first to work on digital technology in Marconi, not only in hardware and systems but also in software. Consequently my account is, especially in its earlier parts, an account of the development of digital technology and projects in Marconi.

In 1958 we had done a study for the Swedish Air Defence agency, suggesting how a digital system might meet their air defence needs. This led to a bid which proved successful in early 1960, leading to the Fur Hat project, probably the first major digital system to be produced in the UK. It included the "TAC" digital computer, a twenty-bit parallel machine which was the first to be developed in the Marconi Company, a pioneering effort in which Digby Worthy played a leading part.

Fur Hat was also a pioneer in employing a small range of logic circuits universally rather than designing many special circuits. Eric Atkins designed the circuits and Derek Watkins organised the logic design and its standards - an art being attempted for the first time in Marconi. Until that time it had been usual for the drawing office to draw all the wiring used in racks; when it was attempted on this occasion most of the drawing became uniformly black as there were vast numbers of wires. Eventually wiring was listed using a computer program written by Donald Gill for the English Electric Deuce computer - another innovative venture which assisted testing and handbook production.

I doubt whether the Company ever again took such a leap in the dark, selling a vast and complex system in the hope that it would all go together even though most of the techniques and circuits were untried. The bid and project were both done under great pressure and it was some years before we began to see that it should all go together and work - a very worrying time. In all this Roger Shipway played a leading role, a man greatly liked and respected by his staff, a leader who did not look at all like one. As the customer finally bought two Fur Hat systems the finances were not, in the end, too bad despite our alarms.

By about 1963 and 64 we were starting to apply the TAC computer in bids, selling a system to Wylfa Head for nuclear power station monitoring, among others I think. Moreover we had also started up the Myriad range of computers, 24 bit machines developed by Eric Atkins' engineering group to our specification. Eric had devised some highly compact logic circuits built from miniature discrete components, enabling the computer to be housed in a desk; it was also designed for use with peripheral equipments chosen to order, and was extremely competitive. By basing our bids on it we got a number of major orders - for mobile defence equipment in South Africa, for message switching in Cyprus, for Flight Plan Processing at West Drayton, to Plessey for Australia, and some others.

With Bob Donaldson, a somewhat eccentric figure held in high regard, I went to South Africa to assist the sale. Bob had used his kit allowance to buy a tweed sports jacket of great  weight; this he wore throughout our month in Pretoria, despite their summer heat wave.

This flood of contracts seemed to have a number of effects: it was noted that the technical effort needed to implement them was all in the Research Division; and it created a need for the writing of software - a new venture for us all. Thus a reorganisation was necessary - one in which Roger Shipway was no longer in line management, a change which I thought bad for the Company and for us all. Due no doubt to Roger's advice it was seen that software would have its problems and I was appointed to manage the "Central Automation Systems Department", a service department required to write software for the projects which had been gained. In this role (Manager, Central Automation Systems Dept), I reported to Dr O'Kane through Peter Way.

As the Department had little experience of this new software art, and as the number and quality of its staff were puny in relation to commitments, life was quite difficult. To add to our problems a squabble developed as the people in the new Line Comms Division were quite convinced they could write software better and cheaper than we could. They were allowed, in the end, to take over management of their message switching system for Cyprus. The less you know about software the easier it seems, and Lines got into quite a mess. My department was lucky in fact to lose the work as we had more than enough on our plate, with a number of projects to execute as well as the need to establish bureau facilities and software tools. We were also supporting a few bids.

As the Flight Plan Processing system had to be highly reliable,(out for no more than 15 seconds in any five year period) we had to write a new operating system to coordinate the actions of three computers; one computer was enough to carry the whole load, allowing one or two machines to fall, be repaired and be re-introduced into service virtually imperceptibly. Results were checked between machines before they were used.      

During this period we were developing new technology at a very high rate, both in the modular organisation of computer peripherals and products and in the use of operating system software. Our computer technology was well up with that of the major players but was exploited only for internal projects. As many of the applications were military the equipment was built to rather special standards of construction as it was produced in small quantities only.

In 1968/9 the English Electric Company had taken over Elliott and was trying, very much too slowly, to digest it. I spent a few months at Borehamwood in the employ of Marconi-Elliott Computer Systems (MECSL), a firm which was supposed to make a range of computers for mandatory use in projects throughout GEC. I noted that the Elliott people had used their time to establish commitments which would guarantee their future, often through contracts which seemed extremely unattractive. Further the party line at MECSL was that they had devised a range of machines which would now be designed and produced. I tried, I thought constructively, to make some counter-suggestions as I thought their plans were seriously flawed: first because I thought their plans to discontinue production of all past computers did not take account of the disruption to existing systems businesses; and second because I thought their plans placed undue emphasis on computers when the world was moving on towards intelligent terminals and the like. Because we clearly were not going to get on I arranged to move back to the newly-formed Marconi Radar Systems Limited, becoming responsible to Ian Donaldson as Manager of a Systems Group with a staff of two: Brian Partridge and Tom Gaskell.

It was immediately obvious that the MECSL computers would not provide a sensible basis for Radar projects. Moreover the first real silicon integrated circuits were just becoming available - first a memory chip and soon afterwards an arithmetic chip - ­and Brian Partridge saw, with quite exceptional insight, how intelligent terminals would develop. We started a new system, Locus 16, in which a printed circuit backplane would provide communication between computers, memory, display drivers and other devices each housed in a smaller printed board plugged into the back-plane. To custom-build an intelligent terminal you chose the necessary devices and plugged their boards into the backplane. Of course the software would still present a problem, often a serious one as designers kept trying to sell a quart of facility in a pint pot of Locus 16. However even this was less intractable than in previous generations of equipment as integrated circuits of increasing power allowed us to introduce higher capacity storage boards, and more elaborate processors, progressively modernising the product range while retaining the backplane unaltered.

We tried to interest MECSL in Locus 16, without success. This was probably a great lost opportunity as this backplane technology was introduced by Digital Equipment Corporation at about the same time and has become virtually universal, notably in the IBM personal computer and its clones. Although a substantial marketing investment would have been needed a major business could have been built on the Locus 16 architecture. As things turned out, Locus 16 became the mainstay of the Marconi Radar display business for more than a decade, some hundreds of machines being sold in Radar projects during the nineteen seventies and early eighties. The potential was not recognised, and for that reason was never fully developed.

The first customer for Locus 16 was, I believe, the Admiralty Surface Weapons Establishment, ASWE. Far from interesting ASWE in further purchases of Locus 16 we succeeded only in helping ASWE to control Ferranti's design of a very similar system called Military Argus; ASWE appeared to use Locus, and a similar Norwegian system, only as a source of guidance in deciding the specification which should apply to Military Argus. Military Argus was soon adopted by the Navy, and later by the Ministry of Defence, as a standard computer to be used throughout all MOD projects unless a sound contrary reason could be advanced. Naturally a lengthy and tedious political war developed in which I led the technical effort in GEC. The policy never really took hold, there being too many adverse cases and too much protest; it was defeated, after some years, more by personal attitudes than by its technical merits.

A bizarre aspect of this battle was the MOD view that the machine used in military equipment should also be the machine on which the applications software for the project should be developed: thus in developing software for a microcomputer built into a shell one should forgo the software tools available on much larger machines with disk stores and line printers, accepting that methods would have to be quite primitive even though the software might well be quite extensive and demanding. Software for use on Military Argus must be developed using Military Argus. In the end the technical absurdity of this view became clear to a point where it was abandoned by MOD. As the need for elaborate software tools had been the central justification for the policy, their attachment to the policy became increasingly questionable.

In 1980 or 81 I was asked to take over the Data Systems Division of Marconi Radar, a Division in which the software projects of the Company were to be implemented. I was responsible to Iain Butler. 

A major new venture, started in the early eighties, was UKADGE - United Kingdom Air Defence Ground Environment. We were in consortium with Hughes and Plessey, and I arrived on the scene in time to manage the preparation of our bid for the display terminals and their software, Hughes providing the "central processors" and Plessey the communications between them. A Company, UKSL, was formed to manage the bid and project and it was our fate to attempt to satisfy a fiery gent named Emmet Burnet, a gent seconded from Hughes to run the project. Our bid proved successful, no doubt due to the pressures exerted by Emmet.

From what I remember, our software was required in around three years within which a preliminary period of six months was to be devoted to "clarification" of the requirement. I appointed a team leader to work with UKSL in "clarifying" the requirement we were to meet, giving the team leader strict instructions to flag up all requirements in excess of our quotation and to reject any which might overload our hardware or cause us excessive costs or delays.

Thus for some years our main effort was to obtain, and to assess the consequences of, whatever requirement might emerge. We tried also to identify components which could be specified and implemented without undue risk, but for a long time there were not many. Inevitably we fell behind the required progress and a war developed: we said we could not get a clear requirement and Hughes/UKSL said there were just a few details subject to change - I think in 1984.

Our display software was completed several years before Hughes completed their "central software", despite all the Hughes' posturing. My impression remains that in the essentials of software engineering we were more advanced than Hughes; Hughes had more gadgets and outclassed us at political engineering and showmanship.

While all this was going on we had initiated work to modernise our methods and software tools, partly by imposing a disciplined procedure, agreed on by our senior designers and partly by introducing tool systems in support of this procedure. Among the tools was one for recording successive versions of software components and for building assemblies incorporating a nominated version of each component. We achieved tool systems which were reasonably effective by the standards of that time. We also developed "host-target" methods allowing software for the small Locus machine to be developed using a larger multi-­user machine connected to a Locus.

Thus in 1983 or 84 I became Assistant Technical Director in the Radar Company, responsible initially to Roy Simons and later to Reg Beckley. I undertook some minor activities connected with software, but had also developed some ideas, very vague at first, which might simplify software engineering very considerably. Being quite unable to convince anyone that these ideas had any future I was pressed to find some more immediately rewarding work but preferred to leave the Company to pursue my ideas on my own.

That was in 1987. Since then I have refined the ideas, proposing a physical model to explain the mechanisms which enable an observer to reason about the behaviour he observes. This theory suggests that any intelligent system must conform to a particular architecture, a form of "entity life" model. The theory suggests improvements in existing technology and I have applied for patents in this country and in the USA. The British application will be published next month and I have approached GEC Research to find out whether they are interested in supporting my efforts.

Looking back on my working life with Marconi I would say that I enjoyed the pre-GEC years enormously but regret not having done something else during the GEC years. GEC never got its act together in digital systems technology: instead its engineers were encouraged to improvise wildly, basing bids on whatever computer seemed the most attractive at the time. Partly for this reason most major real time projects sooner or later got into dire trouble, GEC trying to support more kinds of computer than did IBM. By setting up MECSL in 1969 an attempt at rationalisation was made. This foundered for two reasons: first because the need was for a Systems Group, rather than a Computer Company, a Group to manage the products, tools and methods applied by systems companies; and second because the MECSL computers were unsuitable for most systems business and were especially weak in the "bread-and-butter" area central to business requirements. As it was the scale of the business never afforded enough resource; what we saved in investment we tended to lose through blunders, spending much of our effort in looking for smoke and fighting fires. There was little sense of achievement or of partnership; Amstrad achieved more here than GEC.

Although I enjoyed much of my working life, and especially my many friendships, I do not feel that I was one of Nature's engineers. I drifted into Electronics partly through knowing no better and partly through being entirely unable to visualise objects in three dimensions so as to draw them. I am happy that neither of my sons has gone into technology though I think they would have gained, in their understanding of the world, from a more scientific education. Whatever is said to the contrary, engineering is still among the more menial occupations, the engineer seen by his or her manager as just another resource to be deployed. Only by abandoning engineering in favour of accountancy, or salesmanship, or law, can the engineer shake off his menial origins, a feat which may be easier for those who do not subject themselves to this arduous training in the first place!