Interview with Rob Witty

Rob Witty is a renowned IT professional, his working life spans over 45 years of IT development. Born in 1949 in Yorkshire, he pursued a career in computer science, starting as a computer operator at BP. He studied at Brunel University and later worked on the Distributed Computing Systems (DCS) program, coordinating academic and industrial efforts.

Witty played a crucial role in the Alvey Programme, focusing on software engineering. He later joined GEC, where he implemented the Capability Maturity Model (CMM) to improve software engineering. Witty also led the UK’s National Air Traffic Control system and continues to research in software engineering in retirement.

+ Early Life

Rob Witty describes his early life in Yorkshire as idyllic, with a practical, hands-on environment influenced by his family’s building trade background. His father was a plumber and electrical engineer, working on radar during the war. At school he passed the 11+ exam and benefited from the state education system, which allowed him to attend a grammar school and excel in maths and physics. Rob has no memory of using computers at school, but recalls learning binary and octal number systems, which helped him later in his career.

My grandfather was a plumber. My father was a plumber and electrician. His friends and colleagues were carpenters and joiners and bricklayers and blacksmiths and those sorts of people. So I grew up in that very practical kind of milieu, and I think that influenced becoming an engineer as I liked building things.

+ BP

After school, thanks to the careers advisory service, Rob joined BP as a computer operator. Initially he was unaware of computers, so he researched them at his local library. In his interview, he quickly realized he knew more about them than his interviewer! Rob worked shifts at BP mounting magnetic tapes, (which interfered with his cricket!) but soon learned programming, leading him to pursue a degree in computer science at Essex University. When his manager learned about this, he got him an interview for a university place and a job at BP at Brunel University, so he left Essex to join Brunel, and worked on a simulation model for a petrochemical plant.

It was very unusual. I mean … I was 21 when I got to Brunel. I think I was the only person who was an employee. There were one or two people who had sort of, small bursaries from people like ICL, but they weren’t employees. They had the sort of additional top up grant to their state grant, but I was an employee. And I was three years older than all the other students. They called me grandpa. It’s very kind of them. And I felt utterly different to these kids. I’d been at work for three years. I was a professional.

Here he gained insights into safety-critical systems, including the potential danger of a chlorine tank. He recounts an incident where he was asked to program a computer to win the football pools for three Welsh men, which he found amusing and educational!

When he graduated, Rob returned to work for BP and was involved in building an operating system for a special-purpose graphics system, the FR80, while simultaneously pursuing a PhD under Professor Bob Hopgood. He also used a compiler-compiler, TREE-META, to build the operating system, which he found to be a better approach than traditional methods.

I had been taught as an undergraduate by Professor Bob Hopgood … he was one day week at Brunel and four days a week for the Atlas computer laboratory, which was part of the research council and literally ran the Ferranti Atlas computer. And he said, ‘Come and do a PhD with me in the Atlas lab.’ I… got to that realisation that BP was actually in the petrochemical business, and I was interested in safety critical software systems, and so I said to Bob eventually, ‘Yeah, okay, I’ll take the deal’. And he offered me half my BP salary to become a government scientist.

+ Research and the Alvey Programme

Rob moved from BP to the Atlas Computer Laboratory, where he continued his research and built an operating system. He was involved in the Distributed Computing Systems (DCS) programme, coordinating academic and industrial work to create a coherent research programme, liaising with industry, government and academia. Rob recounts the establishment of the Alvey Programme, where he was the second employee. Alvey was a national initiative to address the Japanese Fifth Generation Programme, and Rob wrote the software engineering strategy.

+ Xerox and GEC

Next, Rob decided to take a sabbatical from research and joined Xerox Park. Here he educated himself in AI and took classes at Stanford, including with John McCarthy, the founder of Lisp, an AI programming language. This sabbatical soon became a return to industry when he joined GEC, working for Peter Gershon to improve software engineering across the company. Rob helped implement the implementation of the Capability Maturity Model (CMM) at GEC, which significantly improved software engineering practices and productivity. This wasn’t without its challenges and he worked to overcome initial misunderstandings and resistance. Rob worked  for Lord Weinstock at GEC, and describes the challenges and successes of driving world-class performance across the company.

I loved working for Lord Weinstock. He has a sort of very fearsome reputation, but he was, he was a twinkly-eyed grandfather to me… His eyes sparkled. He was very bright, but he had a wonderful sense of humour. And at my job interview, he looked at my CV, and he said to me, ‘It says, here that you’re colour blind’. ‘Yes, Lord Weinstock’. ‘Can you tell red from black?’ And the answer is that … I’m very colour blind. I can’t tell red from black. So I looked at him, and I said, ‘I grew up in the family business. Lord Weinstock, profit and loss are second nature to me’. and he beamed at me. We got on famously from there!

+ Air Traffic Control

By this point in his career, Rob describes himself as a ‘poster boy for safety critical systems’ and it seemed a natural step to move to  the National Air Traffic Control (NATS) in 2002, where he took over the control room operations and worked to ensure the reliability of the system. He describes the successful launch of the Swanwick control centre, working with international partners, including the Spanish and Germans, to streamline air traffic control across Europe.

+ Centre for Software Reliability

In retirement, Rob continues to work with Professor Tom Anderson on proof of termination of software with safety-critical applications, aiming to improve software reliability.