Interview with Mandy Chessell CBE

Amanda Chessell was born in Bath in 1965. She is the oldest of three sisters. Her mother was a nurse and her father designed warships as a naval architect.

As a result of her father’s role in the civil service, the family moved regularly and primary school for Mandy was St Christopher’s service school in Gibraltar. Mandy says of the experience: “We had a wonderful 3 years there. The opportunity to live abroad and talk to people from different cultures is a very important part of growing up.” Mandy describes St Christopher’s School as a superb place. She says: “They push everybody to be as good as they could and it was a great school.   The whole environment and the place to grow up was very special.”

Upon returning to the UK, Mandy went to Warblington School, near Havant where she studied for her O’Levels and then Havant College to complete her A’Levels.

Mandy says one teacher, Miss Bowie, stands out for her. She taught social and economic history of Britain. Mandy says: “That was fascinating because it went through the revolutions; Agrarian, Industrial and then into modern Britain. That was fantastic background to understand what is happening to us today because we’re going through another revolution as our whole lives are being digitised.  The Covid pandemic has accelerated that for many people. So to understand what happened in the previous revolutions as employment changed, the way they lived and where they lived changed, helps us to understand what is happening now.”

While Mandy saw computers in the cupboard at Warblington School, it wasn’t until she was in the 6^th form at Havant College that she got to use one. Students were encouraged to study three A’Levels and one O’Level. Mandy opted for physics, maths and biology A’Level and O’Level computer science in her first year. In year two, she dropped biology and opted for A’Level computer science alongside maths and physics.

 At the age of 14, Mandy had a very clear idea of what she wanted to do, following a visit to the school by IBM for a careers advice session. She says: “The chap from IBM was the very first person who told me being an adult was going to be fun. All the other jobs seemed dull and I got the impression that you might do some training, but you really stopped learning and you just did your job. The IBMer said the world of technology is constantly changing, you are going to be constantly learning new things, it’s very exciting. You’re going to see the world change, you’re going to be able to travel, you’re going to have colleagues and friends all over the world. At 14 that was so inspiring; I wish I knew who he was because he set my life on a very good course.”

She adds: “It gave me direction, it gave me an insight into an industry that the school knew nothing about. I would not even have thought of this career, if he hadn’t come.  That is an incredible lesson for all our engineering disciplines.  School children are not inspired to do engineering because they don’t know what it means. They think of it as fixing washing machines and they don’t appreciate how interesting it is, the wonderful people you work with, how important what we’re doing is. It’s something that we really have to think about, because how can most teachers know what we do?”

After gaining her A’Levels, in 1983 Amanda went to Plymouth University to study for a degree in computing and informatics. She took a summer job at IBM’s Havant plant and did her industrial training in one of IBM’s Portsmouth offices and was offered a bursary for her final year of her degree. After her degree, she was offered a job with IBM.

Mandy says: “I was very, very focussed and I’ve not worked for any other company from that day.”

Mandy joined IBM at Hursley Park, IBM’s research and development centre in 1987 after completing her degree. She has spent thirty five years working for the company, changing roles every four or five years and is today a Distinguished Engineer and has been awarded the title of IBM Master Inventor.

She says: “Even though I’ve been in the industry a long time, I am doing the ground-breaking work that will be the standard part of the business going forward. I’ve changed jobs every four or five years and I’ve worked on every new wave of technology.

“Typically, I work on what we’re going to be selling in 2 or 3 years’ time. I do the first stage implementation of new products, new ideas and I work with companies that are ahead in their industry or they’re planning to make a disruptive change. Together we identify the technology we have today that will help, but also, what are they going to need going forward. When you think about those leaders, they’re going to force the rest of the industry to change and if we’ve got products in place that support that new type of business, then, of course, that’s our next generation of revenue.”

Mandy’s first job was working on mainframe assembler code for the original Customer Information Controls System (CICS) machines. She adds: “A very valuable, mature product which is still running today and still doing incredibly well. This was really good training for a career in software because you had to be very disciplined in your thinking.  In those days was memory was very expensive, so, depending on whichever program was in charge, we used to copy the memory over. You really had to understand the whole system because when it failed it would give a dump of the memory and you needed to be able to interpret what the fields were.”

Mandy’s next move was into a role as an advisory software engineer working on the CICS product on AIX, which was IBM’s Unix. She adds: “I did some of the first commercial transaction processing work on Unix boxes. At that time, no one believed you could use Unix for business, you had to have a mainframe. Of course, Unix is now a mainstay for the business.”

Mandy enjoyed her new role and the opportunity to learn new things and advance. She adds: “I had to learn C, I had to learn the VI, which is the editor at the time.  It was a completely new world and 18 months in a testing role gave me a strong foundation for the development leadership role I was about to embark on.”

Even though Mandy was still very junior in IBM, she was one of four who were the most senior in the product. She says: “We each were responsible for a quarter of the product and we did everything; planning, design, led the programming teams, spoke to customers, helped with servicing the product if there were problems with it, wrote the documentation. Consider what I learned at such a junior level – to have such a broad responsibility across an entire product. We were really lucky.  Normally you would have more senior people above you who are making those sort of decisions, but we did it all ourselves.”

Of learning C, Mandy says: “The most important thing with any language is its symmetry.  I was used to managing memory so C allowed me to think in bigger blocks, I could do bigger programs. As far as I was concerned, that was a major step forward. We were heavily criticised for using C because it was considered to be inefficient compared to coding in assembler.  However, within 18 months, the C compilers were faster than anything that a human could produce because they were doing types of optimisation that were not economically viable for a human to do.  It was exactly the right decision. This is the world and Moore’s Law, of course, is one of the reasons why we got away with it because there was plenty of processing power to allow the optimisers to run on our code.”

March 1997 Mandy became one of the lead developers of the component broker where she worked on the early object-oriented distributed object for the CORBA project work which led into J2E; object transaction processing.  This then morphed into WebSphere Application Server. She explains: “There had been a whole set of standards created called CORBA, the Common Object Request Broker Architecture, I think is its proper name. These were allowing us to write objection-oriented programs but each object could talk to an object in a different program.  It’s a distributed architecture. There was a thing called the ORB, which was the broker. We were saying that we could build a middleware system that’s transactional.” Mandy says this came from the realisation that “everything was now broken down into lots of little bits and distributed but businesses needed to work coherently, so, how do we start to bring it together?”

This was the first commercial implementation of the CORBA Object Transaction Service (OTS) open standard and there was scepticism that it would work, especially as it was using the new coding language C++.

At the same time as working on this project, Mandy started a modular master’s degree on learned object-oriented programming. She adds: “So, doing a master’s degree almost 10 years after graduating was actually very good for this industry because I learned a whole new set of techniques that were not even thought about when I was doing my degree.  I was working at the same time, and it was a very good way of feeding in new ideas into the work that I was doing.”

In the meantime, Java was developing a set of very similar standards for Java-based component objects and IBM realised that  a commercial battle was developing between CORBA and Java. Tony Storey, an IBM Fellow, advised that they bring the CORBA and Java standards together.

Mandy adds: “We decided to stop working on component broker, which was a C++ run time and we built WebSphere Application Server. This became incredibly profitable, we had WebSphere for the mainframe, WebSphere for the distributive platforms, particularly Unix. It became the basis of a huge software empire for IBM.  It was a massive revolution in terms of how distributed systems communicate and are broken down and described.”

Asked of the culture of IBM at the time Mandy joined IBM, she says that there was a lot of concern about losing intellectual property, so teams tended to remain in ‘closed bubbles’. She explains: “It was very compartmentalised, you sat in your department, you went to lunch with your department. Your view of IBM came only from your manager. We were in lots of little self-contained bubbles within the laboratory and there was a very strong focus on quality; we were very strongly trained in terms of software engineering.”

 In the mid 1980s IBM lost some of its power in the marketplace as technology moved a pace. Its CEO was replaced by Lou Gerstner who set about shaking up the company.

 Mandy says: “He did a number of things including creating the Distinguished Engineers, which is the job I have today, and the IBM Academy of Technology (AoT). This created a horizontal organisation of the majority of the technical executives. It was how we did a lot of the cross-organisational technical work, particularly the advance work that I was a part of, and that was where I started to build that global network that was promised to me when I was 14. I thought Lou Gerstner was a very great leader and he made some very bold decisions.”

A further decision made by Gerstner was to scrap Systems Application Architecture (SAA); a project IBM had been running to solve what was then considered to be the programming or coding crisis.

 Mandy says: “I was actually working on one of the API’s. The idea was that rather than having a different programming language for every product, we could identify a programming language we need for managing files, one for managing databases, and one for managing communications between programs. So, let’s have standard interfaces and get all our products to implement them, then customers can plug and play between their different products.”

The idea of standardised interfaces for common service was right, it was just ahead of its time and only within a single company. We have been trying to do it as an industry ever since.

Part of IBM’s problem was that, as the dominant leader in the industry for a long time, the IBM teams competed internally for dominance of products.  This internal focus meant they failed to keep abreast of the growth of external competitors and emerging open standards such as TCP/IP which was beginning to be adopted industry wide.  Mandy says: “For a long time, we supported both our own proprietary network protocol called System Network Architecture (SNA) but now, of course, TCP/IP is the industry standard.   This has been a theme throughout my career; we might do something proprietary to start with, but what takes off is the open standards and the use of open source. Such open systems allow everybody to play, leading to growth in terms of use and economic value.”

From its early dominance when IBM created competition within its own organisation, it is now much more focused externally and benefits from greater cross organisation collaboration and communication.

Mandy explains the dual career path at IBM with opportunities to be technical or non-technical. Mandy says: “What IBM observed was that the last thing they want are their top engineers wasting time doing admin.  I’m a technical executive, a Distinguished Engineer.  In IBM, you can be technical and not a manager, all the way up to executive and then typically, you work for a manager who is one level above you.  We have technical people right the way up to the most senior teams in the company which means that there are people who can both speak executive and can speak technical.  New ideas come from the people who are with customers, the people who are coming into the company so, you need technically eloquent people to work at all levels in the company.”

Mandy has always led small teams who support her in her technical executive role, most of her peers are in the US. She says: “I haven’t done the sort of director of IT type role at all.  What I am responsible for is the development and the insight for what are our next generation of products; what’s the next disruptive change?”

Mandy’s work on the next generation of products has often made her unpopular.  She explains: “This is a standard innovator’s dilemma; you have a successful product, here is the next generation and, in some respects, as you work on that next generation you start to damage the income of the old product.  However, soon the next generation starts to make the revenue to replace any loss from the older product and at that point I move to the next challenge.

In developing its technology, IBM has a very clear focus based on three scales:

Horizon 1 technology, Horizon 2 technology, Horizon 3. Mandy explains: “Horizon 3, focuses on customer satisfaction because this is the money engine.  Horizon 2, we’re always looking how do increase our market share. This is going to be the engine that’s going to pay the salaries going forward. Then we have the Horizon 1, where we’re working with specialist customers, building and proving the next generation and improving your ideas.  … The company has been around a very long time because of that constant recognition, that you have to have different waves of technology.”

It’s a cycle. Each generation of technology is slightly different, there are a lot of similarities, and as you move through your career, you pick up the next generation faster than a lot of people who are new, because you’ve seen it before, and all you’re doing is looking for the differences between what we had before and what we had now. It’s a constant a move to open-source which is creating those de facto standard implementations of technology.  I mean the Internet is almost all open source now and think how successful that’s been.”

 On the question of software development being subject to fashion and fads, Mandy says: “It’s frustrating because often what drives the fad and the fashion is the desire to make things simpler. Each generation starts with a very naïve programming model and then, for example, it doesn’t have security, it doesn’t think about upgrade, versioning, all those things that we know we need. Then of course, we have to bolt these complicating matters on eventually and it becomes exactly the same as the previous thing.  So, although some of this is reinvention for its own sake, often each generation does give you something slightly better. However, there is one area that is particularly bad at the moment that we’re suffering from and that is the user interface. There are so many frameworks, and the frameworks have a fashion for 5 years and then there’s a new framework and you have to rewrite everything.  So, yes, we have a lot of fads and fashions, we are moving forward as an industry in an incredibly fast way if you compare us to other industries. However, we do waste a lot of resource on redoing things with the new spelling.”

On the subject of programming models such as object orientation etc, Mandy says: “These are a way of looking at the world. What we’re doing when we’re building software, is creating a digital version of the world.  Within that scope we come across some problems that the programming model does not represent well. So, there is beauty in each of these programming styles, but none of them work for every aspect of business.  The real trick is, to know when to use each style.

Asked between formal and informal methodology, Mandy believes it is best to apply formal and informal methods where you need them. Adding: “I’m a ‘use the right thing in the right place’ type person. I’m an engineer basically, which is about making things work practically. When designing a system, formal methods give you a way of looking at the system in different ways. That’s key to quality, we need to step back and look at things from different perspectives.  But there are other situations where formal methods are completely over the top and in fact, can take you into too much detail too quickly.”

On the subject of Y2K, Mandy says: “It caused a clean-up of an awful lot of old code and got people thinking about the future. …  If we’d done nothing, many things would have shut down. We would have had so many bizarre errors and trust in IT would have been deeply shattered in the eyes of the public, if we hadn’t have done this. It’s one of those things that when you do a good job, nobody notices because nothing fails. So, I think it was important. … From an industry point of view, I think we should be proud.”

Mandy says that initially she did not want to be involved in schemes encouraging women in technology. She says: “Most women do not want anything special, they just want to be treated equally and are not aware of the fact that every day, they’re actually not behaving enough like a man to be considered technical.  So, despite everything that I’ve done, I still have people asserting that I’m not very technical. I typically have no trouble with anybody that I work with, but it’s people who are on the peripheries, and they are often people with new jobs and other opportunities.

Reluctantly, in 1999, Mandy attended to a women in technology conference in Milan and she says; “it was extremely important for my career.  They talked about the way that men and women communicate, the politics of where to sit at the table in a large meeting, all the sort of things that I just had no idea about.  It started to open up my eyes to the fact that I’m different, I do things differently.  In fact I didn’t even know I was doing things differently because often, I was the only woman and so, I couldn’t see how different I was. My fellow colleagues knew of course, but I could never see it because everyone I looked at was male.”

She adds: “There were very few technical women at IBM but that was true with my degree, and when I was 15 making my subject selections for O’Level.  I didn’t know that girls weren’t supposed to do maths and physics. I got a horrible shock.  I thought there was something wrong with me when the number of girls in the class just dropped.  I thought, oh, dear, I didn’t know I wasn’t supposed to do these subjects.  I didn’t get that memo and I’ve declined to get that memo ever since.”

Today, Mandy believes that the pandemic has opened people’s eyes to the fact that we are all different, all at different stages of our careers and all bring something different, she adds: “The stuff we’ve done for women in technology benefits everybody because it allows people to be more themselves and doesn’t create that single cookie-cutter view of how you have to behave if you want to be an engineer.  That to me is the most important thing; the opening up, the recognition of different communication styles and ways of working.”

She believes that there is still more to do on wider inclusion. She adds: “It’s not just gender. The main reason that there was a big focus on women in technology is because it’s easy to count.  It is much, much harder to count different communication styles, as in neurodiversity-type issues, different cultural and ethnic backgrounds.

“All the schemes need to work on allowing people to be as diverse as they need to be because it’s too easy for a manager who is not an engineer to focus on behaviour not results.  Whereas when you have managers who are engineeringly capable, they focus much more on results. I’ve done very much better with managers who were technically capable and could understand what we were doing.  I think the future lies in more focus what people achieve and how they work together.”

Among Mandy’s many honours are:

CBE for services to engineering (in the 2015 New Year’s Honours List)

Honorary Doctor of Science at the University of Bath, Honorary Doctor of Technology, University of South Wales, Honorary Doctor of Technology, University of Brighton, Honorary Doctor of Science, Plymouth University

Fellow of the Women’s Engineering Society, Fellow of the Royal Academy of Engineering

Honorary Fellow of the Institution of Engineering Designers

Royal Academy of Engineering Silver Medal

Fellow of the British Computing Society (BCS)

MIT’s Technology Review magazine “TR100”: a group of one hundred young innovators

Interviewed by: Richard Sharpe

Transcribed by:

Abstracted by: Lynda Feeley