Interview with Muffy Calder OBE

Muffy Calder (nee Thomas) was born in 1958, in Shawinigan, a small town in Quebec, Canada. Muffy says: “I was conceived in Chile, because my father was an engineer working in the nitrate mines in Chile, and they were expecting a medical emergency with my birth, so they emigrated to Canada and ended up in this obscure village in the north of Quebec without a hospital.”

Muffy’s mother was half-Norwegian, half-Swedish, and her father was of Scottish, Welsh, Dutch and German descent. She has an older sister. She describes herself as being very happy as a young child, taking apart in everything in the house, including her cot.

 From the age of eight or nine, Muffy knew that she wanted to be a scientist or engineer. She says: “I was always building things, making things and studying things. I used to collect rocks and climb trees.  I was quite a tomboy.”

Muffy attended school in Canada where she excelled and ended up jumping years to take her exams early. Her favourite subjects were maths, chemistry and physics and because she was ahead of her peers, she also had extra time to study other subjects, in particular music, learning to play the French horn. Muffy won awards for gaining the highest marks in her studies, not just in her school but across the local area. She says: “I did really well at maths, and  I loved trigonometry.  I did very well at chemistry, and it, but what I really loved was rules. I loved manipulating things on the page.  I actually didn’t care about the chemistry per se, I just loved balancing equations. I look back and think there was a computer scientist within trying to get out.  I loved rules and computing with them. I loved calculus because of what you could do with it, not what it philosophically represented.”

Muffy says her father’s own passion for science and engineering was very influential on her. If she asked him questions, he would encourage her to work out the solution with him, rather than simply giving her the answer, she explains: “He often said ‘Let’s get a clean sheet of paper, let’s get a sharp pencil, let’s write down everything we know about this and let’s see if we can work it out.’  And, actually, that stood me in good stead. When I lectured, I would try and share this with students.” Despite his support, her father did not want Muffy to study engineering because she might end up dating the boys on the male-dominated course!

In her late teens, having left school Muffy went to work as a lay person in a Dominican convent in Bodø, Norway, the most northerly convent in the world. It was run by six Sisters and one priest. She describes it as one of the best times of her life. Muffy’s work there was mostly manual, helping to run the hostel. She says: “It was such a peaceful life, it was such a different life.  The joke was, they were always trying to match me with young men in the village. I look back and I liked them so much, I just loved that life. I thought of trying to stay and then I realised, of course, I was seeing it from the outside.  If you are actually a Sister you have to sign up and you have to obey the Church, and then I thought, ‘I’m not so good at the obey bit’, but I’ll never forget it, one of the best times in my life.”

Muffy went to the University of Stirling to study maths and physics. She chose Stirling because she loved the hills the physical surroundings. In her first year, Muffy studied maths, physics, and geology as her third subject; a subject she disliked because it lacked the rules and logic that she liked so much about the other sciences. She eventually got to do a computing course, but she says it infuriated her as it focused on buying computers, not programming them. She followed this up with a course on programming which she says was “just fabulous, fantastic.”

 She explains about the first computers she used; “It was programming Cobol on punched cards, and there was one computer and   about a mile away; a man came with a suitcase once a day and picked up your punched cards. I’m not sure that I ever got a program actually compiled that term, because they were so full of syntax errors.  The next course, we moved to the Kent online system, with teletypes, that was fantastic. However, we still didn’t actually see the computer.”

Muffy’s love of her subject is for the computation and  programming, not the physical computers, she explains: “I’m not interested in the physical machine. The first computer that I really loved  was when I took a course in computability and learned about the Turing machine.  I still think that was the most fascinating, beautiful construct in the entire world…. that’s the closest to me getting interested in an actual machine.  I quite like the idea of circuit design but,  I’m much more interested in programming, programming languages  and concepts like data structures.  I just love all that aspect of computing.”

Following her success in gaining a BSc in Computer Science, Muffy went to St Andrews to study for a PhD in Computational Science. Her thesis was entitled the ‘Imperative implementation of algebraic datatypes’.  Muffy was given a few supervisors before Roy Dyckhoff “offered to rescue her”, as she describes it.

She says: “I will never forget what he did for me, never, never forget.  Sadly, he just died a few months ago, but I feel like I owe my entire career to Roy.  He turned me into a real computer scientist.”

Roy was also responsible for introducing Muffy to Ursula Martin, a Reader at London University and the most senior woman in computer science in the UK at the time. Roy asked Ursula to be the examiner of Muffy’s PhD thesis. Muffy says of the experience: “During my exam, it was a bit odd for me because I had never looked across a table and had a woman critique what I was doing. The whole man thing doesn’t bother me at all in computing … but I was always used to working with men, and men being my superior.  I had not had a woman be my superior and it was actually a bit unnerving.”

Muffy and Ursula became friends and colleagues, Muffy adds: “She’s really been my mentor ever since then, I feel, she’s looked out for me and she’s suggested things to do, she’s given wise advice.”

During her PhD Muffy took a number of part-time jobs including roles at the British Ship Research Association’, (BSRA), in the Swan Hunter shipyards in Wallsend, Burroughs Computers where she worked on a B80 machine and assembler, and the Open University, where she taught T101, microprocessor programming.

Following the success of her PhD, Muffy went to be tutor of computer science at the University of Edinburgh, followed by Stirling, and then in 1988, Glasgow, where she took a temporary lectureship. She explains: “I nearly went to Nijmegen, in the Netherlands instead, because I only got offered a temporary job at Glasgow.  But suddenly Glasgow was the place to be, there were all sorts of things happening there,  there was  a “buzz”.  But they only had a temporary job and it was  Roy  who said to me, ‘Muffy, do you want to be in the Netherlands in three years, now looking over into the UK and saying, “I’m trying to get a job there”, or why not take this job in a very exciting place now and see what happens.’  Good advice.”

 Muffy says she “came to a halt” at Glasgow in what she describes as a “tour of Scottish universities”. Today, Muffy is still at the University of Glasgow where she holds the position of Vice-Principal & Head of College of Science & Engineering and Professor of Formal Methods.

Muffy has applied her love of formal methods, which are traditionally used for specifying the behaviour and also the hardware, to open up new problems, experiencing and developing and applying models in domains as diverse as communication systems, biology, and human computer interaction. In addition, she has built and maintained connections with the industrial and commercial world. She explains: “I do theoretical computer science, I do formal modelling but I’m always trying to do it with industry, with other people.  That’s why I’ve gone out to DEC in Silicon valley, why I went to BT. I’ve always looked for application areas.”

Of her DEC experience, Muffy adds: “I’m really glad I did it though  I didn’t like it very much. I thought the work I did in DEC was very interesting and I liked many of the people, but I didn’t like the culture of Silicon Valley at all.”

As a Research Consultant for British Telecom Research and Technology in Ipswich Muffy worked on a language for protocol descriptions, called the ASN1; she was asked to write the semantics of it.  She explains: “I spent ages doing this algebraic semantics, I loved it, it was, like, fifty pages of symbols and black ink and I just thought it was the bees’ knees.”

However, when BT received it, nothing happened. This prompted Muffy to write an interpreter which she started in in imperative language (C) but moved to Miranda, in a time before Haskell.  Muffy continues: “I had such a ball doing it. I wrote the software and then I gave it to BT and then all sorts of things happened because people started using it.” It was a result of this exercise which helped her realise the power of implementation; “bringing ideas to life”.

She continues: “The implications of what you write, it’s so easy to write stuff that looks really impressive to people but actually means not much.  That’s the beauty of computing, you make it come alive, you animate it, people will then type in it, submit ASN1 specifications and see what happened.  It was a great lesson to me, and I’ve never forgotten it and, almost to this day, ever since then, I never just write symbols without making them come to life.  So that’s what I’m really into, computational models, models that you can compute with.”

Muffy worked alongside specialists at the Beatson Institute for Cancer Research, investigating biochemical pathways along which signals pass from a cell into their nucleus. A project that Muffy says emerged through serendipity when she was introduced to Walter Kolch, a cancer researcher whom she got on well with.

Her idea was to transfer the methodology behind the modelling of communication networks into molecular biology. She explains: “The commonality is, you have a signal and the signal has possible routes to get to a destination, and in molecular biology, the signal is just phosphorylation from ATP to ADP.  The route this takes to get to the nucleus affects the effect it has. That’s the basic thesis.  I could then start using all sorts of tools from modelling in comms networks to model these cell signalling networks and then start looking at how these networks can interfere and overlay with each other.”

She continues: “A lot of people, when they think about computing and biochemistry or biology, they think of data and bio-informatics.  I was going from completely the opposite end, I was looking at control mechanisms. Basically, what I’ve been doing is modelling the control mechanisms and reasoning about the control mechanisms, and one reason to do that is, the more you understand the biomolecular pathway in control you can start to think about, well, these are the places where we could interfere through some molecular intervention, for example, drug therapy, because why you want to do this is when pathways go wrong.”

Muffy is currently leading a five-year project called ‘the Science of Sensor Systems Software’; a programme funded by the EPSRC and which involves Glasgow University, the University of St. Andrews, Liverpool University and Imperial College in London. Muffy says of it: “It is one of the greatest things that I do, great in the sense that it’s most meaningful to me.”

Together with her four co-Investigators, Professors Michael Fisher,  Julie McCann and Simon Dobson, from the partner universities, Muffy is looking at the reliability of the data that is being collected by the many sensors in the many systems that are networked in today’s world. Muffy explains: “Because sensors, both individually and collectively, they break down, they lie,  they cheat, they move, they decalibrate, the protocols break down, the software is upgraded.  Yet increasingly we are using these systems, we are relying on the data and information that’s coming back from them and I want to ensure, ‘How can we trust these and how can we make sure that we can ask meaningful questions and be assured of the answers?’”

In 2018, Muffy was involved with the Blackett Review which set out to write a report assessing the computational modelling capabilities in the UK to give future direction to policy makers and government. One of the outcomes was the demonstration of the sheer amount of different types of modelling taking place in the UK today, including; logic modelling, algebraic modelling, differential equations, machine learning, AI, economic forecasting, and weather forecasting.

Muffy says: “Some of the things that we explored were very much about being very clear about what is the purpose of a model.  What questions do you want to ask of it, because if we’re not clear about that, models can take on lives of their own, and we can end up using models years into the future and asking questions for which they were never designed to give sound answers. We can become quite misled by a model if we’re not clear about what is the purpose of it. There’s a lot of ‘who’, who has responsibility all through this chain, from commissioning right to design, to maintenance, to use, to interpretation of questions.”

Another application of models that Muffy is involved in is how humans interact with systems, not just human computer interaction in the strict sense, but as interactive systems. She explains: “Although we design the system to perform a function, human users often end up using systems in ways that we never imagined; they appropriate the system.  One thing I’ve been involved in, with others, is, if we instrument their interactions, we now then have streams, traces, of all their interactions, and I can use machine learning techniques to infer models from those interactions over which I can then start to reason about different kinds of sub-populations of users.”

The research is looking at modelling existing systems that are evolving because they are not just pure black boxes, but evolving interactions. Muffy adds: “It’s basically, the system becomes evolved into other kinds of systems because we’ve put it in some environment. Models can then help us re-design, re-engineer what we have built to more accommodate, either to make it harder for users to do something we don’t want them to do, or make it easier for them to do something that we find that they are doing. So, instead of having a long computation path to carry out something we could make it a one-step action to get to that state.”

On the subject of   machine learning, Muffy warns that there is a need to understand the implications of machine learning models and how to record them for the future as they become more and more embedded in systems. She explains: “Machine learning has been around for a long time.  It’s very nice to see that it’s having its day in the sun.  Many of the techniques are brilliant.  I think we just have to remember, what is machine learning about?  We often use it when we don’t understand the mechanism, we don’t understand control.  It’s because all we have is evidence and we’re trying to infer some mechanism.”

She cites a conversation with someone from BT about the number of machine learning models, embedded models, in their communication networks. They questioned how anyone in the future will know what each model does, what role it plays and what the algorithm is using. She continues: “The phrase he used, he said, ‘little model’, ‘that little model there, model number twenty-two, little model, where did you come from?’ I   think that’s a really good catchphrase, ‘Little model, where did you come from?’  How do we document, who do we know?  Everyone’s getting caught up in making these little models but they’re going have huge implications, and how are we recording?  Where did they come from?  I think you can wrap that up in a word called ‘provenance’, but I think it’s really important and it might come back to bite us in the next decade.”

From 2004 to 2011, Muffy was involved in the UK Computing Research Committee, UKCRC, as part of the Executive Committee, and acted as the Chair between 2001 to 2011. She became involved through Ursula Martin and was involved from the formation which was the result of an international review of research in computer science by EPSRC who needed a description of research going on in the UK. UKRC formed to construct that landscape and then evolved into a body that looks after research in computing science and represents it to government and policy makers. Muffy says: “I feel it’s just so important, not that I’m particularly good at it, but someone needs to do it, so that’s why I got involved. Then, to my surprise, I was asked if I would become the Chair and I was quite taken aback but I did it and I just loved it and I think I learned so much doing it.  I learned a lot and it was really satisfying doing it as well.”

Muffy was in the meeting at the University of Cambridge when Karen Spärck Jones said the now infamous line: ‘Computing is too important to be left to men’. Muffy says: “I didn’t know Karen very well, I looked at her from a distance and had great respect for her.  It’s a lovely phrase and I think it’s one that we latch onto as a community, but I would say something further; computing is too interesting to be left to men. And that’s why I do it.  It happens to be important, at the same time it’s great, it’s a beautiful career because it’s important, it pays well, there are lots of jobs, but the really important thing is, it’s just so interesting.”

One of the aspects of computing that Muffy enjoys so much is the ability to create at any time of the day or night, in any place, without the need for special equipment. She explains: “You can do it any time you like.  I mean, that’s the beautiful thing, I think for me, compared to chemistry or physics.  I don’t need an elaborate lab.  You can write a programme at three in the morning, you can write it in the bathtub, you construct data structures in your head, it’s all in your head and then you type it in, and you create this thing.  There’s nothing more powerful than that, I think, and that is why we don’t want to just leave it to men because I want women to actually be able to experience that as well, if they want to.”

For those considering a career in technology, Muffy says: “Just do it, but I think the really obvious thing is, do you like programming?  Find out if you do, and if you do then computing is for you.  And enjoy programming.  Don’t be put off by the speed at which someone else programs.  It’s just like mathematics,   just do it yourself, it doesn’t matter how fast they do it or what their program looks like; it’s your programme, just take the sheet of paper, do it quietly, it’s your program and it’s loaded into your head.  I think there’s sometimes a bit of competition, when people are programming together.  I used to feel it, I didn’t particularly like to be in a room when other people were doing it at the same time, I didn’t find it terribly easy, and I don’t actually find it easy to work on other peoples’ code.  I don’t want to use someone else’s library, I just want to create one myself.  But, it’s like doing maths, it’s your problem, it’s your solution.”

Muffy has numerous honours to her name, including an OBE, and Fellowships of the Royal Academy of Engineering, the Royal Society of Edinburgh, and BCS, The Chartered Institute for IT, which she says have all been really meaningful to her. She wished her father had lived to see her recognised by the RAEng. She says: “I always wanted to be a scientist.  I always called myself a scientist.  Yet, when I started looking more at the breadth of science and engineering and working with the biologists I realised, I wasn’t a scientist, I was an engineer.  I don’t look at systems that have evolved through physical circumstances.  I am dreaming of systems and building them, that’s what engineering is, and then I realised I really am an engineer.  But now I’ve gone another step further, as now I’m starting to think, ‘but we can use scientific method to study and to understand the objects that we have engineered’ so I think this is beautiful interplay between science and engineering and I try and be scientific about the things we’ve engineered.”

Of her OBE, she says: “I think the OBE was just the total shock, absolute total surprise, I had no idea. … It was kind of a turning point because, at that time I was just finishing as UKCRC Chair, and I was thinking, maybe I should just go back to research.  Maybe this isn’t for me, this kind of contribution to society, maybe I’ve done my bit.  But, actually, when I got that I thought, ‘No’, I feel even more obliged to keep giving, to keep thinking about how is our science funded, what are the policies for it, what are the impacts on society, who speaks for our science in government and policy makers and more broadly for science and engineering.”

In 2011, Muffy she was awarded the OBE for services to computer science. She is a Fellow of the Royal Society of Edinburgh, the Royal Academy of Engineering and BCS, The Chartered Institute for IT.

She is Vice Principal and Head of College of Science and Engineering at Glasgow University and Professor of Formal Methods.

She is a Member of the governing body of EPSRC, Engineering and Physical Sciences Research, Chair of the DCMS Science Advisory Committee, Member of REF Main Panel B and Co-chair of EU FET Science Advisory.

Until 2015 she was on secondment to the Scottish Government as the Chief Scientific Adviser, where she also co-chaired the Scottish Science Advisory Council.

YouTube Videos

Interviewed by: Elisabetta Mori on the 11th December 2018 at the Royal Academy of Engineering

Transcribed by: Donna Coulon

Abstracted by: Lynda Feeley