Oral-History:Hilary Kahn

About Hilary Kahn

Hilary Kahn was born in South Africa in 1943. She grew up there and moved to England in 1960 to begin her university studies. There, she majored in the classics – Latin and Greek. She received a computing degree at the University of Newcastle where she learned to work with the KDF 9 computer and ALGOL. Kahn later worked for Manchester University on COBOL and computer-aided design and software engineering, which she continued to do for the rest of her career. Kahn was also involved with the design of the Manchester MU5 computer system. As an integral part of the Computer Science Department at Manchester University, Kahn headed several research projects on computer-aided design and information modeling. Kahn passed away in 2007.

In this interview, Kahn talks about growing up in South Africa and her initial interest in subjects other than math and science. She also describes how she got into computing, her experience with the KDF 9 computer and ALGOL, and her interest in computers. She goes into detail on her adventures in London, arriving at Manchester University, and the different computer programs and software she used and mastered. She explains her work and experience with teaching and lecturing on computers. Kahn talks at great length about her time at Manchester University.

About the Interview

HILARY KAHN: An Interview Conducted by Janet Abbate for the IEEE History Center, 18 April 2001

Interview # 605 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, 39 Union Street, New Brunswick, NJ 08901-8538 USA. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

Hilary Kahn, an oral history conducted in 2001 by Janet Abbate, IEEE History Center, New Brunswick, NJ, USA.

Interview

INTERVIEW: Hilary Kahn
INTERVIEWER: Janet Abbate
DATE: 18 April 2001
PLACE: Hilary Kahn's office at Manchester University

Background and Education

Abbate:

It’s April 18th, 2001, and I’m speaking with Hilary Kahn.

So, to begin at the beginning, when were you born? And where?

Kahn:

Eleventh of July 1943, Cape Town, South Africa.

Abbate:

Did you grow up there?

Kahn:

Yes. I left there in 1960. About at the end of 1960.

Abbate:

Was that to go to school?

Kahn:

It was to avoid the politics of South Africa as soon as I could, and my plan was to come to England and do a degree. Keep talking?

Abbate:

Sure!

Kahn:

Well, my plan was—The theory was—I was, I didn’t have what—In England, at English universities, you require something called A Levels, and I had South African qualifications, which were not A Levels. But I had a place at Durham University, to study maths, which was very nice of them. And I went and took a look at it—and didn’t fancy it at all! —I thought it was awful.

Abbate:

Durham, or . . .?

Kahn:

It was cold and bleak, and I’d come from South Africa, and Durham in winter is quite challenging. So I decided against mathematics, and I would go forth and do A Levels, and then try again. So: I wanted to do my A Levels in one year, not two, which meant I couldn’t do science A Levels, because I couldn’t get the practical work done for in physics and chemistry; so I went for—and I didn’t know English history, and I didn’t want to do English, and I had never learned French; so I did Latin, Greek, ancient history, and maths.

Abbate:

Wow!

Kahn:

And, having done that, I was bit stuck, didn’t know what I was going to do for a degree; so I switched to Latin and Greek for my degree. So my qualifications, which I got at London University, were in Latin and Greek—and not much use to anybody!

Abbate:

Which college did you . . . ?

Kahn:

Bedford College, this was, which is now swallowed up into Royal Holloway. Very nice, right in the middle Regents Park next to the zoo!

Abbate:

How did you choose that?

Kahn:

I went ‘round lots of places, because it was an excuse to tour England—remember, I was relatively new in England. So I went to places like Exeter—because I wanted to see the cathedral; I didn’t actually have any plan to go to the university there—and all over the place. I think I would have chosen—I did want to go to Bristol; but the professor who was head of Greek there was retiring, and I, for some reason, believed that I should only go there if he was going to be there to teach me Greek—[which was] nonsense—so, then London! Because that’s where all foreigners go. Yes. London equals England!

Abbate:

What did your parents do for a living?

Kahn:

My father was a cardiologist, and he died when I was little; and my mother was . . . well qualified—she had an M.A. in Hebrew, I think—but was a housewife; and I have a sister who’s a librarian. And that’s it really.

Abbate:

How did she support you? Or did she remarry?

Kahn:

Oh, she did [support us]; no, she didn’t remarry. We had reconverted our house in South Africa into a block of flats—apartments—and I guess that was part of the funding of what kept us going. And when we came to England—everybody came to England initially; that’s my mother, my sister, and myself—but they went back, and I chose to stay on. And that was it, really!

Abbate:

Were you interested in maths as a child?

Kahn:

Oh yes! I liked maths, I liked science, I liked archaeology. I didn’t like Afrikaans; I hated it!—and that was a political dislike, not anything else. Well, it was terrible at literature, as well—absolutely awful literature! [laughs.] In my day, it was the only literature was westerns, and it was just dreadful, absolutely dreadful!

Abbate:

So, okay. You ended up here with a degree in . . .

Kahn:

Latin and Greek.

Abbate:

Classics, Latin and Greek.

Introduction to Computing

Kahn:

So, my plan was: At that point I wanted to be an archaeologist—and I wrote to the British Museum, and I wasn’t asking for a job; I was quite specifically asking for advice on what further training I should have so that one day I could have a job. I was very keen on the Greco-Roman period, and I was a rather pathetic student who signed up for the Society for the Promotion of Greek Studies and the Society for the Promotion of Roman Studies, and I went to their learned libraries in the middle of London on a regular basis, which as an undergraduate was an eccentric thing to do. Anyway, so I was very keen on that. And they wrote me a fairly nasty letter saying there was no point in my ever imagining that I could ever get a job with them, because I was not British, and it was a civil service job. And that was the end of that.

So, although I hadn’t applied for a job, they also didn’t give me the sort of advice I thought I was asking for. So I saw a sign up saying, “Go to Newcastle-upon-Tyne and do computing!” And I had no idea what a computer was. With hindsight [I realize] I had seen tape decks at 3M’s [sp?] in London—in Whitmore Street? Or Wimpole Street?—somewhere in that area, where I’d walked past when going to concerts or something. Down in the basement I’d seen these machines whirring ‘round. And that was my sole exposure to what a computer was. So I thought, “Well, I’ll go into this post-grad course in computing for a year, and if they take me that’s fine. If they don’t . . .” And they did, and that was fine!

Abbate:

So just, on the spur of the moment?

Kahn:

Sort of spur-of-the-moment, yes.

Abbate:

Did you have an idea that would be a good way to get a job?

Kahn:

No; no, no, no; no, absolutely not. It had nothing to do with careers; it had to do with that looking quite interesting. I don’t really think I understood what a career might have been. Never, never seriously thought about it, other than thinking I’d be an archaeologist—which I’m not! [laughs] Which is a pity!

Abbate:

So, so this is the mid-sixties?

Kahn:

Mid-sixties, yes. Right in there. I can never remember when I graduated, but I can look it up: ‘65 or thereabouts. Or ‘64, whatever. I got older.

So I went and did a year’s course, and they had a bit of a problem with me, because for a start, everybody else, bar one bloke, were doing a numerical analysis course, and they wouldn’t let me on that course because I didn’t have the maths background. They did very kindly get me on a course, which they eventually concocted a title for, which was Diploma in Business Processing, or some such rubbish, which had a bloke on it as well; I don’t remember much about him.

But I didn’t attend lectures on how to program. They gave me an absolutely excellent little book that had been published by what was then RSRE, which is Malvern area; and this little booklet told you how to write programs in ALGOL; I think ALGOL 60. And it introduced you to procedures and recursion very early on, so I didn’t know they were a problem. Everybody else was struggling because they were being taught the traditional way: that you learn your assignment statement, and you were to handle your declarations and so on. So I just had this book I learned from—which was very good!

Abbate:

So that you taught yourself, sort of?

Kahn:

Not really, because there were people to ask. It would be unfair to say I taught myself. I was the only girl on the course; I was the only arts graduate they’d ever taken; so it was—it was an interesting experience.

The students were used to run the computing service in the evening, and Day One they said, “This is a computer; these are tape decks; and this is a printer. This is how you load paper in the printer; and this is how you change tapes on the tape deck; and that’s a console”—which was a Flexowriter—and then they said, “Right! We’re off down the pub!” And that was it, more or less! [laughs.]

Abbate:

So you were operating the computers?

Kahn:

Yes! Absolutely!

Abbate:

For just . . . for yourselves? Or for other people as well?

Kahn:

No, no no, for other people: It was running the University’s computing service, such as it was. They used students out-of-hours.

Working with the KDF 9 and ALGOL

Abbate:

So as soon as you came there, you basically had . . .

Kahn:

Got put on with it, yes. And it was great big vacuum tape decks, yes? You’d hook the bits of tape together, and then they go pfft! [laughs.]

Abbate:

What sort of a computer was that?

Kahn:

KDF 9—which was an English Electric LEO Marconi machine. Very nice machine architecture, with a stacking architecture designed to be useful for compiler writing, and quite fun to program by hand. It didn’t go anywhere, in the long run, but it was a nice machine. I think in general the first machine you program is the one that you think was a nice machine! [laughter.]

Abbate:

I think that’s probably true. So you were doing this in ALGOL? Or in assembler?

Kahn:

ALGOL and KDF 9 machine code, yes. The assembler for it was quite a nice assembler, which gave you a lot of control over the stack; and the games were to put things into the stack and then, when you wanted them, just at the right moment they would appear at the top and you could pop them out! And that was—that felt very elegant. I’d spend hours doing useless things like making sure it was more efficient from that point of view.

Abbate:

From the memory management point of view?

Kahn:

Yes, absolutely! Which mattered in those days—when machines were not fast. I don’t remember what the memory capacity was, but it wasn’t large.

Abbate:

What sort of memory would that be?

Kahn:

Core store memory, in terms of type of memory; but what it was in terms of size, I don’t remember now—not large.

Abbate:

It’s interesting: Everyone seems to have a memory story, whether it’s rotating drums or core memory or something else.

Kahn:

Well, memory was a real issue! I mean, if your program didn’t fit or the data didn’t fit, then you spent a lot of time trying to make sure that it did fit; it was an important issue for you! And software did very little to help you. So, ALGOL 60 was a very nice language to have learned, but you had to worry about how things were being stored in order to be appropriately efficient, even in ALGOL 60. Yes.

Abbate:

When do you think that stopped? The overwhelming importance [of memory]?

Kahn:

To some extent it’s never stopped, because people keep complaining they haven’t got enough memory for X—whatever X is. Certainly the last 20 years, 15, 20 years, it’s ceased to be an issue. When I first came here, it was . . . We had a very small PDP—16K memory, probably—and trying to run simulators on that, originally, was quite, quite tough! Yes. There wasn’t a lot of room. It was a different sort of machine from the Atlas machine, which was the first [Manchester computer?] that I used.

Abbate:

When you were at Newcastle, you were taking this course, and it was called Business Computing?

Kahn:

Right. Business . . . Diploma in something-or-other to do with data processing; which meant that I learned COBOL—which was a great benefit to me—perhaps—[laughs] in future life.

Abbate:

And this was like a Masters?

Kahn:

It was equivalent of a Master’s, but you can’t give—in those days you can’t give an arts graduate a Master of Science degree—because they don’t have a Bachelor of Science. [laughs.] So it was a Diploma.

Work with Computers

Kahn:

University of Newcastle were very nice: they arranged for me to get some funding from English Electric in Kidsgrove, which is just South of Manchester, and in return for that funding, I had to work for them for a year. I actually finished the course, and then I stayed on for three months over the summer to write data processing software, in machine code, for some-or-other company—stock control software. Looking at it now, it’s sort of extraordinary that people did that sort of thing.

And then, I went to English Electric in Kidsgrove for a year, where I worked on peripheral controller software, and . . . Oh, they set me on a learn-to-compute course, originally, a whole month or something in London; which was great fun, because the course work—having paid for me for a year, you’d have thought that they would realize I didn’t need an introduction to computing course—but they forced you to take it! So, you sat there all day, did your so-called homework on the tube, came back into the middle of London, and then you were free! [laughs.] And of course, because I’d been at London University, I had friends and so on in London—so I enjoyed that! And then I did the year at English Electric, which I didn’t enjoy at all.

Abbate:

Why not?

Kahn:

Poor management. Stupid instructions. They kept on having me write little bits of some piece—I was supposedly maintaining the operating system. It was on a computer called the System IV, which was based on the UNIVAC something-or-other and the IBM 360, all have the same architecture. It’s an RCA machine. It was a ridiculous operating system. It was undocumented, more or less, and I was supposed to sit there, and should there be anything wrong with it, I had to fix it. At the time, it didn’t have a lot of bugs, so I was bored out of my mind. They then gave me this peripheral controller software to write with some bloke helping me, but it was not very interesting stuff, and they’d keep changing the specification of what I was supposed to be doing. I even got told off once for having done a variant, and told to abandon it and do the next variant; and then the time we came around to the third or fourth variant, it was a lot like one of the earlier ones—so I dug out the earlier one and finished it—and got told off for finishing it. And I was pretty annoyed, really! [laughs.] It seemed pathetic! I’m sure from the management point of view it all looked fine, but from where I was, at the lowest of the low, it was not much of a . . . I didn’t have much interest.

Abbate:

Were there a lot of other women there?

Kahn:

No. There certainly were some other women in more senior positions than I, but those were the days when the Factories Acts said that women couldn’t work in a factory. It’s a factory environment; they built machines. So, although we were in a different building, working late at night was not permitted. I used to do [it], and then I’d be told off regularly for working through till midnight or whatever. When there was stuff to be done, I’m happy to work, and by the Factories Act, you weren’t supposed to be doing that.

Abbate:

Is that just in a factory, or anywhere?

Kahn:

I don’t know. I really don’t know what the law was at the time. Yes. But it was certainly something that I was told off for a number of times.

Abbate:

Although we all know in reality that people working on computer problems are working at night.

Kahn:

Absolutely! Yes. And it was perfectly good; if you wanted access to machines and things, you worked at night.

Anyway, so that was English Electric. It wasn’t terribly interesting. And I applied for a job here, as an assistant lecturer. I don’t know why. I don’t believe I ever had a desire to be an academic; it was just an alternative.

People had applied on my behalf, and encouraged me to apply, for jobs in industry in the States. Mostly, I think, because they wanted me—I had the car, and when we went for interviews in London, I was the person who’d do the driving to London; and as part of that I needed also to be part of the application process. I think that’s why everybody filled in the forms for me! [laughter.] Maybe. I don’t remember, because I really didn’t want to go; it was in New Jersey—and I couldn’t imagine why I’d want to go to New Jersey.

Arriving at Manchester University, Working with COBOL

Abbate:

Was it at Bell Labs or something?

Kahn:

No, no no no no, it was at some oil company or other, wanting lots of computing done, and it was very well-paid, and sounded absolutely dire! So I had no real desire to go there, but I was a good taxi service for London! And anyway, I applied for the assistant lecture post here.

Abbate:

Here in Manchester.

Kahn:

Yes, right.

I cannot, I really cannot say that I had a plan to be an academic. I just had a plan not to be at English Electric. So in so far as I had a plan, that’s what it was! [laughs]

Abbate:

To escape.

Kahn:

To escape from English Electric, yes!

Abbate:

Just to back up a little: So when you’d gone to Newcastle, did you immediately say, “Oh, yes, computers are for me,” once you got a chance to use them?

Kahn:

All right! Yes, I liked them. I like an awful lot of things though, so . . . [laughter.] Quite seriously, I think if I’d been plunked in archaeology, I’d have been just as happy; if I’d been plunked in a lot of things. So, I was perfectly happy with it; I enjoy computers. I like problem-solving, and programming is problem-solving. It certainly felt like it then. Whether it still does, I’m not so sure. These days it’s web-searching and seeing if you can find somebody else who’s solved the problem for you! [laughter.] So I enjoyed that aspect of it, but I—with regret, no real plan for a career. Which I feel is bad; I should have had a plan, you know; I should have followed it through. But I didn’t.

Abbate:

I don’t think many people do.

Kahn:

So that’s as far as it went!

Abbate:

So you end up as an assistant lecturer . . .

Kahn:

. . . which is the lowest you can be, which is essentially probationary grade. And it was fundamentally very brave of the department to appoint me, because why would they appoint a classicist, with one year in industry, one year post-grad diploma in something?

Part of the reason was, I think, because I said I’d done COBOL. At this point, we were designing the fifth machine here in Manchester, MU-5, and that was a machine designed to be good for the support of compilers and processing; software support. I thought the theory was that I’d work on operating systems, which I fancied; their theory was that I’d work on compilers, which I didn’t fancy; and their particular theory was I’d write a COBOL compiler—which I truly didn’t fancy. [both laugh.] I went along with this for a bit, fairly unhappily, but, you know, I was in no position to argue.

At some point or other they said they needed somebody to help the engineers. A very basic simulator [for the MU-5] had been written to run on Atlas, which was the university’s fourth machine, and it was the university’s workhorse; and they needed people to just help the engineers do paper tape punching and generally be a dog’s body and run the staff. I volunteered, because it seemed better than writing COBOL compilers. Of course, anything was better than writing COBOL compilers.

Abbate:

This was a simulator of the new machine [MU-5]?

Kahn:

Yes. And that was really only partially effective. The person who’d started to develop that, he and I then volunteered to rewrite it. (I’m not sure it was a volunteer; we’d like to feel it was a volunteer, I’m sure.) We were told to rewrite it, and Atlas was going to be switched off at some point, and we’d turn and use a 1905E, which was an ICL machine, which was being used as a workhorse in the department. So he and I . . . So at this point I started writing logic simulators, and about that point nobody said I had to write COBOL compilers anymore, which was great! [laughs] So I ended up becoming a computer-aided design person—but it’s only because I was avoiding COBOL compilers! [laughs]

Abbate:

That’s a good enough reason! [laughs]

Kahn:

It’s a good enough reason, absolutely.

I used to lecture on COBOL. It’s the only time in my life I’ve fallen asleep while giving a lecture!

Experience with Lecturing on COBOL

Abbate:

No!

Kahn:

Absolutely! It was . . . Our lecture theaters have a pattern of two boards that switch on each side, and a single board in the middle; and I would use the two boards—the middle board and the other two boards—and then I’d have to go back to the first two. In the days of chalk, lecturing with blackboard and chalk. And I got to the end of a board, and I knew it was the end, because I had such a regular pattern of how I used the boards that when I got to here, then it must be the end. And I had no idea what I’d been saying—absolutely none whatsoever! I had to go to the back of the lecture theater and read it, so I’d know where I had to start again! Very embarrassing. [laughter.] Falling asleep in COBOL lectures!

Abbate:

So you were teaching the whole time you were doing this?

Kahn:

The students I was teaching initially were third-year students, and they’d been in computing at least as long as I had. It was all quite embarrassing! It was very funny actually, because of course my only training really had been as an arts graduate—an arts undergraduate. And the professors of Latin would pick up a book of poetry that they were going to talk to you about—and they were very old books; not old because they were old and important, they were just old and badly looked after, so they’d fall to bits—and the professor would pick up a bunch of pages and he’d walk around, lecturing from these bits of Latin—it was Horace, I remember in particular—and because he didn’t always pick up a consistent bunch of pages, you know, he’d turn the page and it wasn’t even the same poem—but he’d move on to the next poem! [laughs]. I was used to this peculiar style of lecturing; and of course when I came to lecture to science students, they thought you had to write everything on the board. If it wasn’t on the board, then they didn’t have to write it down; they had no concept of taking notes and things. After I’d lectured the first time, maybe for about twenty minutes, one of the students asked, well, was I ever going to write anything down, so that they’d know what they had to remember? [laughter.] Which I thought was sweet.

The first lectures I gave were on sorting techniques. It just happened to be something that I’d been interested in when I was at Newcastle. COBOL came later.

Abbate:

Were you making up the curriculum as you went along?

Kahn:

Oh, yes. It’s very interesting how little guidance there was. There were no standards for how you should give your talk. What you lectured on were the things that were—either the expertise you had, or an interest you happened to have. . . . There was a clear . . . That’s unfair; there was a very basic curriculum that went through what had to be taught in the first and second years, and obviously the maths base—because we inherit perfectly well from the maths experience over centuries—so there’s a sort of good maths stream. Because our department came from electrical engineering, not maths, there was also a good basic electrical engineering stream—so all our students learned about hardware; all our students learned logic design. There was no business of “Are you going to be a hardware or a software person?” You just learned that computer science was a combination. But by the time you came to the third year—which is still true, now—there was considerably more freedom in what lectures you could give. And you were able to lecture on the things that were of interest to you.

And over the . . . The course would have started in, whenever it was, ‘65 or something; so by the time we’d been running for maybe two lots of graduates graduating, there was a fairly solid central syllabus—and an absolute understanding that anybody should be in a position to lecture on any part of the syllabus; it wasn’t a matter of “I only lecture on my favorite things.” As it is now! [laughs]

There was a lot more expectation that if somebody needed to give lectures on a topic, they could do it—and if you had to learn it, then you learnt it! So, for years I lectured on—not Gaussian elimination per se, but basically on methods related to that. And as a non-mathematician, it was an oddity to ask me to do that operations research type stuff. But it was quite fun; I didn’t mind.

Abbate:

So you must have learned a lot, to be able to teach it . . .

Kahn:

Yes. The best thing of being in the department is what you learn while you’re here. Still true! Absolutely still true. If you’re not learning, you shouldn’t be hanging out here, I don’t think.

Abbate:

And did you also do the engineering side?

Kahn:

Oh, we had a little experiment . . . Well, yes and no. Very little of the lecturing, but we did decide at one point that all the software people should do a little bit of hardware design, and all the hardware people should do some software. This is the staff [I’m talking about]. So we all had to do labs; we all did the first year lab experiments one summer, and it was absolutely fantastic! It was really great fun, and you could do these bits of design, and then you had to build them on patch boards. And I had great fun color-coding the wires, so all my clock signals were a particular color, and a particular data path was blue, and another data path was green; which of course caused much amusement among the staff, because it’s not necessary to make it work. It was just—it was like knitting, really. It was quite fun! [laughter.] Oh yes, really, we all had a go at something like that, which was fine.

Computer Software Design and the Department

Abbate:

Was there a division of hardware people and software people in the department?

Kahn:

Yes, yes. That division must have started relatively early on. The first of the software people, way before my time—in fact, he left as I joined (it wasn’t my fault he left!)—was Tony Brooker, and he really built up the software side through the ‘50s, in developing proper manuals and developing various languages, in particular Atlas Autocode. He really built the department up so that there were a number people around him who were software developers. A compiler compiler was one of . . . You know what that is?

Abbate:

I’m sorry?

Kahn:

The Compiler Compiler.

Abbate:

Like YACC [Yet Another Compiler Compiler]?

Kahn:

Yes, but the first one! [laughs] And that work was work that he’d started. So had I done a COBOL compiler, then I would have been using the original Compiler Compiler, or variants of it, in the processing, which at that time would have been very unusual.

Abbate:

I would think so. Did it have a name other than . . . ?

Kahn:

Compiler Compiler! And then there was another one called the Revised Compiler Compiler, which was what my husband was—his research was in that; and I used that language for a long time for simulation, because it was a good language, its compiler-writing facilities being essentially just one aspect of it.

Abbate:

Once you got into the computer-aided design, you just continued that ever since?

Kahn:

Yes and no. I mean, we did a simulator. The guy who I worked with was called John May he and I did a simulator, which was very unusual; it certainly . . . We gave it to ICL—gave it, please note—and it became the basis of all their simulation systems until very recently. With hindsight, it’s extraordinary how little we knew of what we were doing in relation to what was going on elsewhere in the world. At the time we were doing it, there essentially weren’t logic simulators around, and the basis of it was unusual. We didn’t actually realize how unusual it was; we just got on with it, and eventually wrote a summary paper right at the end, years and years later, but never really got anything out of it in terms of papers.

We did a lot of work on simulation, and then, because we were teaching the students, hardware design as well. And they were switching eventually from— Oh, I did stuff on printed circuit boards, so tracking software and layout software for printed circuit boards. From the early ‘70s we moved into what is now called the Computer Building, and we had—before that, and then there as well—a printed circuit board fabrication facility in the department, which is very odd for computer science. So I did a lot of the software for that. Then I did software for— The students started using gate arrays, which were—logic arrays, they were called, from Ferranti, ULAs [Uncommitted Logic Arrays]—which only had a very small number of gates on, but it seemed fantastic in comparison with what we had been doing with black tape and what-have-you on boards. So I wrote gate array design systems, and all the time there were all these users; so all the time that I was doing stuff, people were using it. It’s quite hard work—and less and less obviously research, as you become more of a dogsbody.

Abbate:

I don’t know that expression.

Kahn:

I don’t know what the equivalent would be. Basically, you just keep on doing what you’re told to do; you’re the lowest of the low; you’re a gopher. A little bit like a gopher, yes? [laughter.]

Abbate:

So the design things were always targeted towards some specific need in the department?

Kahn:

In the teaching of the department, or whatever, yes. I never saw it as a—I never really had an opportunity to see it as a research exercise; it was much more: people needed something, so I did it. And I had a small team eventually, and we did it. We needed more tools to do something? Well, you just work harder and more hours, and then you produce tools that do something, and then people use them.

I was fortunate in the sense that because I was to some extent isolated—I didn’t feel isolated, but I was—I learned how to build the next system based on my previous experience, in an effective manner. In particular, I learned that if I could put as much of the intelligence of my system into external files of data, which I then used to configure or control my fundamental system, then I could switch to new requirements a lot more flexibly than writing lots more code. So I became interested in data-driven systems. But it was a self-defense exercise; it was not a big moral judgment. I’m sure I’ve made no big moral judgments in my career. I did was what was necessary in order to provide something that, when they changed what passes for their minds, I could produce some new tool relatively quickly.

Abbate:

When you said you were isolated, do you mean from other people doing computer-aided design?

Kahn:

Well, there weren’t very many in the world anyway doing C.A.D., so it’s not that; no. I was just—I was the person who did this sort of thing in the department, and occasionally I had people to help me, and I’d get students and so on occasionally; but it was not seen as a research thrust. Because it was just useful stuff that should work—and when they changed their minds it should work immediately!

Abbate:

“How could it be research if it’s useful?” [laughs]

Kahn:

Yes, yes! Absolutely!

I tried at some point, in the end successfully, to persuade the then-head of department . . . that’s not true; I don’t know the dates in relation to when he took over as head—but I did try very hard to say, as computer-aided design systems became more common in industry, “Wouldn’t it be nice if we could buy one?” —so that the students and the staff could use this other stuff with lovely graphics, and I could maybe get on with doing something which was a bit more—a little bit more opportunities for thinking, and what-have-you. We did look at it, but the person in charge, who was Dai Edwards, just couldn’t get his head around the idea that these systems cost thousands of pounds. He kept saying things like, “But a floppy”—we’re talking about these big old floppies—”a floppy only costs a few pounds,” he’d say. [laughs] And he really meant it! He wasn’t being . . . It’s just— The mindset that software had value hadn’t arrived; only hardware had value.

Abbate:

How much of a software industry was there at this [point]? I’m not sure what point this is, exactly.

Kahn:

We’re in the ‘70s. So there was an adequate software industry. People were writing operating systems; there was lots of application software around.

But the culture in this place had been that we wrote our own software, therefore it was free. And . . . well, that’s the end of it, really, isn’t it? Whereas with hardware, you knew you had to buy the wherewithal to make . . . MU-5, for example, was made of huge platters, and that cost X amount per platter, and they had money for a hundred of them—and that was it! So you could put a value on MU-5; it was a hundred platters. (Of course, that’s nonsense! It’s value was not that.) But there was no value in software, because you were going to pay the people anyway, supposedly; and then they had to do something, because otherwise they’d be bored. [laughter.] So we’d write software! That was truly how it felt. I’m not complaining about it, you understand; it seemed perfectly all right to me.

Abbate:

I wonder if some of it is . . . It seems like manufacturers had a lot of software sort of bundled in with the hardware, so that it wasn’t separately charged. I’m not quite sure how that worked here [in the United Kingdom].

Kahn:

They did until . . . Well, because . . . It was IBM that actually switched the mode, in terms of a upwards-compatible series of computers, thereby giving longevity to whatever it is: you buy one of these machines and you can live with it for some time. It’s a dead good plan, you know; you can’t fault the plan. Before that—and to some extent until people caught on to what IBM were doing—machines and their software were inextricably bound up together, and you got your machine and it had its specialist software. And yes, there might be applications over and above that, but this whole idea of portability, and “CD arrives in the post and you just load it onto your PC,” just wasn’t there. Everybody didn’t have the same Windows operating system, et cetera. And there was a lot of—I think there was a lot of tight interrelationship between the software and the hardware.

I think the department’s particular attitude—which I’m not complaining about; it’s just it was interesting—was: Because we’d always developed our own software, paying for software seemed an extraordinary thing to do. Why would you?

Abbate:

Right.

Kahn:

I mean, just: Why would you? [laughter.] It’s a little bit like saying, you know, “Why would you go out to a restaurant for a meal if your wife can cook a meal?” It’s just clearly stupid! And then you can stay in and do something, yes?

I genuinely think that was the attitude. I don’t think it was . . . It would be very, very unfair to say they were being narrow-minded, because I don’t think they were; it’s just where we’d come from. We’d come from the environment, “If you want it you write it; if you want it you build it!” Because nobody else was likely to have written or built it. So get on with it!

Now, that became less and less true, as the industry became so much broader—so I was very glad in the end when I could get out from under, and I was no longer the provider of the CAD tools. And I could look at CAD as an area of research rather than as an area of survival. [laughter.]

Abbate:

When did that happen?

Kahn:

In the ‘80s—as late as that. Largely because I didn’t . . .

There’s something else I should explain. The department had really—throughout its history, its funding for what it was doing was always [for] working on a single large research project. And the goals of what were going on [in that project] were therefore the department’s goals. Originally Tom Kilburn’s goals; his 2IC (second-in-charge) was Dai Edwards, and the two of them, with the other people who worked with them—Derrick Morris, and so on—they had the goals of where they were going, and the rest of us fitted in around that. Particularly if, like me, you were interested in writing stuff that people use—and I did like that, and I still do. And it was quite a difference, when you start saying, “Well, what else is going on in the rest of the world?” if your whole philosophy has been, you know, “We’re all doing this single project.”

The department eventually got funding from the government for MU-5: the first time it got, I think, a really huge research grant—a traditional research grant. And when I came, there was really no feeling—this was into the ‘60s—really, really no feeling that it was my responsibility as a lecturer to go out and get my own funding. My responsibility was to help the corporate goal of building MU-5 and doing the MU-5-related things. And that’s a very—doesn’t square with what you’d—if you talked to an academic now, there’s no working point of contact with that sort of attitude. And it took a long time for that to go away: ten years or so.

Abbate:

And that’s because people were doing separate projects and getting their own grants?

Kahn:

Well, originally they didn’t do separate projects; they just kept all working to the single goal. And then slowly people came in as the department grew; and they were not part of “the team” who’d been doing this exercise, so they would be looking for other funding.

And in the ‘80s—I forget the dates in the ‘80s, but in the ‘80s there was something called Alvey, which was a deliberate attempt by the U.K. government—which I guess must have been following an American model—but anyway, a deliberate attempt by the U.K. government to link industry and academia. So you had to do projects that had industrial partners; and if you did that, then they’d give you loads of money.

I actually argued against it, and refused point-blank to play for about two years before I got summoned into—by then, it must be ‘82 or something—Dai Edwards’s office and told to shut up and work with what were then Ferranti. Some part of Ferranti became ICL, but some part of Ferranti survived. And they basically sat me down with the guys from Ferranti and said, “They want this work done; you’re the person who can do it. Get on and you have a research project.” And it was quite a shock, really, because I didn’t know how to do that sort of thing!

Abbate:

What did you have to do?

Kahn:

I had to write a research proposal! All of my own, right? Which was a surprise to me!

Abbate:

So you didn’t write it with Ferranti?

Kahn:

Oh yes! Ferranti was equally involved, because it was joint money, as it were. And that’s how I started doing Alvey-related projects. Then I started doing lots of Alvey-related projects, when I could see that I could get funding to do things that were interesting.

Abbate:

So the first one . . .

Kahn:

It was some time in the ‘80s.

Abbate:

Ferranti had something that they wanted to do?

Kahn:

They wanted; and they knew we had this piece of expertise. And I’d been just refusing to play ball, and I can’t really work out why.

Abbate:

It was some sort of CAD thing?

Kahn:

Yes, oh yes. Yes, yes, yes: To do with their ULAs, or something like that. It must have been a ULA-related project. I can’t remember now. “Uncommitted Logic Array”: It’s a form of gate array.

They’d so trained me—not intentionally, but then I’m so brainwashed or whatever, that I thought my worldly life was to do whatever the department needed me to do. And this sudden switch to “Well, you’d better go get some research money, and become an independent human being”—was fine, but it was a surprise.

Abbate:

But once you got used to it, you had more . . .

Kahn:

Oh, it was no problem; it was fine.

Abbate:

. . . more autonomy?

Kahn:

“If that’s the game, then I’ll play the game.” Absolutely!

Abbate:

So it was better, in a sense.

Kahn:

Oh yes! At that point, the department then started buying in the CAD tools it wanted to use, which was fantastic, because I really didn’t want to do that. One person, maybe with one research associate or maybe with one student, cannot write sophisticated CAD software; you just can’t. Just as one department now cannot build a computer system—MU-5 was the last machine, not because there were no ideas, but because it just ceased to be possible for a university department to compete, in time-scales or anything else. Yes! The money requirements are just too great.

So it was great! And so I switched mode in the ‘80s. I can’t remember when the first independent grant was; it would have been ‘83 or something, I suppose—I don’t remember.

Abbate:

Were you able to come up with a project you wanted to do and then persuade some industrial partner . . . ?

Kahn:

Oh, eventually. Eventually, yes, absolutely. But not initially. Initially it was: Ferranti had this problem; how about?

I became interested in something EDIF, which was Electronic Design Interchange Format. I did a lot of work with research students on database work—my interest was databases to support computer-aided design of electronic systems, and other things like that—and I met this thing called EDIF, which was being proposed as a standard.

But you know, again, because of my lack of plan in life, I started looking at it because I just wanted to make sure that the capability of this proposed standard was a subset of the capability of the system that I was designing—that’s my only interest. I had no particular concern about moving designs from A to B. So I started asking these EDIF people questions, and they started . . . Initially they didn’t reply, and then some time in the mid-’80s they said, “If you’re asking all these questions, you’d better come and talk to us. How about you come and meet with us, and we’ll give—we’ll all give a tutorial.” Because by this time we’d written some software for processing it, mainly because if you want to understand something—you can turn the pages until you’re blue in the face, [but] if you write some software that interprets it, you have to really understand it. So it was a way of concentrating the mind. And so I became involved with EDIF—again completely by mistake, as always.

And I still in theory chair the EDIF stuff now, but I can’t be bothered! [laughs] I don’t do too much of it now. But it certainly was a—it meant that I had a lot of work thereafter with the States and Japan.

Abbate:

Whose standard was it?

Kahn:

The Electronic Industries Association, now known as the Electronic Industries Alliance: EIA—people who make RS-232 lines and things, and they’d never made a software-related standard before. So, there was a whole path of my life which then went along with EDIF.

Among the other things I’ve become interested in is: How can I formally specify the sorts of . . . I said I was interested in data-driven stuff, and that’s remained. How can I formally specify the data-driven stuff so that I can generate the software, rather than have people write the software? In looking at specifications and things, we started using a particular modeling language called EXPRESS. And when they were doing a . . . EDIF was being enhanced into the ‘80s, early ‘90s, and I proposed to the EIA that they should do a formal specification before they did an implementation in terms of a standard. And, largely due to the fact that they had no idea what I was talking about, they said yes! [laughter.] So it became the first of the electronic standards to have a proper specification behind it.

Abbate:

This was the mid-’80s?

Kahn:

Late ‘80s. By then that’s the late ‘80s, early ‘90s.

So I’ve worked with EDIF from the mid-’80s through till more or less now, on and off. In theory, I have the title “Director of the EDIF Technical Center”: Ooo, such an important title! [laughs]

Abbate:

Which is here?

Kahn:

Yes, yes, yes, yes yes. And I run the EDIF web sites and things. But that’s as may be. The main reason for mentioning it is more to do with the fact that you asked, how did I establish things that I became interested in? Through that route [EDIF] I did a lot of work with Japan and the States. So I’ve had quite a bit of DARPA funding, which is eccentric of me and regarded as very odd in the U.K. And not given much credence, either! [laughter.] Well, it’s just because it’s not the norm, so you can’t be measured by things that are not the norm. But it has been an interesting experience.

Professional Organizations and Teaching

Abbate:

Have you been very active in professional associations? Standards bodies, things like that?

Kahn:

Standards bodies, yes. [Mainly IEC.][This last bit got cut off; I took a guess from context.]

[TAPE 1, SIDE 2]

Abbate:

You didn’t have to deal with the ISO?

Kahn:

Less ISO and more IEC, because IEC is the ISO equivalent for the electronics domain. My dealings with ISO have been because of the modeling language that we use, EXPRESS, which is really ISO 10303, part 11. Yes, I have had dealings with them, and yes, I have been turfed out of meetings for not being sufficiently well-behaved—thinking that things are really not good and saying so! But that’s too bad. I was right, though! [laughter.] So that’s okay.

Abbate:

How has the teaching program developed here?

Kahn:

Well, of course it’s become considerably more formal. And now it’s dreadfully formalized in what we must and mustn’t do. We have something called the Quality Assurance something-or-other—QAA is whatever it stands for—teaching quality assurance and all that stuff; which means that, as with any . . . Understanding your process is important—and becomes more important than whether the process is good, bad, or indifferent. There’s always a problem; it’s a little bit like ISO 9000. There is real value—and there is genuine value—in understanding your processes; [but] there’s even more value in making sure that the process you understand is a good one.

We have a lot of—you’ll hear this from any academic now—we have an awful lot of requirements in what we have to do in documentation: justification of things we do; explanations of things we do; feedback to the students on things we do—which old people like me are not entirely convinced are to the benefit of the students! There was a time—until the age of consent, I think, dropped from 21 to 18—that we were in effect in loco parentis when the students came here, and it was genuinely possible to be more helpful to them than one can these days.

But the teaching curriculum, a lot of it, is set by standards bodies. In our case, the BCS—British Computer Society—and the IEEE both have proposals for what a computing curriculum should be.

Abbate:

Really!

Kahn:

Oh yes! Big, fat documents. And then, if you don’t follow those in some appropriate way, then you don’t get accredited by the appropriate societies. So if you want Chartered Engineer status for your graduates, then you have to do certain things. The world has changed from the sorts of stuff that . . . . Much less mathematics. Our students—our particular students—are still required to have maths A-level as their basis, but we [computer scientists], generally speaking, teach considerably less based on the mathematics, and your average student can avoid an awful lot of it. And it’s difficult to know whether that’s good, bad, or indifferent: it’s changed. It can’t be all bad, and it for sure ain’t all good. [laughs.]

Abbate:

Do they still do the electrical engineering side?

Kahn:

Much less, but we have a computer engineering course. And we just, because . . . After this department came out of electrical engineering—and it has, about a year ago, it’s reabsorbed the light-current electrical engineering, and microelectronics has come back in; so we now . . . There is no separate electrical engineering in this university; it’s within this department that’s to do with light current. So we have two or three new professors who’ve come in.

We’ve always kept the stream of computer design, and that’s very important. It gives our particular courses a balance, as it were, which is different from other people. Whether students want that balance is another issue. The jobs are not in designing chips. The jobs are going down in . . . At least ten years, five years ago, the jobs were in accounting. Now they’re only going and working for the big banks in the City of London, or being a consultant—it doesn’t matter what you’re a consultant at, but being a consultant gets well paid. But I think we are right to maintain our balance, because it is a unique arrangement.

Now I do less and less CAD stuff, and more and more—I’m interested in using the data-driven approach for wider issues than this modeling approach that we’ve been developing over the years. So I’ve been doing very little on CAD now, and more on integration of systems, and I’ve worked with different sorts of companies—construction industry, projects recently with the construction industry. I’m always serving electronics manufacture, that’s true, but I’m more interested in the manufacturing process than I am in tools to doing any design.

I don’t really have a proper purpose in life, do I? [laughs]

Abbate:

So it’s very applied research, still.

Kahn:

It’s applied, yes; oh yes. I enjoy very much if people are using what we do. That gives . . . And the people who work with me enjoy it as well. I guess that’s self-selecting, isn’t it? We don’t like—we’re not theoreticians—I’m not objecting to theory, but I’d just rather apply it. So we will use some of the more formal techniques and then apply them, rather than just research into the formal techniques themselves. Just how we are! [laughter.]

Women in the Academy

Abbate:

Have you noticed that female students end up in certain areas more than others, at least at the graduate level?

Kahn:

We have so few female students, we had so few female staff members, [that] it’s difficult to judge. There are a small number of women who are interested in the engineering side, and there always have been and always will be. But there’s a small number of men who are interested in the engineering side, yes? The majority of where people end up is on the software side. And I don’t think I see much of a difference in—hearing what the students are going to do with their careers, I don’t really see a huge difference in where they’re setting off to.

There may be selection at the—I don’t know how many women get into the big banks, for example. I think if they’re good enough, they will.

Abbate:

So [selection might occur] after they graduate . . .

Kahn:

In terms of what courses they do: not really. I haven’t noticed—but I’m perhaps a bit insensitive to it.

I have a problem in the fact that I tend not to worry if I’m the only woman at the meeting, or . . . I was brief—I was for a period the only woman professor in the science faculty, which is. . .

Abbate:

Of the whole university?

Kahn:

Of the whole university.

Abbate:

When was that?

Kahn:

Oh, until last year! For a period of— There was maths professor, and she retired, so— I don’t know, maybe it was a couple years [that] there was just one of me. Before that, there were just two of us. And that’s horrific! That’s wrong.

Abbate:

When did you become a professor?

Kahn:

Early ‘90s? I can’t remember. I’m not very good at these things, am I? [laughs]. It didn’t make a lot of difference.

Abbate:

So it didn’t really make . . .

Kahn:

Not from my point of view. I’ve been very independent. Since I discovered that I had to go forth and find money for my research and things, I’ve operated essentially very independently, so I haven’t belonged to anybody particular. I haven’t . . . In the tree, in the hierarchy of the—they’ve randomly appointed me—before I was a professor—randomly appointed people who logically I should report to, but there’s nobody [who] ever bothered to ever ask me what I was doing! It didn’t really matter! [laughs.]

Abbate:

It hasn’t mattered too much . . .

Kahn:

It hasn’t mattered too much. And I think, I reckon, if you do what you’re supposed to be doing, and you do it okay, people should leave you to it, basically. As long as you’re not, you know, messing up the system, undermining the system or whatever.

So I’ve tended to be—incorrectly, I’m prepared to believe, but nevertheless, I’ve tended to be insensitive to the gender imbalance in the department. My husband and I argue quite a bit, or used to argue quite a bit about that: he’s certainly more of a feminist than I am, and he would go for positive discrimination, and it annoys the hell out of me, basically! [laughs.]

Abbate:

Is he in this department?

Kahn:

He was, yes, he retired recently. He was the Revised Compiler Compiler man, as opposed the Compiler Compiler man. [laughter.]

But I certainly am not—I’ve done my stint in chairing committees in the university, or chairing committees in the Science Research Council—the equivalent of the National Science Foundation, right?—and I’d rather hope that the reason I was chairing the committees was not because they wanted to get the quota of women up. Because I’m not that soft a touch, so they shouldn’t do that if they were just trying to get women! [laughing] They will send me in occasionally as the hatchet man, when there are difficult projects for review and so on; because I have a reputation that I will be a hatchet person. In fact, I do it very rarely; but I do tend, if you follow me, to say what I think. I’m not very good at the politics!

Abbate:

Well, is that a disadvantage?

Kahn:

Oh, I should think so, absolutely! Yes! Oh yes, if I’d wanted to progress up—which I don’t—but if I’d wanted to progress up the university hierarchy, then you had certainly better behave a lot better than I did. And I’m quite happy to say something (forgive me) is bullshit if it’s bullshit. And I’m not at all happy to just let it go. And that’s . . .

Abbate:

That’s worked for you!

Kahn:

I put it down to being South African. I put it down to being South African. [laughs.] It’s very un-British; it’s very un-British.

Abbate:

Ah, that could be.

Kahn:

Oh yes! It’s very un-British. So, I certainly have a reputation—people assume that I eat students alive, or something! [laughs.] Which is completely untrue!

Reflection on Women in Computing

Abbate:

So you don’t feel that being a woman has been an obstacle to you in terms of advancement?

Kahn:

I really, really, really don’t believe so in my case, and I— You can put it down to my being tough or something; I don’t know what you put it down to; maybe you just put it down to the fact that I don’t perceive it as a problem. So maybe it has, and I haven’t noticed because I’ve chosen not to perceive it as a problem.

I think that you should be judged on what you do and what you’re able to contribute. Given that’s what I genuinely believe—and I really believe it; I don’t [just] say it, I really believe it—it has really not been an issue; and I ought to worry, but I don’t. I certainly would be quite happy in every environment, and I don’t mind if it’s all women or men or whatever. If I want to say something, I will say it.

Abbate:

But the department’s been a good . . .

Kahn:

Oh, maybe they’re scared of me; I don’t know! [laughs.] I don’t know! I think in general people have let me get on. As long as I don’t make a pig’s ear of things, that’s okay! You know the phrase “pig’s ear?” Mess!

Abbate:

Have you had any difficulty balancing your career and family responsibilities?

Kahn:

I’ve been very, very, very fortunate. We have one daughter. And the . . . When we decided that we would have a child, we actually asked the university—asked the head of department—if we could both go two-thirds time. This was my husband’s idea—and a very good idea, as the theory being that we would share the looking-after of the child and not demand full pay from the department. And the then-head of department, Tom Kilburn, said, “Well, are you going to be doing two-thirds of a job?” And we said, “No, no, no, we just realized that because of peculiar time-keeping, we could inconvenience people.” And he said, “No, if you’re actually going to do a full job, then you should be paid a full salary. If it turns out you can’t do that, then we’ll reconsider it.” So they allowed us to work flexi-hours, long before flexi-hours were invented.

Abbate:

When was this?

Kahn:

My daughter was born in 1977.

So we worked flexi-hours—half of each day each, every night and every weekend—and that way we kept the jobs going. I have a huge admin load and always have had; that was fine. And I was at work the day my daughter was born, so that didn’t inconvenience anybody too much! [laughs]

Abbate:

Wow!

Kahn:

Well, it’s not far to the hospital from here! [laughs.] I only survived until about twelve o’clock or twelve-thirty, then I had to go off to the hospital.

So, it was nice writing in and asking for my maternity leave to start the day after my daughter was born. And the department was very, very good about it. We actually kept a playpen in the photocopying room in there, and when we had to both be at a meeting, my daughter got dumped in there! [laughs] —with the nice ladies who ran the print room. [laughs.] And the department was excellent about it, absolutely excellent! I have no complaints whatsoever. And I think they would argue that they got benefit as well, because we didn’t—we still computed all—I still gave all my lectures. I was marking exams the week after she was born. I was sitting at four o’clock in the morning and feeding a baby and marking exams, and it’s just one of those things you do if you have to.

Abbate:

What does your daughter do?

Kahn:

She’s a biologist. She’s doing a Ph.D. at the moment. And one hundred percent doesn’t want to have anything to do with computers! [laughs]

Abbate:

When both her parents . . . [laughing]

Kahn:

You can’t blame her at all! But she uses computers, so that’s all right.

More on the Department at Manchester

Abbate:

Did you have any role models or mentors who encouraged you?

Kahn:

No. I mean, I wouldn’t have known what to do with one, I don’t think! [laughter.]

The department was, from my point of view, an absolutely excellent place to be, because I didn’t feel pressurized by things. I really didn’t. Now, maybe that’s purely because I wasn’t looking for it. I don’t understand why I don’t have any of the same hang-ups that other people seem, or problems that other people seem to have had. So either the department was really very good to me, or I didn’t observe people being horrible, so it didn’t really make much difference.

Abbate:

But in terms of—if you needed assistance, you could find it?

Kahn:

Yes! But I would go to whoever. I certainly was not—it was not gender related. If I wanted . . .

Abbate:

I didn’t mean necessarily [gender-related mentoring].

Kahn:

But if I wanted . . . I mean the . . . I got on . . . Tom Kilburn’s second-in-charge, as I say, was Dai Edwards; and Dai Edwards then took over when Tom retired in ‘81. And although Dai Edwards is a very fiery Welshman, and he does lose his temper like that [snaps fingers], I got on with him really really well, mostly because when he didn’t say anything sensible, I was quite happy to tell him it wasn’t sensible. I don’t think I—I really didn’t do it because I wanted to establish a point; it was part of the same thing, that if it isn’t sensible, then you should say so. And I was rather pleased that when he retired, he actually thanked me for the honesty with which I had talked with him over the years! [laughter.]

From my point of view, he was probably the—he was a very strong person, and he was probably the person . . . I got told off once by somebody in the university—saying, “Oh my god, you were trained by Dai Edwards, weren’t you?”—for arguing about something in the same way that he would have done. And I’m happy with that! If you have an opinion you should . . . You shouldn’t hurt people, but you should certainly state what you think.

So he was probably I think the closest. And also I’ve been very fortunate—again, this will sound sloppy—but my husband and I are both—are friends, and he’s been an enormous help to me. Particularly, I was organizing the celebrations for the fiftieth anniversary, and that was very very hard work in ‘98; for a year essentially I did almost nothing else but work twenty-four hour, twenty hours a day, planning all these huge celebrations we had. And it was very helpful having somebody around who A) understood what I was doing, but B) could give a technical opinion; and I still rely heavily on the technical opinion of “Well, you know, it won’t come across very well if you do it that way”; or “No, you can’t say that”; “The wording of that needs to be changed or something, because it isn’t quite right.” That helps; that helps enormously.

Abbate:

Did you meet him here?

Kahn:

Yes, yes. Boring, huh? [laughs] “Boring!” [laughs] We both like cricket. [laughs]

Abbate:

What’s your husband’s name?

Kahn:

Brian Napper.

Abbate:

Brian Napper.

Kahn:

Right. So half the time I have to remember that I’m Mrs. Napper.

Abbate:

Are you?

Kahn:

Oh yes! When I go home I’m Mrs. Napper; when I’m at work I’m Kahn. It was his idea, not mine, because he thought it would confuse the students to have two people called Napper. Instead it didn’t confuse the students; it horrified the students that Miss Kahn was pregnant! It was so funny! They were really, really embarrassed. [laughs.] In the ‘70s, yes? Really funny! They’d look anywhere, like this! [mimics student avoiding looking at pregnant professor.] Ah, times have changed!

Abbate:

Well, it was a probably a good lesson for them.

Kahn:

It was interesting! Because most of them didn’t know that we were married; why would they? We didn’t have the same name. [laughs.] That was the theory, anyway: not to confuse the students.

Abbate:

To keep them guessing!

Kahn:

We just didn’t think they were bright enough to cope with two people with the same surname, they wouldn’t get the—they wouldn’t hand in their work to the right one.

Abbate:

Well, if you were both Dr. Napper . . .

Kahn:

No, I don’t have a Ph.D.! I cheated; I didn’t get a Ph.D.. Which was, again, an argument I had with the university. In order to get a Ph.D. I would have needed to start . . . I was . . . By the time . . . I basically was supervising a lot of students, including Ph.D. students; but if you supervise Ph.D. students, you cannot also be a candidate for the degree. And when they said to me, “You really, really, really must get a Ph.D.; otherwise you’ll never get promotion; it’ll be the end of the world, blah, blah, blah”—they then said “You have to get rid of all the Ph.D. students until you’ve got your Ph.D.” And I said I didn’t want to do that, because we were doing this interesting work, et cetera et cetera. So—I don’t have a Ph.D. But I have supervised lots and lots of Ph.D. students. Successfully!

Abbate:

Well, it doesn’t sound like that was a disadvantage to you.

Kahn:

I don’t think so. It was, but . . . I’m sure it will have delayed something or other, but I can’t say that I’m too bothered. I enjoy very much working with the good students, and I would have hated to have parted with them for a period of two or three years while I dutifully wrote up a Ph.D.—to what end? [pause.] Don’t tell my daughter! [laughs.]

Abbate:

Is she getting one now?

Kahn:

She’s studying for her Ph.D. at the moment, yes.

On Chartered Engineers

Abbate:

[Remembering something from earlier in the conversation.] What was I going to ask you? Oh, the “Chartered Engineer”: I don’t really understand . . . As far as computer science is concerned, what is the significance of being a Chartered Engineer?

Kahn:

It’s more to do with legal protection, I think, than actual genuine qualification. If you’re a Chartered Engineer—there are certain jobs, particularly in the European context, that you need to be a Chartered Engineer in order to get those jobs. So it is a measure of a certain degree of qualification. And it matters for some of our students; it ought to matter, if they want to move around in Europe.

Abbate:

So even on the software side?

Kahn:

Yes, yes, yes. And it’s to do with understanding. It’s to do with understanding legal responsibilities as well as good practice and so on.

It’s one of the reasons—it’s not the only reason, but one of the reasons that we do have a—we are obliged to put into our undergraduate courses and our taught Master’s courses stuff on legal issues, on company issues, on accounting issues, so that they get a better—a less “techie”—view of what it is to be a computer scientist.

Abbate:

That’s not bad.

Kahn:

It isn’t bad. It just doesn’t come naturally to an awful lot of us. The majority of the people here are here because they are really techie computer scientists, to be honest! [laughs.] Right?

Abbate:

I think that’s true most places.

Kahn:

It is. Yes it is. Okay. So we find these quite difficult to go along with, but it’s—from the bigger perspective, I think it’s the correct thing.

Abbate:

So is every student who graduates also automatically [a Chartered Engineer]?

Kahn:

No, I don’t think so. They certainly take a course that is accredited by the BCS. I think they may have to apply for the Chartered Engineer status. Very few of the staff members have —which we should.

Abbate:

I didn’t realize it was that common, in computer science.

Kahn:

There’s no reason why the staff members shouldn’t apply; it’s just that, why would they bother? Because, as I said, the main reason for wanting it is to work in Europe.

Abbate:

So if they want to work in the U.K., they don’t need that?

Kahn:

Not really.

Abbate:

Ah, okay.

Kahn:

Maybe some companies require it. But I think the Germans are much stronger on it, and possibly the French as well. I can’t imagine the Italians care for it; I don’t know.

Abbate:

All right. That makes more sense . . .

Kahn:

Have other people mentioned it to you?

Abbate:

A few people had mentioned it, but I didn’t—I hadn’t gotten the impression that it was that much of a necessity to have . . .

Kahn:

It’s not.

Abbate:

. . . and I wasn’t sure why one would have it; but I guess it’s the European context.

Kahn:

In the European context it’s useful, and I think it also helps the students from abroad in general, because it’s another thing you can say: “Chartered Engineer”—sounds a lot better than, “I don’t know what I’m doing, but I’ve got a degree,” you know! [laughs.]

Abbate:

To have those initials after your name.

Kahn:

So that must help in some context.

Reflection on Working with Computers

Abbate:

What have you found to be the most satisfying aspects of working with computers?

Kahn:

Wow! [pauses]

Doing things that people use. In the end, that’s what I like doing: doing good things that people use, not garbage that people use. And that remains the most interesting thing—persuading people that they shouldn’t be doing things the cheap and nasty way, they should try and do it properly; and proving to them that in the long run, it’s more effective. You know, I like the practical nature of it. But I’m not a techie anymore. I’m not allowed; I haven’t written a program in years. It’s so sad! [laughs.] I’m just a manager.

Abbate:

Are there programs—the stuff you did when you were working with industry: Are there things that became commercial products that you worked on?

Kahn:

It’s less like that, because it’s mostly infrastructure, rather than things that they can sell on themselves. We ourselves do supply software to the industry, for not much money, which they incorporate in tools—I’ve done a bit of that. But it’s more to do with infrastructure.

Having said that: We’ve currently got a project that’s about to start, which has supposedly got a wonderfully clear route to exploitation, with a venture capitalist involved and things—and I’m really not sure I’m that confident! [laughter.] Because while I’m very happy that the people should use it, there’s an awful lot that comes with the need to talk to venture capitalists that—things which I’m not sure is my scene. However, it’s all about learning, yes?

Abbate:

How do you think the field of computing has changed since you started?

Kahn:

Out of all recognition! [laughs]

It’s the endemic nature of computers [that] is just amazing. It used to amuse me that Tom Kilburn would say he didn’t have a computer—which he didn’t; he was given one, but he must have given it to his son. But of course the video player has a computer in it; everything has a computer in it.

The nicest thing, I think—the most entertaining and the most beneficial, from my point of view, thing that’s happened is the Web. I think it’s . . . Five years ago I really wasn’t that interested, because there wasn’t that much to find; but it’s so huge now, it’s just like having . . . I love encyclopedias and books, and it’s like having this spread before you, and you can look for almost anything, and do almost anything—and then of course try to exercise judgment on what you’ve been given back, to see if you can tell the good from the bad. Just like encyclopedias, and reading, and things. The endemic, the completely endemic nature of computing is amazing. It’s very difficult to see—if one could ever step back now . . .

But I don’t do anything exciting with computers myself, other than to look on the Web for the things I want, and write lots of documents, because I’m a manager really. So that’s it.

Advice for Other Women in Computing

Abbate:

Do you have any advice for young women starting out in computing, or thinking of going into it?

Kahn:

Same advice I’d give to anybody: Do what you want to do! Not what anybody else tells you is good for a job or anything else. Just do what you want to do!

Abbate:

Do you think that’s a problem with students today, that they’re [too career-oriented]?

Kahn:

Oh yes! I don’t think it’s particularly true of women, but it certainly is true of students. It’s perceived as a route to a good job, and we have too many students who are not natural computer scientists or computer people, who are really not happy with the subject. I’m not sure what they would be happy with, because I’m not competent to judge, but I think they are doing it because it’s seen as a good job prospect.

Abbate:

So you’d rather they do it because they . . .

Kahn:

. . . enjoy it!

Abbate:

. . . enjoy it.

Kahn:

Yes. Enjoy it. More than games. That’s fun as well, I’m sure, but it . . .You should do things because you enjoy them and you think you’re going to continue enjoying them, so you don’t mind getting up in the morning and going to work. I bet that’s true for you! [laughs.] I think it just should be what people do! And maybe I’m very fortunate; maybe everybody can’t fall into that category. Maybe you have to do jobs you don’t want to do. But if you have the opportunity, then do what you will enjoy doing—whatever it is.

[End of interview]