Oral-History:Kathy Humphry
About Kathy Humphry
Kathy Humphry was born in Liverpool in 1944. She attended the University of Saint Andrews and became involved with computers right after received her degree. After graduation, Humphry worked for Standard Telephones and Cables as a telephone switching technician. Afterward, she moved to the USA to work on the same machine at Blue Cross/Blue Shield. Humphry eventually moved back to Britain to work in the university system, first at Open University and then Edinburgh University. At Edinburgh University, as a Computing Officer she was involved in using the university's own high-level language, Implementation Language (IMP). At the time of this interview, Humphry was working at Xilinx.
In this interview, Humphry describes her interest in math growing up in Liverpool and her time at the University of Saint Andrews, where she earned a degree in psychology and economics. After graduations, Humphry worked for Standard Telephones and Cables as a telephone switching technician. She goes into great detail in this interview about her job in telephone switching, working with computers, and her experience in the field as a woman. Humphry also talks about her job in message switching at Blue Cross/Blue Shield in the USA. She also talks about her time at Open University where she did front-end work onto the university’s mainstream computers, linking them up so that the computers that the students were using for their studies could in some way be linked up to the main university computers. In this interview, Humphry also reflects on her career and her work with computers.
About the Interview
KATHY HUMPHRY: An Interview Conducted by Janet Abbate for the IEEE History Center, 17 September 2001
Interview # 603 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, IEEE History Center, 445 Hoes Lane, Piscataway, NJ 08854 USA or ieee-history@ieee.org. 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:
Kathy Humphry, an oral history conducted in 2001 by Janet Abbate, IEEE History Center, Piscataway, NJ, USA.
Interview
INTERVIEW: Kathy Humphry
INTERVIEWER: Janet Abbate
DATE:17 September 2001
PLACE: Kathy Humphry's office at Xilinx
Background and Education
Abbate:
This is an interview with Kathy Humphry on September 17th, 2001.
To start at the beginning, can you tell me when you were born, and where you grew up?
Humphry:
Right. Fifth of July, 1944, outside Liverpool.
Abbate:
What did your parents do for a living?
Humphry:
My father was a Managing Director of a plywood manufacturing company —or he built a plywood manufacturing company outside Liverpool, with government aid; because Liverpool was a Development Area. He was highly successful during the war, which made him a bit guilty.
Abbate:
Interesting.
Humphry:
Happily, he was able to pay his taxes back.
Abbate:
Did your mother work at all?
Humphry:
She worked for him until she had kids.
My parents were Austrian. They came from Vienna in 1938, so they were starting afresh in this country. And she could work for him because she could type in German as well as English.
Abbate:
Did you grow up speaking German?
Humphry:
I’m afraid not! My parents spent too much time with each other, and spoke very good English themselves, so they spoke English with us in order to get our attention immediately.
Abbate:
How many children were there?
Humphry:
There was myself and my sister, my younger sister. They had a bit of a gap before children as they waited to see how the war went. They weren’t interned; they were very lucky there. The factory was so important that they were allowed to remain running it, instead of being interned as other aliens were.
Abbate:
What kind of schools did you go to?
Humphry:
I went to private schools. Again, I think because my parents were foreigners, they didn’t know better.
Abbate:
What do you mean?
Humphry:
Well, I don’t send my kids to private schools. I’m very against it, and I think now my parents would be; but in those days just what the first person they met told them to do, they believed. So I went to a private “prep school,” it’s called, for probably seven years: a day school, near to where I lived, and all the kids around me went to the same school. And then to a private boarding school, about an hour from where I lived; that would be for six years.
Abbate:
Were these all-girls’ schools?
Humphry:
Yes, yes.
Abbate:
Were you interested in math or science as a child?
Humphry:
I liked maths. I was always good at maths—not brilliant, just good. So I sailed through maths at my first school, and again at my second school; but I wasn’t very good at other sciences. I was no good at physics, and I didn’t go on to do maths at all at university. I did maths, history, and English for A Level, which is our final exam, used for University Entrance, so that didn’t suit me for a technical degree.
Abbate:
Was that encouraged in your school? Did they have a lot of science?
Humphry:
Maths was—and I think still is—considered both arts and science, so when they’re dividing up the subjects, it’s one of the very few crossover subjects. I chose to do that, but I was the only one, and that caused great time-tabling problems. The only reason I could do it was there was another girl doing maths who was very, very good—far better than me, actually; she’d be really interesting to talk to; her dad was one of the famous first computer scientists—and she didn’t need much help, so I could just sort of fit in and get help from the teacher any time. And apart from her, all the other people who did science didn’t do maths. Nobody did maths, either the scientists or artists.
Abbate:
Really!
Humphry:
I don’t know why! The people taking science mainly thought they were going on to do medicine, and the people doing arts thought that maths was too hard.
Abbate:
So, you must have been pretty motivated to have taken it?
Humphry:
I had a brilliant teacher: old lady, a spinster lady, who would give me lessons by myself for three hours on a Saturday night. She was dedicated.
Abbate:
Did your parents encourage you?
Humphry:
Yes. My father was very interested in maths, and sort of hoped that I would follow him. He died when I was fifteen. Before he died, he said “You should be a computer programmer when you grow up.”
Abbate:
Really?
Humphry:
And that was quite early; I mean, he died in 1960. My sister and I said, “Oh, Daddy, that’s stupid! You have to be clever to do that,” and dismissed him. But he was right!
Abbate:
Now, why did he think that? Did he use computers in his business?
Humphry:
No. I think he was very far-sighted. With hindsight, I think that.
Abbate:
Had you even seen a computer, at that point?
Humphry:
No, not at all. I didn’t know what it was or did. I knew nothing about it.
Abbate:
Did your parents encourage you to have a career of some kind? Was that expected?
Humphry:
Yes. I think probably what was expected of me was to follow him into the business—being the eldest, and there being no sons. If there had been a son, the heat would have been off; but there wasn’t.
Abbate:
So you were going to go into the plywood business.
Humphry:
I did not want to, and I’m glad I didn’t have to, but I think that would have been expected.
Abbate:
So what happened instead?
Humphry:
Well, he actually had stopped the business by then, and then he died. And I just went and did a really wussy, “female” degree: psychology and economics; social sciences. Pathetic, when I think back on it! I went to the University of Saint Andrews and did it there, and I realized while I was there that really, anything to do with maths was where I was best at. So I did pretty well in psychology, because I wasn’t scared of the hand-held calculators: you had to press little things down with your fingers—they really hurt—and then wind the thing; everybody else was terrified of them. So all the experimental work was easy for me. I told the Careers Officer that, and he sent me job specs for applications, and I just noticed the names on all these jobs were ICL (ICT, it was then), IBM, STC [Standard Telephone and Cable], Plessey: they’re all computer companies. And I asked him, “Why are you sending me to them? Am I going to do personnel, or as a programmer?” And he said, “Try programming. They do aptitude tests. See how you get on.” So with this entirely dotty degree—boring, uninteresting degree—I went for these aptitude tests and did fine, and got accepted by all these companies. This was 1966, and I had seen a computer by then; an engineering colleague at the university had taken me in to see a Stantec Zebra: an enormous thing!
Abbate:
I’ve never even heard of that.
Humphry:
This was a fabulous machine! I knew nothing about it; for me it was just a big box that took up a whole room, and you had to sort of slide down the sides of the room because it was so big—absolutely terrifying thing! Afterwards I heard that it was designed by a Dutch man who was both blind and deaf, and he was one of the few people who could actually run the thing. There may be one around somewhere still, in a museum; it is a fabulous machine. But I never got closer to it than squeezing past it one day with this guy, and that was it.
Using Computers at Standard Telephones and Cables
Abbate:
Were you interested in it?
Humphry:
I was, yes!
So then I got all these jobs, and I could take whichever one I wanted. Life was very easy then: come out of university with a boring degree, and the jobs fell into your lap! We were really lucky. I went to Standard Telephones and Cables, which is a subsidiary of ITT (well, it was then; I’ve no idea what’s happened since), and for the first time ever was confronted with computers. I was absolutely terrified! I didn’t know what it trying to do; you know, I didn’t see any cause and effect. But they taught us. We were called “the graduates.”
[recording pauses]
Humphry:
So where was I? Oh yes, right: I’d just started at this place, and there were ten of us, and we were called “the graduates.” We weren’t all women: four women, I think, out of ten.
Abbate:
Ten new hires?
Humphry:
Ten new hires. And it was very funny, because I had spent my whole life trying to become a graduate—I thought this was my ultimate ambition—and when we got there, that term was used as a derogatory term: in other words, “the new boys.” But they put us into three groups, each with one or two women and two fellows, so that we were split up like that. It was very, very interesting, and we soon got the hang of what was going on. Two groups were working in a message-switching group, and the group that I was in was telephone switching. It was really interesting. We were putting in one of the first digital exchanges in London, in Dollis Hill, and this was also pulse-code modulated. I suppose all this is—everybody knows about it nowadays.
Abbate:
But it was pretty advanced at the time, though.
Humphry:
But then it was so exciting. Also, the information, as well as being coded, was stacked and time-sliced. I mean, all that stuff is so obvious now, but then it was great!
Abbate:
“Stacked” is frequency division?
Humphry:
Yes.
Abbate:
I just hadn’t heard the term “stacked.”
Humphry:
No? Well, maybe I shouldn’t have used that term. Each conversation was about 1/25th of the bandwidth, so there’s a lot of switching to be done. We were working on the front end, on a . . . This was such a long time ago; I can’t remember the name of the machine. It may come to me; I’ve lost it right now.
[recording pauses]
Humphry:
One of [the machines we worked on] was really brilliant: it was a PDP-1, the original DEC machine; eighteen bits, great big core banks of memory. The memories were in core, and we discovered the hard way that it got dry joints, so when you had a table lookup and execute instruction—and everything was in assembly language, so I’m not as old as some, obviously [laughs]—it would go the wrong way! So you’d suddenly find that it wasn’t doing what you expected. We discovered it was because of the hardware: it was actually physically dropping a bit on these indirect jumps. So what you would do when that happened was go open the back of the machine, and take your hand and [run it] along the memory banks, and shake them up; and then it would work again!
Abbate:
[laughs.] I have never heard that!
Humphry:
[laughs.] We really did that!
Abbate:
That must have been quite early.
Humphry:
We’re talking about 1968, somewhere about that time. You know, if machines got hot, you would put a fan inside, and we could also pull down blinds against the sun, etc. It was just crazy, but it was wonderful.
And everything was visual, in bits, or sound. We managed to hook up sound to some of our transfers, so that you would walk along and you’d hear the sound of the bits, and you’d recognize a pattern and know, “Well, everything’s going okay.”
Abbate:
Just from hum of the machine?
Humphry:
The transfers, when you were talking across from one machine to another: we had hooked up some sort of sound thing on it, so you just knew that it was the normal chatter between the two machines, and things hadn’t gone violently wrong. And that was linked up to lights. You know how you just see lights and—well, you don’t nowadays, but you did then.
Input was paper tape. I’m very good with the unipunch. I can patch and alter files with these little bits of tape very well. [laughs.]
Abbate:
So you were working in assembler?
Humphry:
Yes, that was assembler, yes; various . . .
Abbate:
And punching it onto cards?
Humphry:
Well, we didn’t punch it ourselves, actually. We used to write it all out, and then someone else punched it up (and would make mistakes), and you would then get a run back, telling you what happened and where it went wrong. So that we must have moved along very slowly.
Abbate:
So there were punch-card . . .
Humphry:
. . . operators, yes.
Abbate:
Probably women?
Humphry:
Yes, always women.
Abbate:
And then operators who ran the machines?
Humphry:
Sometimes they did. In some cases they did, and in other cases we did it ourselves. We could probably get an hour a day ourselves, so that was quite good. Or we could come in nights and weekends, which was excellent—but terrifying, because one weekend I was in all by myself, and when I turned the machine on—doing everything I was meant to, so I thought—it suddenly went POOF! And I thought, “No! Thousands and thousands and hundreds of thousands of pounds I’ve just killed!” But unknown to me, there was a safety trip—and it was that big [holds hands about six inches apart]—and it had flipped with some surge that had occurred.
Abbate:
So they had this huge, six-inch-long circuit breaker?
Humphry:
Exactly. And that’s what happened, but I didn’t find out until Monday morning. I had a horrible weekend!
Abbate:
So you heard a big bang and everything stopped!
Humphry:
And I was baby-sitting for my boss that evening, and I said “Guess what? I’ve just broken everything!” And he just laughed, and on Monday morning we discovered I’d broken nothing at all. Thank goodness!
Abbate:
So could you have just flipped the switch, if you had known?
Humphry:
Yes. Yes, I could have probably started again.
Abbate:
So next time you were in on a weekend, you knew?
Humphry:
Yes. [laughs.] I think I kept clear! I made sure someone was with me after that.
Abbate:
Did you work a lot of nights and weekends?
Humphry:
A bit. I mean, I didn’t have kids or anything, so yes: a bit.
Abbate:
To get the time [on the machine]?
Humphry:
Yes. And the peace.
Abbate:
Well, what was the working environment like? Were you all in the same room?
Humphry:
Well, as I say, there were three different teams, at least, in the group that I worked in. I can’t remember how many employees there were. We were in an office building in the north of London. We were actually hooked up to an office building of ITT, but we never met them at all; even though they were the parent company, they were completely separate from us. So maybe there were thirty or forty of us, and in my group: I don’t know, ten, maybe. All ten of us were in one room—plus, presumably, there were bosses around.
Abbate:
So it was pretty busy?
Humphry:
Yes.
Abbate:
And how was that organized? Was there one person in charge telling people what things you were working on?
Humphry:
Oh yes.
Abbate:
Did you work in teams?
Humphry:
Yes. And we had a couple of guys from Germany who joined with us—who would have come from one of the associated companies to work with us—but they still sat in this big room with us, even though they were more senior. But the person who ran the project was not in the room; he was somewhere else down the building.
Abbate:
What types of things were you working on?
Humphry:
Well, this was telephone switching, as I say; and the other people were working for Reuters on a message switch.
Abbate:
Were you designing this switch?
Humphry:
Just making it work. The switch was designed, but just making the communication bit of it work was our job—the front end.
Abbate:
So how did that translate into programming? Did someone else figure out what it should be doing?
Humphry:
Yes. I have to say, I was very bad—me personally; I’m sure other people in the group weren’t—at seeing the overall picture. So I took my little chunk away, and I could program that and understand what it was meant to do, and understand the interface between that bit and other people’s bits; but I could never stand right back and see the whole picture. I always found that very hard. So I knew that this switch was going to work for the trunk dialing for the whole of London, or half of London, but I didn’t really know too much more than that.
Abbate:
Did it seem important?
Humphry:
Yes. And fun!
Abbate:
Was there some point when you actually made a telephone call and knew it was going over your switch?
Humphry:
No, because I left before that. But another group who were working for Reuters actually picked up the emergency message that Robert Kennedy had been shot. They were doing a live test for the first time, and they got this special code up on their screens, and they thought, “Oh no, our whole system doesn’t work.” And then they read the message, and it gradually dawned on them that it not only worked; it worked very well, and they’d received one of the very first messages into this country.
Abbate:
Was it fun, the work?
Humphry:
Yes. Oh, yes.
Moving to the USA, Working for Blue Cross/Blue Shield
Abbate:
How long did you stay?
Humphry:
I was there two years; and during those two years I got married, and we wanted to go to America just for an adventure—for no good reason at all, just for the sake of it: wanted to travel. My husband had ulcerative colitis, so we couldn’t go to Africa, or anywhere where there was a problem of him getting ill; and so we went to America. He was old enough not to be drafted, thank goodness—just—and so we asked the company to send us there, and they didn’t do it. They may have done eventually, but they hadn’t got around to it, and my boss kept saying to his senior people, “If you don’t take them on, they’ll leave anyway. They’re young and they want to go; they’ll go.” But whilst that happened, I got a job which fitted me perfectly at Blue Cross Association, which is a service bureau for the Blue Cross/Blue Shield plans, and that’s in Chicago. They were going to pay our fare, both of us; and I was just going to fall into a very similar-type job, this time message switching, on the same machine—also front-end. They’d come across to Britain to recruit. There were about six of us they’d picked up; one came from Sweden and the rest of us from Britain.
Abbate:
Why did they come here?
Humphry:
It was at the time of the “brain drain.” You know, we had the graduates, and they wanted them, and they came looking for us. They advertised; we saw the ads; and we turned up. They gave us temporary visas and a return ticket. In fact, in my case I got a permanent visa. We were very lucky: we heard that the green card system was changing, and that it had changed, and it was too late to apply; but we applied anyway, and we were just in time: we were the last month of people who got green cards from this country straight away, without having a job.
So, we went to Chicago, and I was working there. I’ve forgotten [what computer system I worked on]. It was a bit like the IBM machines, but it wasn’t; it had a sort of subset of their operating system, and I can’t remember what it was. My husband just came with me, and he just looked around until he got work, which didn’t take him very long either.
Abbate:
Had you met him at STC?
Humphry:
I met him at university. We were at Saint Andrews together. But he actually did join STC for a wee while. Yes, that’s right.
Abbate:
And he was doing computer things?
Humphry:
Sort of. He had done maths at university and then gone into Operations Research, but he was gradually sliding more and more towards programming as well. And he got a job in the States with Montgomery Ward. He got a series of jobs that seemed pretty boring to me: all data processing. [both laugh.]
Abbate:
But you were working for the Blue Cross.
Humphry:
BCA, yes, and that was great, again. I was very lucky. That cut over live just as I was leaving. So, I got to the final stages of that project, which was super.
Abbate:
And you got to see it go into operation?
Humphry:
Yes. Well, for me, it actually worked quite well. Chicago was where BCA is, and we were in the final stages of testing when I and my husband, Mike, moved to San Francisco, because he had lost a job in Chicago and then got one in San Francisco. I was not going to carry on working for a wee while because we were having our first child, and so I actually—the night that I left to go to San Francisco, we had the first real full-blown test, sort of in parallel with the old system; and we sat up all night monitoring it from our office, and so that was very exciting—but exhausting for me, because I was seven months’ pregnant. So I left work at nine o’clock on a Saturday morning, having worked through the night: very peculiar.
But I’d seen all that. As soon as we got to San Francisco on Monday morning or whenever it was, I went into the local Blue Cross / Blue Shield plan—the California one—where they had a desk for me, and I carried on monitoring the system from the other end, which is very, very useful: to have people doing that. We’ll have had people at other plans, like we’ll have had somebody in the Illinois plan doing it, because it was in the same building, and maybe at one or two others; but, we were such a small group, it was really brilliant to have somebody at an outside plan, monitoring it from the other end. We had communications going in those days. It was exciting, and I could still debug and put changes in—I’d probably have to tell people; I couldn’t probably do it myself from the far end, but I could still pore over the code and think what was going on.
Abbate:
And this was the administrative system for the plan?
Humphry:
Yes. It was the message-switching system, so that patients like myself, who moved from one plan to another—and of course, I was a typical case, because I was about to use it—their information could be passed between the plans, and verified, and taken over and whatever. They did whatever transactions they wanted, and we were the mechanism for it.
Abbate:
So you got to actually see it in use?
Humphry:
No, we were still at the final stages of testing when I left.
Working for Friden, Singer
Abbate:
What did you do after that?
Humphry:
So then I looked at advertisements—because now I’m sort of stranded in San Francisco, looking for a job—and finally got taken on by Singer. Well, actually, no: Friden. Friden made paper tape readers.
Abbate:
Hmm.
Humphry:
Oh, everybody would know Friden! They’re these little—ASR-33 they were called: well, one of the models, the most well-known one. It’s a paper-tape reader, and they had a printer as well, for communicating with computers, and also you could hook modems up to them—possibly later than that. So instead of just entering things into a machine with switches and paper tape, you now had this typewriter, which also included a paper tape reader, and a paper output—so it was beginning to look a bit more like a typewriter. They built that, and almost immediately everybody used those. It was almost a generic term: the way you would say “Hoover,” you would say “Friden”—almost. And then they were bought by Singer.
Abbate:
The sewing machine company?
Humphry:
Sewing machine people. I got a very nice cheap sewing machine as a result of it—which I bought, but at a knock-down price.
They started to develop their own computer, called the Singer System 10—but there may have been other computers as well—and they had other bits of hardware; and this group was now called Singer Business Machines. After I left, they were taken over, I think, by ICL. So I think that’s where they’ve landed up.
Their computer was a very strange thing. We were still in assembly language—I’m now moving into the very early ‘70s, for me—and it had memory partitions that physically could be moved. So, I’m working late at night one night, and I did a run, and there was a mistake. We were still poring over core dumps—still at that stage of debugging; so I pored over the core dump, understood the problem, did it again, pored over the core dump—ooh! Nothing’s working, and I can’t understand why. And after about ten minutes a little guy comes out from behind the back of the machine, and it was an engineer, and he’d taken out my memory while I had been working!
Abbate:
[laughs.]
Humphry:
He hadn’t told me! I guess he didn’t know I was there. So it was an amazing machine that it was partitioned that well that that sort of thing could happen.
It had a very interesting machine code. Now, how did it work? It had ten six-bit words—six bits going like that—but it had ten of those, and the machine instruction went along the top two bits of these six bits.
Abbate:
So sort of vertically . . .
Humphry:
Yes. And then the locations were going along the bottom bit. Yes, you have to draw it. [Draws something like this, with 6 rows and 10 columns across: ]
Abbate:
So six vertically and ten across.
Humphry:
I think it was six down there, yes. And then this lot going along here were the instruction, and that lot going along were the location or address. So it was absolutely extraordinary—and again, great fun.
By 1972, when I left, the idea of higher-level language was just creeping in. People were talking about it there, but I hadn’t met it myself. Of course it wasn’t just creeping in, because people had been using ALGOL 68 and goodness-knows-what elsewhere, but in those places we hadn’t got the hang of it yet. The data processing people were using COBOL and stuff, but for what we were doing, we were still using assembler.
Abbate:
Is that because this was high-speed, real-time stuff?
Humphry:
Probably. We probably felt that it had to be done where we had to have total control. I’m sure that’s what we thought—but we were wrong! [both laugh.]
Moving Back to Britain, Working for Open University
Abbate:
How long were you at Singer?
Humphry:
I was at Singer for about two years. I was in the States for four and a half years, so I was at Singer for two and a half years. And again my husband lost his job, and this time we decided we would come back to Britain. We were sick of this losing jobs; we believed that it didn’t happen in Britain, and we believed it would be nice to bring up our son in Britain. So he came back, and then I came back to join him, and we both got temporary work at the Open University. You’ve heard of the Open University?
Abbate:
Yes.
Humphry:
That was brilliant as well, because it had just started, and the buzz there was terrific. It was the first time you had this distance learning. I worked in their Student Computing Service, so again, I was doing front-end work onto the university’s mainstream computers, to link them up so that the computers that the students were using for their studies could in some way be linked up to the main university computers. In my opinion, a great security risk—but I still enjoyed doing the job.
Abbate:
So the students had terminals somewhere, linked to a computer?
Humphry:
That’s right. That’s right. And that was where we were working, but I personally was working on the join-up between that and—I suppose the Finance Department’s computer. I can’t believe that we did that! Or why! [laughs.] But I had great fun doing it!
[pause to answer phone]
Humphry:
So, right: We decided to come back; we went to the OU; and that was my job; and I can’t remember what machines I was working on. [laughs.] Still assembly language.
Working at Edinburgh University
Abbate:
Because you were doing data transfers between computers?
Humphry:
The actual communications: the telecommunications, the hand-shaking, and all that stuff. And I was very much on my own. That was the best project I did, because for once I actually had to set and do a bit of design myself, and figure out what was going on. My boss, who was a lovely man, didn’t understand what I was doing, but trusted me—goodness knows why! Sometimes I wanted to tell him what I was doing, and I used him as a brick wall when I was making mistakes and needed to sort of figure it out—and it worked, and we got it going.
By then Mike had got a job here at Edinburgh University, and I did too, and so we moved here in 1973. I joined the Computer Science Department at Edinburgh University—and that probably is where you’ll get a lot of the people around here that you’re going to see, because as far as I can tell, it’s the sort of starting point for everything in this area now [i.e., the high-tech start-ups]. Well, it was a good starting point; one of them.
I was in the Computer Science Department for the first time, and that’s where we had people who had never done computer science as students. My boss there had done Greek and Latin in Oxford; that was his background. But that fitted with parsing and languages and compiling; and I was introduced to my first high-level language: IMP. While you’re in Edinburgh you’ll probably hear that many times.
Abbate:
Imp: I-M-P?
Humphry:
I-M-P. “Implementation Language” You probably won’t hear it anywhere else.
Abbate:
I don’t think so.
Humphry:
But you will hear it around here a lot, because it was invented here, [but] it wasn’t well-enough documented, and it never took off. It was just as good as PASCAL any day—better, actually—but it was never publicized. So it’s very insular, and you’ll only hear people here, or people who have been here, talking about it. But it lived a long time, and they built their operating systems in IMP. They claimed that you could write operating systems in high-level language, and they were right!
So, my first day I went on the IMP course. I couldn’t—I mean, I was just shell-shocked; I couldn’t get the hang of it at all.
Abbate:
Really?
Humphry:
Well, you know: How do you do this? How do you do that? What if I wanted to shift left? [laughs.]
Abbate:
You were so used to working very low-level.
Humphry:
Yes, that’s right.
Abbate:
And this is 1973?
Humphry:
Yes. I was a very lowly member of the Computer Science Department—which was very small then, and it was brilliant. It was headed by this person I spoke to you of before, Professor Sidney Michaelson, who had started it in, I think, 1966. It was before my time, anyway. I was called a Demonstrator, because that was all they had; and because I was called a Demonstrator, on my first day here I was told to demonstrate! I was thrown at a PDP-9, I think it was, and told to demonstrate it to the students, who were turning up that afternoon; and I said, “But I’ve never used this machine before!” And my boss said, “Well, you’ve got an hour. There it is!” I was quite unused to that way of working. In industry, you do have your hand—or in those days, anyway, you did have your hand held better than that! I had my hour, and the students turned up, and I said to them, “If I can do it, you can!” [laughs.] And we all managed it! [both laugh.]
Abbate:
So, the people in the department: they were actual Professors?
Humphry:
Lecturers. “Lecturers” we call that here.
Abbate:
So there were Lecturers and Demonstrators?
Humphry:
Yes. The Demonstrator: that was typically a post—not, I mean, in my case; it was wrong, really, for me to have it, but it’s all there was—it was typically jobs that were given to people who were doing Ph.D.’s and needed to have a job as well. The idea was that they would go on to become Lecturers or whatever, and many did.
Abbate:
How long were you a Demonstrator?
Humphry:
Well, eventually they created a post called “Computing Officer,” which was much more related to what I was really doing. There are a lot of Computing Officers there now, and I was one of them, and I was there for fifteen years—and I loved it! We used to actually be allowed to lecture a little bit—we tutored; we lectured—as well as just doing what you’d call sys admin nowadays. So again, there was a great buzz about the place.
But after fifteen years, I wasn’t sure what aging Computing Officers did. I mean, a lot of people would get their Ph.D.’s and go on and do other things, but for those of us who didn’t, I wasn’t so sure. I thought maybe I ought to look for something else, even though I was actually enjoying it; and my boss who had hired me originally started his own company. He asked me to join him at his new company, and I was very flattered, and I went. I thought it was a good opportunity.
Oh: I had been asked a few years earlier to join another company, a very exciting company, by the people who eventually started this one.
Working for Xilinx
Abbate:
“This one” being Xilinx.
Humphry:
Well, the company that Xilinx took over.
Abbate:
Oh, okay.
Humphry:
But when I went to see them I had to confess that I was three months’ pregnant, and that clearly wasn’t going to work. It takes you three months to leave the university; then I could give them three months of my time—waddling around, but still—and then I would be asking for time off. So that wasn’t going to suit them—even if it was only three months, it wasn’t going to suit them—and they withdrew the offer! In fact, they never offered it. They sort of said, “Oh! Well, how interesting.” So that job disappeared. It turned out for me it was quite lucky, because they left that company anyway, and they started another company, which was eventually taken over by Xilinx.
So that opportunity went away, and then several years later, Hamish Dewar asked me if I would join him at his company, and I went, thinking that I might not get another chance to do something exciting like that. It was boring. They designed firmware for laser printer controllers, and there were three of us programmers who did the testing. Testing is a thankless task: if you find errors the designers have to make changes, and if you don’t, you are not doing your job properly.
Abbate:
[laughs.] I’ve never heard that definition before!
Humphry:
Well, it’s actually true, because when you find errors, it means they’ve got to go back to the drawing board, so they always feel grumpy. It’s just the nature of the job.
After about six months I felt trapped, but I actually stayed on because the two partners were friends of mine. I eventually moved on when they warned us that there wasn’t enough new work coming in to sustain all of us. My first reaction was, “Oh, I hope they don’t mean me!” And my second reaction was, “Whoopee! This is my chance!” [laughs.]
At exactly that moment, Xilinx took over a company called Algotronix, which was struggling for finance and had had a brilliant idea in FPGA design. Xilinx took them over, moved into this building, and told them to expand. I joined, but I’m not an electronics engineer, and I can’t design chips.
Abbate:
Can I ask you what “FPGA” stands for?
Humphry:
“Field Programmable Gate Array.” It’s programmable hardware.
Abbate:
So a chip that you can sort of program . . .
Humphry:
. . . on the run—yes. You’ll probably get more of that from Jane [Hesketh] and other people here.
The particular design that we were working on then, or this group was, had been dreamt up as a Ph.D. for one of the original people who started this. The other two were his supervisors.
Abbate:
At Edinburgh?
Humphry:
At Edinburgh University. And then they left there, and started this wee company, because they thought this really was a going concern. It was, but they couldn’t get enough funding, and Xilinx, who was a sort of rival, bought them up—which was great, for a while. That’s when they expanded, and when we moved in here. We’re not working on that exact chip anymore, but we’re still working for Xilinx. Well, some of us are! [laughs.]
But, sadly—and this is the bit which is a shame—I, at that point, stopped being technical: because this chip is just way beyond my experience, and they were hiring electronics engineers, who were experienced. Most of them have Ph.D.’s, and most of them have actually worked on exactly this chip or something very similar. So I’m just the dogsbody now.
Abbate:
What do you actually do here?
Humphry:
I’m the admin. I pay the salaries; I order the coffee; and occasionally write a PERL script if Julian will let me; and I write one or two programs when I have the time—but that’s background, unfortunately, because salaries have to come first.
Abbate:
Now, you had mentioned something that implied that there were a lot of high-tech spin-offs from the University?
Humphry:
Yes. There are a lot. I don’t know how many there are. Julian was at one, which was Harlequin, and Beth was at one. There must be about 30 companies that spun off from Edinburgh University; it’s been very successful—some from the Electronics Engineering Department and some from Computer Science, and some people, of course, who came from both.
Working with Students at Edinburgh
Abbate:
And when you were actually—to backtrack a little: when you were at the university, what sort of projects were you working on?
Humphry:
Well, basically it was sys admin for the students.
Abbate:
So were you creating new software to handle that?
Humphry:
Yes, really just keeping machines going, doing backups, and adding sort of packages so that students could do practicals without the nitty-gritty—which I thought was a shame, because the nitty-gritty is the bit they like doing!
Abbate:
Did it give you a chance to learn new software and things?
Humphry:
Yes. Anything we were showing the students, of course, we had to learn ourselves.
Abbate:
And you had a lot of contact with students?
Humphry:
Yes. Yes, at that time we did. I’m not sure that’s so true now.
Abbate:
Were there a lot of women in the Computer Science Department?
Humphry:
No. There were very few, and I’m sure some are on your list. They must be. Have you been sent to see Rosemary Candlin?
Abbate:
I’m seeing her tomorrow or the day after.
Humphry:
Well, she was the main one. She’s about ten or fifteen years older than me. There must have been a handful of us in the department—five or six, I can’t remember. And again, with the students: the first year I joined, the fourth-year students were all men. Of third-year students, there seventeen people, of which five were women. (One of those women became my boss here. She’s one of the original people who started this place.) Following year, something similar; and then after that, it got worse again, and the women started to disappear again.
Abbate:
And this was the late ‘70s or something?
Humphry:
Yes—early 1970s. 1973 is when I first started there, so fourth-year were all men. I mean, I can remember the start of the year: when students signed on, they’d sign on as computer science, and they’d look around, and the girls see that there were only one or two of them, and go off and do maths instead. I don’t know if that’s happening now.
We used to care about it, and try and do things about it. We used to send women out into secondary schools to tell women that it was a perfectly okay career. We held a Women in Computing Day for several years, and they came, and they played on the machines, and they enjoyed it, because they weren’t being elbowed off by the fellows, but they didn’t seem to come back in droves to do the subject.
Balancing Family Life and Career
Abbate:
But the ones who were in the department . . .?
Humphry:
They were doing fine. They did fine. And it was a very flexible department, so you could work part-time if you felt that was necessary, as I did. But when we were running a Women in Computing Day, and a school kid, who was sort of sixth-year, stands up and says, “How does it work with having kids?”, I was unable to answer that, because those answers are still individual. There’s no formal mechanism for coping. It’s still informal, depending on how good your own colleagues are.
Abbate:
How much of an issue was it for you, balancing work and family?
Humphry:
I went part-time, and as a result I’m an administrator and not a computer scientist! I’m sure it’s had a major impact.
Abbate:
Do you think you would have pursued more education?
Humphry:
I think so. I like to think so. Yes. So I still think it’s a problem.
There’s no real mechanisms. You get your child into a nursery, but a typical state nursery finishes at two: they call that “full day.” So at two o’clock, you’ve got to start finding some other arrangement. If your child’s sick, there’s no official arrangements for that, not like in other countries. I don’t know what it’s like in the States.
Abbate:
Well, it’s a patchwork of . . .
Humphry:
It used to be great in Scandinavia, but I’m told that even there it’s beginning to fizzle out again. I mean, they used to send somebody around if your kid was sick. My husband, who’s the best father/mother/wife you could imagine: when the last child came, he’d say, “Oh that’s okay; I’ll look after that one.” [But] when it came to making arrangements, he’d forgotten. It was always back to me. If I asked for time off, it was socially acceptable, but if he asked for time off it was not socially acceptable.
Abbate:
So even if he wanted to, it was [difficult].
Humphry:
That’s right. He was frowned upon.
Abbate:
Do you think that was any worse in computing than in other fields?
Humphry:
No. No, I don’t. It might be even better in computing than in other fields. I don’t know.
Abbate:
Was it possible to work at home?
Humphry:
Yes, I could. Even in 1975, I had one of these ASR-33’s at home, hooked up over a network with a modem. We live in the country, so we actually shared the phone line with our neighbors—so I’d be working away and my neighbors couldn’t get on the phone! [laughs.] But that gave me quite a kick, that I could do that; I could talk to the Open University, and I could talk to Edinburgh University. I had managed to make a muddle of some work, losing back-up data when a machine became faulty, very shortly before I was to have my next child, and I didn’t dare not put it right, because I only had a day or two left. I retyped the lost data in from hard-copies, and I spent that day or two, including the evenings, at home just typing-typing-typing, and then I could send it straight off. It gave me quite a kick that I could do that, and that the listing would be waiting for me at the other end, which it was.
Abbate:
Did you ever feel, other than the time commitment [for raising children], that as a woman you had less access to training or promotions?
Humphry:
Personally I didn’t feel that. I felt that any constraints were put on by myself, in order to juggle my time to be where I had to be with the kids. But I think the opportunities would have been there for me otherwise.
Abbate:
Did you have any mentors or role models?
Humphry:
No. No, I was very much always in a male society, really.
Abbate:
I’m trying to remember . . . Your first job, it sounded like maybe a third or a quarter of the people there were women?
Humphry:
Yes. I think there were four out of ten, in that intake. Yes.
Abbate:
And was that similar when you were in California?
Humphry:
Yes. Yes, it was about that.
Abbate:
Were there any women in senior positions?
Humphry:
Yes, when I was at Blue Cross there was one senior woman, who was older than us by about ten years. In California, with Singer, I don’t think there were. There were one or two women, one of whom was my friend, who was definitely in a very inferior post to what she was capable of, who had taken a post much lower than she needed, just to get the job. I don’t know why that should be.
Abbate:
Did you notice any difference between the U.S. and the U.K. in terms of the culture of the computing world?
Humphry:
I don’t think so. I think it was all very similar at that time. Yes. And people moving about quite a bit, as they are now.
Abbate:
You mean, between the two countries?
Humphry:
Yes, yes.
No, I didn’t notice a lot of difference. I don’t know if you’ve spoken to anyone else about that.
Abbate:
I’ve only talked to one other person who worked in both countries.
Humphry:
Oh, well some of the people that are here at Edinburgh will have worked in both. Have you got Irene Buchanan on your list?
Abbate:
I don’t.
Humphry:
She’s one of the people who started Algotronix, which was then taken over by Xilinx. She was a third-year student when I joined the Department of Computer Science, and she was in Caltech for a while, so she’s been both ways. I think part of her Ph.D. was at Caltech. I’ll give you her number.
Reflection on Working with Computers
Abbate:
What have you found to be the most satisfying aspects of working with computers?
Humphry:
Getting it right! Knowing that you’ve got a problem and that, even though it seems insoluble, it will be solved—and then solving it.
Abbate:
So you enjoy problem solving?
Humphry:
Yes.
Abbate:
What strikes you the most about the way computing has changed since you started?
Humphry:
Because I’m not very good at having an overall vision, it’s much harder for me now, because we do much more top-down stuff now, because of all the facilities we have with languages and programs that we don’t actually know the insides of.
Abbate:
So you think that it used to be . . .
Humphry:
You just took a tiny little corner.
Abbate:
And now there’s . . .
Humphry:
It’s the people with the vision who really get on.
Abbate:
Is that because there’s so much existing software built up that you’re building systems on top of systems?
Humphry:
Yes! Yes, absolutely.
Abbate:
So it’s really a different experience, different skills.
Humphry:
Yes.
Abbate:
Do you get the sense that computer jobs are more open to women than they used to be? Or less?
Humphry:
I think they’re open to women. I just don’t understand why women don’t come forward for them. I think it’s there. I don’t see there’s any reason why . . .
I think it’s the women who are choosing not to, and when I’ve thought about it, I think one of the reasons is that they actually get switched off at primary school. I actually think that the boys muscle in, and become extremely boring with their computer games from the word “go.” And those women who are more intelligent think, “This is boring.” And they don’t understand that it’s not necessarily just that.
Abbate:
So that would be a change from earlier [times]?
Humphry:
Yes! Because it didn’t exist then. So we thought the boys who were going on about the insides of cars were boring [both laugh], and the computing was still sort of fresh! It was a blank page, so it was sexless. But now it has changed.
Abbate:
So you think it’s become . . .
Humphry:
It’s becoming less—women are taking it up less than at that [time]. You know, there was one mid-point where it was just equal.
Abbate:
Do you have any advice for young women who might be thinking of going into computing?
Humphry:
Oh, yes! If they’re fairly good at it, go for it! It’s as good as anything!
Abbate:
And what does it take to be good at it?
Humphry:
Logical. Not scared of any new bit of equipment; just go for it and try it! You still have these people—and I was one of them—who think, “Oh! If I touch that, everything’ll explode and it’ll all be my fault!”
Abbate:
[pause]. Let me think. Have I left any holes in your [story]?
Humphry:
I don’t think so, no.
On Xilinx
Abbate:
We did jump around a little bit. So, where are things going with Xilinx?
Humphry:
Well, Xilinx is a great company. There’s about twenty-five hundred people worldwide, but it still feels like a sort of community—well, of course, because of email; we feel that we’re all part of one family. We meet our CEO once a year. He comes actually into this building, and seems to know us.
Abbate:
And it’s based here?
Humphry:
No. It’s based in San Jose. That’s where maybe sixteen hundred of the—I’m not sure what the numbers are; the majority are in San Jose. They’ve also got a big site in Boulder, and Ireland; and other than that, they’ve got small sites dotted around. They’ve got sales forces in the Far East, and in various places in Europe, and so on; and one or two small companies like this that have been taken over: ourselves, and another one in Grenoble; another one in New Mexico, I think. And we hope that the company’s going to do well, and expand, and we’ll stick with it!
Abbate:
Do you think the industry has become more international?
Humphry:
Yes, yes. [pause.] Well! You say that, but you see, I worked for ITT as my very first job, a subsidiary of it: that was already international. That company was already dealing with—I mean, it already had offices in Germany, during the war. So it was already crossing political boundaries.
Abbate:
Because of the nature of the system?
Humphry:
Yes.
Abbate:
And you say people were going back and forth [across the Atlantic] to find work?
Humphry:
Well, I don’t know; that was before my time.
Abbate:
Well, people came here [from the US] to recruit, and . . .
Humphry:
Oh, yes.
Abbate:
So maybe that’s been a constant.
Humphry:
I don’t know.
Abbate:
All right. Thank you very much for talking with me!
Humphry:
Well, it’s been a pleasure!
[END OF RECORDING]
