Computer Oral History Collection, 1969-1973, 1977
For additional information, contact the Archives Center at 202.633.3270 or archivescenter@si.edu
Interviewee: Grace Murray Hopper (1906-1992)
Interviewers: Beth Luebbert and Henry Tropp
Date:
July 5, 1972
Repository: Archives Center, National Museum of American History
LUEBERT:
I was interested in seeing you for one major reason and that was in finding out how you
became the one who authored the articles for the journals with Aiken and were
responsible, and he says (in a forward manual?).
HOPPER:
One day he walked up beside my desk and he said, you're going to write a book. And I
said, I can't write a book, I've never written a book. And he said, you're in the Navy now,
and I wrote a book. I wrote about five pages a day which I had to read to him at the end
of the day. If he rejected them I had to start them over again. One section in there, on
division, I've forgotten how much time I spent on it. No, that's how I came to do it.
TROPP:
But Aiken just said, you're going to write a book then?
HOPPER:
I think possibly in writing reports and things I had been a little more, I'd written a little
better English. I had been brought up to write things and I'd had good training (in writing
things?) and I'd always bugged all my students about writing things. When they came
into my probability course, the first thing I did was give them a lecture on Sterling's
formula and then asked them to write it up as a ( ?) and them I'd cover it up
with ink and I would get a rebellion that they were taking a math course not an English
course. Then I would explain, it was no use trying to learn math or learn anything about it
unless they could communicate with other people and they should be able to write things
clearly. So I have always been a little bit of a something or other about getting things
written clearly even back when I was teaching, and I taught thirteen years before I went
in the Navy.
TROPP:
Along with your ( ?) on geometry…
HOPPER:
Computer Oral History Collection, 1969-1973, 1977
2
Grace Murray Hopper Interview, July 5, 1972, Archives Center, National Museum of American History
So it was possible in writing up things he had noticed that my, that I had written a little
more fluently and a little better English or something. But that's the way it was
announced to me, he came up beside my desk and said you're going to write a book. And
I did object and he did say, you're in the Navy now.
LUBERT:
And that's the same way for the articles, which were just a condensation essentially.
HOPPER:
Yes. When I wrote those, he didn't ( ?). because they were jointly authored
( ?).
LUBERT:
I was noticing when I wrote my paper on Babbage and Aiken that there was a lot more on
Babbage and a greater perception on the analytical engine for what it was instead of the
higher power difference engine as Aiken put in his '37 paper.
HOPPER:
I put that in.
LUBERT:
You put that in.
HOPPER:
I wrote all that. He did not write …(voice fades out).
TROPP:
I think I'd like to get on tape the comment you made at lunch about when Howard Aiken
really discovered Babbage.
HOPPER:
Well that's the best of my understanding of reading of him and the way he talked. I can't
be, I can't prove it. But I'm quite sure that he discovered Babbage well after he had the
concept of the engine and that he used it as a selling thing that made it more legitimate. It
was a part of a development of a sequence of developments, that made it more
respectable and it would be easier to get money for it. But I'm quite sure it was his
concept, independent of Babbage's. He could use the Babbage later …(voice fades out).
For additional information, contact the Archives Center at 202.633.3270 or archivescenter@si.edu
Computer Oral History Collection, 1969-1973, 1977
3
Grace Murray Hopper Interview, July 5, 1972, Archives Center, National Museum of American History
TROPP:
Let everything fit?
HOPPER:
Yes. Somebody probably said to him, have you ever seen Babbage's stuff? And knowing
Aiken he'd go find out what it was all about and then he'd …(voice fades out). But I'M
quite sure his concept was independent.
TROPP:
I'm also interested in repeating your comment about the Mark I as a communications
instrument in terms of its overall structure.
HOPPER:
Well that's the way I always thought of it as a whole bunch of different engines which
would ( ?). And I think it's a concept which we are coming around to again
today and essentially it was. Those ( ?) were independent ( ?) and if you notice
them in sign and co-function and cosign or whatever it was, in log and those functions,
those were totally independent functions independently programmed by those step
counters and you simply sent something over to it and they went to work and delivered it
back again.
TROPP:
That's where I, of course, got the impression along with other things as I looked at the
hand written programs of the duplicating, essentially I as a mathematician would do, is if
I needed an expediential function I would go to the table. If I had to interpolate, then the
next thing I would do would be to interpolate and then come back to my problem. And
that's essentially what the machine did. It called the functions, did the interpolation, and
then came back and…
HOPPER:
No, the functions were put in there particularly because of storing the constants, was the
real reason they were built, in rather then. Of course, they were never used, after a very
short space of time. Because they persisted, you see, inn delivering, for instance, with the
cosign, it didn't care what angle you put in, it insisted that the angle must be somewhere
between minus and ( ?) and reduced every angle. Well I had a
problem which did that with the fire control on board ships, so I had to produce the roll of
a ship and I could not tell that computer that I knew my angles didn't go beyond 90
degrees, but I tried, I knew that ship didn't ( ?). (LAUGHTER). It
persisted in assuming that those angles could be anywhere between minus and
For additional information, contact the Archives Center at 202.633.3270 or archivescenter@si.edu