Jan. 18, 1973: pages 1-43

Interview Day 2, Tape 1, Side 1, pages 1-43

Jan. 17th, tape 1, side 1 pages 1 through 43
Jan. 17th, tape 1, side 1, pages 44 through 86
Jan. 17th, tape 1, side 2, pages 87 through 166

Jan. 18, tape 1, side 1, pages 1 through 43

Jan. 18, tape 1, side 1, pages 44 through 55


In 1972, Joseph Desch, Robert Mumma and Donald Eckdahl of NCR were approached by the Smithsonian Institution and asked to donate their recollections to its History of Computing project, now part of the National Museum of American History. Desch and Mumma were interviewed on the 17th and 18th of January, 1973 by Henry Tropp.

page 1

HT: Okay. This is .. the 18th of January and we’re still at NCR in their Patent Office area; Mr. Joseph Desch, Mr. Robert Mumma, and we’re going to continue our discussion from yesterday. And, Mr. Desch just suggested that we talk about the relationship with MIT and I agree. That’s probably a good place to start.

JD: It won’t take very long to –

HT: Right.

JD: to get through their reports because all we have essentially is some correspondence, .. most of which .. .. is of little consequence for your purpose .. because it involved .. fees, and

HT: Right.

JD: contract prices and so forth. And – and then we have their reports of their technical activities

HT: Mhm.

JD: in the field of designing and building vacuum tubes to gerforrn counting operations.

HT: This was the connection you had with Mr. Coleman?

JD: No, no.


page 2

HT: No, not?

JD: No. This – this, this is all commercial that I’m talking about at the moment.

HT: Oh.

JD: None of this is military.

[Recorder off]

HT: While we’re waiting for Mr. Kneiss to dig up the MIT documents, why don’t – why don’t we just talk about some people, like Howard Aiken~ Let’s start with Howard.

JD: Okay.

HT: When did you first get connected with him?

JD: I think that I first became acquainted with him on a visit to .. his computation laboratory at Harvard after the war. He had MARK I. He held some kind of a symposium, and we were invited and Mr. Williams and I went down. We saw MARK I and that’s when I met him. I think Mr. Mumma was also along.

HT: Mhm.

JD: And sort of a friendship developed between .. .. Aiken and Mr. Williams and myself. And we had him come out here and visit and so forth. And.. we made several other trips down there at diffeent times. We even took the president of the


page 3

company down to see – it was either MARK II or MARK III that – well, it was the first vacuum tube job that he was building.

HT: That was MARK III.

JD: Yes, that was a monster, too, that was a big one. In fact, I made a mistake because I scared the president of the company by letting him see it. He couldn’t conceive of his sales department selling anything that – that big, see?

HT: Aha.

JD: So,.. .. that was the beginning of our association with him.

HT: Well, the – the question that came up in the car yesterday was that you and he had discussions .. related to both of your positions in terms of the internally stored program. He had come to it rather late in the development, and you mentioned some discussions. Can you recall how he arrived?

JD: No, I can’t recall his logic, but I did get several lectures from him on the logic of internally stored programmed computers here at Dayton when he’d come to visit us.

HT: Aha.

JD: And I learned quite a bit from him on ..


page 4

HT: Oh, did Aiken discuss with you in any detail how he had finally arrived?

JD: No, he did not. In fact, it never occurred to me to ask him, but he was very voluble on the subject. It gave me a rather broad education in – in the development of logical devices at that time, which were nothing but vacuum tubes,

HT: Mhm.

JD: especially multi-element tubes where you’d use all the e~ements, like in a tetrode or a pentode. The use of various grids, which – which controlled to make a logical device.

HT: Mhm.

JD: So,.. .. instead of using a lot of tubes, we used one tube, but used all the grids to do the control and so forth. I remember that very dtinctly because it didn’t work very well.

HT: Mhm.

JD: But ….. then we didn’t hear anything from him for a while until he showed up one day and – and wanted to see me. And, I went down and got him and he told me he wanted to talk to the – to the president of the company, so I took him up to the executive office and introduced him and – and


page 5

they started conversations, with Mr. Allyn at the time. And they seemed to get along real well and .. they hired him as a consultant. And I think he served in that capacity for .. about two years. I had very little to do with him because it was right at the time when we were purchasing CRC.

HT: Mhm.

JD: And I know that he was involved in judging the efficacy of the design of the computers at CRC, which left me out because I wasn’t doing any work on the – like the 303 .. .. CRC computer or the

RM: 102?

JD: 102. I knew about them, what they were, .but I had nothing to do with the design.

Well, I think that the break came with Aiken when the 304 — I mean the 304 — computer became a reality and the design, the concept, started at Hawthorne, California at our El Sequndo plant. That was the outcome of a set of specifications that Mr. Keenoy had developed and gave a – gave me a copy and he gave Hawthorne a copy and wanted to see who came up with the best computer. Well, Mr. Mumma and I, all we had here was a .. was a


page 6

program called NEAM, National Electronic Accounting Machine — that’s the

HT: Mhm.

JD: abbreviation for it. .. But the specifications asked for things that we couldn’t do with NEAM. I mean, it called for the flexibility of an internally programmed computer.

HT: Mhm.

JD: We would have had to ‘ve been pulling plugboards to change the programs. And so Hawthorne won the contest hands down. In fact, I abdicated in favor of Hawthorne because – that’s one of the reasons why I wanted to buy them in the first place was to get the internally programmed computer.

HT: Bob, do you remember Howard Aiken at all?

RM: Oh, yes. I was along on that trip to see MARK I. I remember that trip up there. And then, of course, I remember when he carne down to Dayton here.

HT: What was your reaction to MARK I when you saw it?

RM: Well, of course, it was a tremendously big machine, but it was basically IBM — I think IBM built it for him.

HT: Yes, yes.


page 7

RM: It was full of IBM plugboards.

HT: Right. It was all IBM built.

RM: Yes, relays and so forth.

HT: Howard Aiken wrote a report, or a memo I guess, in 1937, in which he outlined what a sequential calculator ought to do, what its characteristics ought to be, what it ought to be able to accomplish. And in this same memorandum, which was written to someone at – at Harvard, he indicated that, hav- . ing searched the country, the only company that he found that h~d the technology and the talent to help develop and build such a machine was IBM. And that led to this relationship then between Harvard and IBM which resulted in the building of the MARK I. But that was built at IBM .. and the patent for MARK I has four names on it, one of which is Howard Aiken’s and the other three were IBM engineers.

JD: That’s news to me. I didn’t know that. I thougI thought he built the machine, but you say –

HT: No, it was built – built at IBM

JD: IBM.

HT: And at the time it was being built, he was on Navy duty.


page 8

JD: Yes. In fact, when he – when he demonstrated the machine, aa I recall, he was in uniform.

RM: Yeah, he was in Navy uniform.

HT: Well, he stayed in uniform until sometime after the war, and people often refer to that project as a ship, or, you know, environment. It was run like it was a Navy vessel. And all the logs are still available today that were kept during the – the running, the building of – of that whole sequence.

JD: Is the machine still — do you have the machine now?

HT: There are – the machine is in two pieces, .. both of which look complete with the input and the output, but the shaft is cut down considerable in size. .. One of the halves is in Harvard at the Computational Laboratory; the other half IBM has.*

JD: But you fellows don’t have any thing. *

HT: No, not the – not the MARK I. We have some pieces of it. I’ve got some of the little counters from MARK I.*

*[See note at end of transcript]


page 9

JD: Say, by the way, last night after I left you, I — it occurred to me that those devices that we talked about yesterday — you know, during that document

HT: Mhm.

JD: and it’s in the reports.

HT: Right.

JD: You know the one without a cover, that one,

HT: Mhm.

JD: it’s got the photograph. You may, if – if you want to take the time and trouble — yes, that one –

HT: Yes.

JD: you may be able to acquire some of that equipment from wherever it’s presently located because, for instance, Aberdeen may not be interested, ’cause they were taken care of with ENIAC. They may not be interested in what Bob and I delivered, and you just might acquire it for your archives. .. You might acquire that communications system from the Navy because it may be considered a prototype at this time instead of one of their operational pieces of gear. Or maybe they never did anthing with it, I don’t know.


page 10

HT: It would be interesting to trace these – these items through.

JD: I would suggest that you consider .. .. checking a little bit around to see whether or not anybody is willing to help you out .. to .. gather that stuff together.

HT: I’m also interested in checking at the University of Chicago to see if they saved any of the counters. Those little, I guess, little black boxes.

JD: I only built one, you know.

HT: Just one that was built there? The one that you went up to service, Bob?

RM: Yes, that’s all I know of.

JD: Yes. They

HT: Well, that may still be around.

JD: they wanted more. I think they wanted 40 but we couldn’t handle the job because we — we were snowed. I was limited in manpower and during the war, you couldn’t get technical manpower

HT: Mhm.

JD: because, it was there, but manpower was frozen practically. And I think I hired one man during the war and I think he was a 4-F. So – you just couldn’t get technical people that were good.


page 11

HT: Aha.

JD: So what we had to do was, use our then existing force.

HT: That might have been preserved.

RM: Mhm. I don’t know about that.

JD: What?

HT: Chicago might have saved that.

RM: It might have, I don’t know.

JD: They may have. I don’t know.

RM: They may have given it … It all depends on who took over, you know. Some people throw things away and some save them.

JD: You know they tore that pile down.

HT: Yeah.

JD: And when they tore it down, I don’t know whether, all the peripheral equipment that was with it, whether or not they kept that or not, I don’t know.

HT: It may even be at Argonne because then, about, you know during the post-war period,. that became essentially part of the Argonne National Labotory Development before they got their own facility.

JD: Well, there was nothing secret about that piece of gear. It was simple. It just simply took the time and trouble to build it and it wasn’t


page 12

difficult to make work because by that time we had perfected the tubes to a great degree and we had reliability into the thing. We had no trouble in there at all.

HT: It would be an – an interesting counter to have, with – with your little tubes in it and knowing the speed at which it was able to, to count the rays – the radiation.

JD: Well, it may be since we’ve got a box of tubes over there that we discovered we had and I didn’t know we had until I started cleaning the place out, a box about so big, full of these little tubes, that the company would give you some for your archives.

HT: That’s great.

JD: I mean it could be. You can ask them.

HT: Oh, sure. They’ve been very cooperative on everything and I’m sure they’re — thing is I’m loath to tamper with anything in that laboratory until people decide the best way to keep it intact. And the importance I say is – is the whole entity rather than a piece of this and a piece of that and I’d like to see the whole thing preserved properly.


page 13

JD: Now if you – if you carry that to its ultimate then you’ll have to deal with .. .. the University of Dayton .. .. Chemical Department where the Glass Blowing Shop is located. And I’ve already talked to them about this that they may be approached. And they are willing to – to return these machines if whoever gets them supplies them a new machine.

HT: Right.

JD: Of course, it won’t be like the old one, but what you’re trying to do is preserve the old anyway.

HT: That’s right. That’s right. And I talked to Jay about it and he – he said that was very feasible.

JD: Yes. Well then they would surrender that andand come out a winner really

HT: Yes.

JD: because they would get new equipment.

HT: Yes. But we would have the original and you know precisely where it starts and you can even see the markings on the floor where it was in the laboratory. You can see where that big machine stood. On the floor boards.

JD: Well, that didn’t play as big a part as you think.

HT: Aha. But, it’s just part of the environment.


page 14

JD: Yes.

HT: That’s right. In order to have the whole picture.

JD: …. There were different smaller machines wherewhere we did the sealing and stem making and so forth –

HT: Aha.

JD: that was done on other machines, and one of the machines that’s over there now never was used. I mean we – we, we abandoned the approach and the machine it sittin’ off in a corner, it wasn’t even used. So, that wouldn’t be properly a part of the display.

HT: Mhm.

JD: although you can have it if they make a decision to let you have it. You’re going to have to cut a hole in the roof to get that thing out, I’ll tell you right now.

HT: Well, shall we get through the MIT material?

JD: Yes, if you’ve had enough of …

HT: Oh, no, we’ll get back to these other things but as long as the MIT material is here. .. These are a series of about eight .. small books, which are reports, from MIT of various projects that were done in conjunction with NCR. Is that right?


JD: That is correct.

HT: Were you involved in this at all, Bob?

RM: No.

JD: Well, .. I’m trying to find out about the – the dates on here. Here is a brief progress report on the preliminary study and development of the rapid arithmetical machine. ..

HT: That would have to be in the early forties.

JD: Well, it’s not dated and I don’t know whether it’s even signed. There was one statement on page 11 that said, “The development of counting rings has now advanced to a stage where it can be safely, be temporarily suspended to permit attacking the extremely important problem of developing suitable matrix transformer elements or equivalent devices.” That would be Mumma’s.

HT: Mhm.

JD: “As soon as circumstances permit, however, the work with counting rings will be resumed.” That was a very.. was a very.. marginal counting ring, and now I’m trying to find out who wrote this report.

HT: Here it is.

JD: It was Radford.


page 16

HT: Radford, and it is dated.

RM: Yes, May

JD: May the 1st, 1939.

HT: It’s earlier than I thought. Right. It’s much earlier than I had expected. ‘Cause that correspondence between you and Caldwell and Bush and others in the other book that I was looking at seemed to be about 1941.

JD: I think I got ahold of – of Perry Crawford’s. prposal. Now let’s see if I’m right.

[Pause]

HT: This was about the time you were on deck in full force?

RM: Yes.

HT: But you were working on that prototype at that time?

RM: That’s right. This was just a month after I carne to work there.

HT: You had begun work then on the –

RM: Mhm.

JD: Radford and Caldwell prepared this on October 15, 1939, and it’s a – an investigation on the praticability of developing a rapid arithmetical machine; another proposal.


page 17

HT: Another proposal.. different from the one that was first reported.

JD: Well, I don’t know if it’s different or not. Or whether it’s similar to. But it’s now written by the head of the department

HT: Right.

JD: because.. .. .. Caldwell — see, that was Radford and what was the date on that?

RM: It was May 1st, 1939.

JD: This is October 15th so this one is a later one

JD: and it’s a thicker one, too. But here they propose certain circuitry, and

HT: Mhm. Characteristic curves in there.

JD: Yes. Here’s a scale of four counter and notice there’s two trigger pairs. That was the state, at the time, of vacuum tubes, of course.

HT: They’re interested in – in a state of — scale of four counter as opposed to a scale of two counter?

JD: No. What they’re trying to do is just build a ring. All right. So, they developed a trigger pair and optimized it. Then to see that it could drive another one, they – they optimized a combination, but they wanted to continue to go out to 10.


page 18

HT: Oh, I see.

JD: See the point?

HT: Yes, I see here you get into the square wave configuration.

JD: Yes. But, the point is they’re developing elements of the arithmetic machine

HT: I see.

JD: and not – that is not the whole thing.

HT: I see. Aha.

JD: Here is one .., let’s see, we’ve got another one here. Here’s another report. .

HT: Well, this one has an interesting appendix. Before you qet in

JD: Well, this is the same one, October 15th.

HT: – Yes. Before you get into that, you might look at the appendix, which is dated December 7th of ’38, not ’39, which is a copy of the, they call it the research program, essentially.

JD: This one says “Copy of Research Program December 7” on this one.

RM: Yes. Writer’s memorandum December 7th, 1938.

HT: Mhm.

JD: I think we’re talking about the same report.

HT: Right. I think I have the original and you have a copy.


page 19

JD: That’s right. I think it’s the same tning. Just two copies.

HT: That makes two as the literature .. references. Well, I’m going to separate one of these out. I’ll separate the copy out.

RM: Want to put this one over there, too.

JD: You know I notice — here’s two more identical. Both January 8th.
Well, now we’ll go back to rapid arithmetical machine research, October 15th, 1940. There are two of those.

HT: This is by Overbeck, and this is –

JD: Let’s see what he calls this.

HT: “Work done during the past year” and “duplicates much that was in the previous __”

JD: It’s what they called the Kleiotron and you’ll have to read it to see how it works.

HT: They call it the middle grid, but that –

JD: Yes, what, what’s the grid called? There is the control grid and there is the grid closest to the anodes.

RM: Screen grid?

JD: Screen grid: That’s the screen grid, and for some reason or another, the way this is constructed here’s the tube.


page 20

HT: Yes. This is a duplicate. On page 6, if you’ll notice — the next page from that, after the picture, yes there it is — is the section on the counting rings.

JD: Yes. And also the next page has got a picture of a whole, whole series of them. There is a K1eiotron counting ring. And this is dated October 15th, 1940.

HT: This is a bank of 10 in that figure 3 in the counting ring.

JD: Mhm. And it shows the ….. And further on in this report, they show the digitron and now that is on page –

HT: Schematic here.

JD: [turning pages] One more, I think. There, what would be page 10. That’s the construction of a digitron, and somewhere they ought to show a circuit using it.

HT: Discussion of the digitron starts on page 8.

JD: Does it?

HT: It says “Although our counting ring achievements gave us some satisfaction, we continually heard , remarks to the effect that, after all, counting rings were an established art and our real problems


page 21

lay in the development of storage and switching devices.”

JD: Here again “we wish to” — they didn’t hear that from us, I’ll tell you that. Here again: “we wish to avoid duplication of work being done in our sponsor’s laboratory” — that’s us

HT: Yes.

JD: “and decided to develop purely electronic storage devices rather than the magnetic type of device.” Because I was working on a magnetic storage, using a – an alloy tape;

HT: Mhm.

JD: solid metal tape.

HT: The first digitron is about three inches high, if you look at figure 6. They’ve got a picture of a completed one with all the wires hanging out of the bottom. There it is.

[Pause]

RM: You know they must have had a glass blowing setup up there at MIT, too.

HT: Yes. I’ll have to talk to Perry Crawford, who’s –

JD: It wasn’t very extensive and you noticetheir seals were all made of this type. It was not the type where you’ve got an envelope and


page 22

then it went back up in there with the stem inside. They didn’t have a machine to do that. They had to do it this way.

HT: Here’s a, a more recent one.

JD: That is a magnet around that tube. I think you’re going to find that it’s, it’s magnetically controlled.

HT: Mhm.

JD: That’s a magnet up there. …

RM: That’s a gas tube and they tried to get the, the glow to move around, I think. That was a digitron.

JD: Well, they tried to give it preferential motion with a magnetic field.

HT: Here’s their vacuum system. There it is on figure 8.

JD: Yes, there it is. That’s it.

HT: Do you remember seeing this?

JD: Oh, yes.

RM: I didn’t.

JD: I was there repeatedly. I didn’t know it was in one of these books.

HT: Yes. It’s not quite the free-standing system that yours is.

JD: No.

HT: It looks like a real —


page 23

JD: Lot of framework.

HT: Right. They have it all within a frame. You can see the tank. .. Now, what kind of gas were they using?

JD: They used –

HT: For the tanks?

JD: Argon.

HT: Argon?

RM: Probably argon.

JD: A little further back there, you’ll run into the Kleiotron. .

HT: Right. And to record the test data here, screen current versus voltage graph. There it is.

JD: Yes. Now there is a trigger circuit using one vacuum tube.

HT: Mhm.

JD: That’s what that was.

HT: Mhm. Their data for their ten two’s on both the on-off position. .. And here’s the circuitry for the Kleiotron counting ring.

JD: Figure 12, Kleiotron counting ring. You notice –

HT: Here’s a double triode counting ring, here in figure 13.

JD: Yes. You know – you see what they were doing?


page 24

They were trying this, then they were trying that and then they were trying that. They kept on moving around, trying to determine the optimum condition.

HT: Mhm.

JD: And, instead of bUilding a useful device of some kind — now, we didn’t, our problem was to develop that single tube to perfection,

HT: Mhm.

JD: but we weren’t constantly moving around into all kinds of circuitry,

HT: Mhm.

JD: except as it occurred to us .. .. sporadically, and then we would do something about it to get a patent on it and then keep on going, see. But, these people were truly searching for the ideal element for the ring.

RM: Yes. We’ve got quite a few patents on special counting circuits, counting rings.

HT: On special counting circuits?

JD: We were pretty well settled on what we wanted, but when we thought of other ways to do it, we – we recorded it and, and went on.

HT: What I was looking for, and apparently it’s not in here.


page 25

JD: Well, there may be –

HT: What I was looking for was a – a kind of summary at the end but really all of uhis is a survey of each of the different stages and the data that they got.

RM: There is no conclusion, no conclusion there.

HT: No conclusions and no suggestions, and notno kind of summary material, so there may be a later report.

RM: Yes, there is a later one.

JD: There is. Here’s a – here’s a, here’s a December 9, 1941 and I saw a ’42 –

HT: That’s a year later. That’s February.

JD: and I saw a ’42 here a little bit ago.

HT: No that’s –

JD: That’s ’40.

HT: February.

JD: That’s ’40.

HT: February of ’40. Now that precedes this one. Now let’s see what that .– now this is a confidential report to NCR.

JD: Here – here is a final report.

HT: Oh, there’s a final report. Dated September

JD: ’41.


page 26

HT: September 30, 1941.

JD: But, there is one around here that is marked somewhere. Here it is. High Speed Counter, January ’42.

HT: “A New High Speed Device.” Well, this then must be a descriptive report of the completed device.

JD: Here’s the biquinary electronic counter. We had a bi-quinary, too.

RM: Yes, we sure did.

HT: It shows here – it states here that you had completed an experimental unit which works at one million pulses per second.

JD: It says that to what — that they had done that?

RM: That we had or they had?

JD: Where is that? “One of our own present-day goals”

HT: The last sentence of the first paragraph.

JD: .. “We have now completed an experimental unit which works reliably at a million per second and which we believe is considerably simpler than existing circuits.” Well, I wonder which one it is.

HT: Well, here’s the figure — the bi-quinary.

RM: Yes.

JD: Oh, yes. Okay. The speed is determined by these


page 27

RM: Yes.

JD: trigger pairs out here, because that scales it down by a factor to go into a slower counter. We did the same thing.

RM: With tubes, yes we did.

HT: A signature on this. There doesn’t seem to be one. This looks like an Overbeck report but there’s no name on it.

JD: Here’s another one without a name on it.

RM: Surprising they let them do that. Particularly with Caldwell on it.

HT: This is confidential to you. 1941.

JD: How can it be?

HT: That’s what it says.

JD: No. There’s nothing there that .. magnetic storage put on it. That’s all they did. Let’s see what kind of magnetic storage — let’s see, if this is a pinwheel. Yeah, here – here, here’s what the thing was.

HT: Mhm.

JD: This – this wheel was on a shaft [with] a lot of wheels. Each one represented a digit and they were spinnin’

HT: Mhm.


page 28

JD: and there were little magnets. And you’d put pole pieces out here, and you – any time you get a pulse when the magnetized stem goes by, and that gives you, that tells you what the digit was.

HT: Mhm. This final report, by the way, is not an MIT report. This is the University of Chicago.

JD: Oh, it is?

HT: This is your counters. Yes.

JD: Oh. Well, here

HT: That’s the one.

JD: That’s the one we started with. .

HT: Right.

JD: Yes, there it is. Remember it?

RM: Yes, that’s the one.

HT: Yes, that’s your baby. Yes. But, that’s the one we were just talking about.

JD: Yes, but that isn’t the way it ended up. That-you’ll have to look at our reports to find out how it ended up, because that didn’t work.

RM: That’s their proposal.

HT: Well, it says “completion of contract” at the bottom.

JD: Well, that was to NDRC.

HT: Yes.


page 29

JD: Because they were working for NDRC.

HT: Right. This is an NDRC .. number designation.

JD: I don’t think their heart was in their work and I think NDRC knew it and I think they were very glad to get rid of it because they weren’t a bit upset about it.

HT: Aha.

JD: In fact, they gave me work afterwards .. — the University of Chicago did.

HT: Aha.

JD: And that – that was, when

RM: Yes, I remember this now.

JD: we got that, we hadn’t built this other device.

HT: Oh, I see.

RM: This is the concept, I think, you’ll find, the first concept to recycle the – the binary counter in which you count up to ten and then kick it back to zero.

HT: What you were talking about yesterday?

RM: Mhm. And this is the University of Chicago ccuit that did that. And then you even had the recording device worked out for it, so you’d scan the pattern.

JD: Excepting that we spent about a year, or half a year, perfecting it.


page 30

RM: Yes. It wouldn’t work right.

JD: It didn’t do what they claimed.

RM: No. We had to put some extra diodes in and work it out so that the pulses wouldn’t get mixed up going back recycling the counter.

JD: No, there’s nothing secret about it.

HT: I was just kidding. I just …..

JD: Here’s January the 18th, 1942, “New High Speed Counter.” Is that the one you’ve already looked at? Yeah, you did already look at it. There are duplicates here.

HT: Yes.

RM: September 30, 1941.

JD: I think you’ve seen that one by Overbeck.

HT: Yes. Mhm.

JD: And,

HT: Yes, this is the October 15th that we looked at.

JD: here’s the ’41.

HT: Right. This is the one we went through together.

JD: Yes. Here’s a ’41. .

HT: I think we looked at that, too.

JD: No signature and no heading. On that though who wrote this? Gives absolutely no idea.

HT: No way of knowing. No identification.


page 31

JD: No, and there’s a brief report on this one. Now, this is the type of work they were doing for us. That’s what I wanted you to see. You might as well put that with that.

HT: Right. Right.

JD: .. This is the type of work they were doing for us, and I wanted you to know it as part of this whole program.

HT: Right. Well, I’m glad Gene was able to find these. That’s the one I had taken off.

JD: Well, I saw them just last week when I had them home. ,

HT: Well, fortunately there are a lot of duplications here.

JD: I’m trying to find the one that Perry Crawford wrote.

HT: Oh, I think that was the first one I took out. Was –

JD: Oh?

HT: See if this is the one. That’s the one you have a copy of there. No, this is by Radford and Caldwell.

JD: Radford. That’s right. Now, I haven’t seen anything by Crawford and I know there is one by Crawford.


page 32

HT: I thought I saw one by Perry. It’s not in this collection. Let’s see if it’s in here. This is the correspondence. There’s material by Crawford in this correspondence file — research contracts.

JD: Maybe Perry didn’t write the original proposal.

HT: Maybe.

JD: Maybe it was Radford.

HT: Mhm. I’ll have to check with Perry. I’ll get a chance to see him.

JD: Perry may not have written it. I may be wrong and Radford is the fellow that wrote it. Now, you want to pull those over and put them back in this file?

HT: [Moving] Get this one out of here. That goes in there too. This – I guess that was separate.

JD: That doesn’t go in there.

HT: But that was with this. This is the corresp- this white folder is the correspondence.

JD: Oh, well, they shouldn’t go in there.

HT: Related to MIT.

JD: I don’t think it will fit in anyway.

HT: No. No. While Joe is putting some of this stuff together, . Bob, you remember some of our discussions yesterday,


page 33

were there any major holes or gaps in the early period that we talked about that triggered your memory after we bro~e up yesterday? Were there any things that you figured ought to be major topics that we ought to get more detail on?

RM: I can’t think of any.

JD: We’re not going to go through my reports?

HT: Not at the moment.

RM: I worked on those three computers and then I worked on the device that went to Aberdeen which was a straight counter. .

HT: Mhm.

RM: And then the device that went up to, to the University of Chicago. And then we got on this . communication bit. As my memory of how things went, I worked on the counter circuits and that, and somebody else worked on other parts. I had a lot of help on that. I think Frank Bucher was involved in that. See Patent Nos. 2,425,307; 2,428,089; 2,451,812; 2,451,859; 2,462,613; 2,466,467; 2,556,614.
JD: Those would be recorded in those, in those reports that were uncovered just last night or this moing, .. which extends the work over the longer


page 34

period of time that you did

RM: Yes, that’s right.

JD: and not just that first machine. Isn’t that right?

RM: That’s right.

JD: So there are other reports backing up what you say.

RM: Oh, yes. In the notebooks, and things of that sort. But that – we did that along with NDRC work after I broke away from that, that big -did you ever see that big computer that looks like a great biq box? I think that’s down in our warehouse.

HT: I’ve seen a picture of it crated. I’ve not seen a picture of it

RM: I see.

HT: with the, you know, with the box off.

RM: With the keyboard, and that sort of thing?

HT: No. I saw the – the picture of the prototype that we looked at in this

RM: Yes.

JD: Cover folder.

HT: folder with the cover on it, but I haven’t seen –

RM: Is that the vertical one? Are there pictures of that other device?


page 35

HT: The one that stands?

RM: No, the other one, the one that is spread out like a box.

HT: Yes, that’s in here.

RM: Yes, okay. That’s the one. That’s what we took up to Minneapolis. But I didn’t — Larry Killheffer finally finished that job up. I laid out the plans for it and [pause] I don’t remember seeing it.

HT: I thought it was. I know we looked at a picture of it.

RM or HT: I thought it was in there.

RM: ‘Cause it was a great big piece.

HT: Yes. I guess there isn’t any. I remember looking at a picture. But I guess it was in one of the reports we looked at.

JD: No, it’s in there.

HT: It’s in here?

JD: Yes, I’ll get it for you.

HT: Okay.

RM: Mm. Must have – something. You needed a layout of it right there.

HT: That’s where I thought we had seen it. And it would be fairly early.


page 36

RM: That’s the plan, but I don’t remember seeing a photograph in there.

JD: You see, that – that –

HT: Oh, there it is. Sure.

RM: No, no. That’s the original –

HT: That’s the first one.

JD&RM&HT: There it is.

RM: Yes, okay. That’s about it.

HT: Right. This is the one labeled exhibit 7 in that

RM: Yes.

HT: box

RM: That’s the one –

HT: and I’ve only seen this in a crate.

RM: Yes.

HT: So, I’ve not even seen that much of it in a recent photograph.

RM: That’s the one I had up in Minneapolis and we added, subtracted and multiplied on it for the judge up there.

JD: I don’t recall when I – this right here, these blueprints and this write-up, when I sent that to Warren Weaver, I don’t recall if I sent up a photograph of this along or not.

HT: Aha.


page 37

JD: I can’t recall.

HT: Yes.

JD: Unless I said so in the write-up.

HT: Well, I think it might be .. interesting now to talk about this .. material — not necessarily the suit, but the material related to the interference suit with Mr. Dickinson and – and IBM. When did you first learn about his work, and whenhow did you get into this whole – this patent clash with Dickinson?

JD: Well, of course, it occurs when you file a patent on a device and the interference develops in the U.S. Patent Office between two or more inventors,

HT: Right.

JD: where they see similar claims. And, on two dissimilar approaches,

HT: Mhm.

JD: both claiming certain things. And.. I don’t call anymore all the areas of- of 11 claims in question, but there were 11 and there were preliminary skirmishes between the lawyers in determining the extent of the interference — that is, how many things were truly legitimate interferences. And – and this occurred, I think, in


page 38

Cinc– .. .. this occurred in Cincinnati, did it, or was it in Washington?

RM: The interference?

JD: No, the determination

RM: Oh.

JD: of what was going to be involved in this.

RM: Probably Des Jardin, he was our outside Counsel? (Mr. Des Jardin was a Patent Attorney from Cincinnati, Ohio.)

JD: That was down in Cincinnati.

RM: Cincinnati, I think. Yes, ’cause — .

JD: So, he was our attorney, Des Jardin. He’s dead now. Well, anyway, .. after the lawyers had their day and settled and, and arrived at a .. .. a satisfactory position where each agreed this would be the limit of the test; there were some contentions or other that were thrown out. .. Then the discussion of preparing to go to .. to take depositions. So, the first depositions were taken in New York at the World Headquarters of IBM. We spent two weeks down there while Dickinson testified on his approach and, and justifying his claims. The claims were always under fire during these – during the testimony.


page 39

The lawyers were trying to prove these claims and the validity of them, and the work that was done was brought forth. Mountains of blueprints and so forth. I think they are in, in the, in these files here. I think you pick one that’s loaded ~ith blueprints, folded up, and you’ll find the blueprints that were being involved in the interference. [Walking across room] For instance, here’s one. I can tell by looking at it. See it?

HT: Yes.

JD: And this is ….

HT: It’s about four inches thick.

JD: Brief with Dickinson at the final hearing. If you go to looking at it, you’re going to find. it just full of blueprints.

HT: Aha.

JD: And.. these look like Dickinson blueprints. They are. So-

HT: Without going into all of the detail, what can you remember about what Dickinson’s approach and early motivation was?

JD: That I can’t help you on,

HT: Do – you know?


page 40

JD: because I don’t remember.

HT: Aha.

RM: He had a cathode ray tube device, remember?

JD: Well, he, first of all — I’ll tell you the reason that I don’t remember is that in reviewing his work, he had a starting date. First of all, he didn’t build — that was paper patent work that he did.

HT: Right.

JD: He didn’t build this stuff and he, he started out with devices that – that he thought would work. And then all at once he thought of another one and he thought of another one and they were all dated and so forth. And – we took testimony -his testimony took about a week, I would imagine. And at night, Bob and I would spend our time examining the technical aspects of what he was presenting, because we had copies of it everyday.

HT: Mhm.

JD: Stenographic copies were available at 5, 6 o’clock in the evening. And, one night — this is the reason that we acquired eight out of eleven .. .. .. .. claims was the fact that in examining one blueprint — Bob and I were on the hotel floor


page 41

crawling on our knees; it was a big blueprint — we found the error in the diagram that would make it inoperative. What it was; he tied all the grids together in a string of tubes, and all the plates together in a string of tubes, and he had no way of selecting the tube. We found 17 or 18 drawings that had that mistake in it.

HT: That same configuration?

JD: Yes. And.. it was just copied from one to the other, to the other, to the other, as –

HT: Right. And had he been building it, he would have caught that error instantly.

JD: That’s what tipped us off that he wasn’t building it.

HT: Aha.

JD: Then our counsel started to go after him about building it and .. .. we had him in a box on that one because we found also one of his other early ones that wouldn’t work because of a defect in the design, because he never built it.

Well, we knocked out his early dates and we became the early — we, we got ahead of him by a couple of years, I think, as a result of doing that.

RM: See, we had a working model down at that hearing.


page 42

HT: Mhm.

JD: We had that tall model sitting there always threatening to use it.

HT: To add two and two for you.

JD: Well, anyway, they finished and we cross-examined and so forth. The whole thing took two weeks. Now, during this period, Bob and I were practically prisoners for the simple reason that we had to keep studying like crazy to get ready for our testimony, so that we wouldn’t fall into a trap. Some of our testimony is in there, Bob –

HT: That’s the volume that you showed me

JD: Yes.

HT: and Harry Williams’ testimony is in the same volume.

JD: Yes, all that. And we presented plenty of drawings and blueprints, too. And, that was done in Dayton. We – we took depositions in Dayton. This was all depositions we’re talking about now. We’re not in a courtroom,

HT: Right.

JD: and we .. we testified down at the .. .. Biltmore Hotel. It’s now called the Sheraton.

RM: Sheraton of Dayton.

JD: Yes. And we spent two weeks here. .. So there


page 43

–were four weeks altogether in which we’re contesting the claims of these .. two approaches.

RM: Wasn’t their attorney Kolish?

JD: Yes, Kolish was counsel for them and they hired him just like we hired Des Jardin. We was not an employee of IBM. He was a New York lawyer. He died, incidentally.

RM: I think Dickinson married his daughter or something like that. There is some relationship there.

HT: I hope I’ll be seeing him in a few months. But, the interesting thing — and that’s w~ I’m glad I not only had a chance to talk to you Bob, but I’ll get to see him, because he was working in a similar environment. He was interested, I’m sure,

RM: Oh yes.

HT: in improving the company products of that –

JD: Certainly.

HT: of that day.

JD: He certainly was. You’ll notice if you look at his circuit diagrams, which I wish you would, you’re going to find that they’re electronic and, and there, there’re strings of tubes and things of that kind.