An Interview with Jim Dimmick

Jim Dimmick


JD = Jim Dimmick
GAM = George Michael




JD: As of the 7th, I am 79 years old.

GAM: Really? Oh that's wonderful. Today is May 2, 1997 and we are going to interview one of the great persons who came to the Laboratory because we had IBM equipment. This is Jim Dimmick. Jim, why don't you tell us when and how you got to the lab and what you worked on initially.

JD: OK, I came out with another fellow and I can't remember who he was, but we came out when the IBM 701 was installed. I was not part of the 701 world, but I wanted to see it. I felt that things were moving too fast, and I was not a part of it and, therefore, I would come out and witness this. I'd heard about it a lot, and I'd worked with people who did computations, Joe Brady for one. He and I worked together on the IBM 602A. He did things with the 602A that just baffled me, that's the only way of putting it. And so I asked him. He had trouble with the machine and I was aware that, with electronic machines, you really could bias them. He would operate them at different voltages. So I looked at the blueprints for this mechanical machine, the mechanical relays and all the rest of it, and decided there were power supplies that I could make adjustments on. There were three different voltages, and if I ran it at the three voltages with the test pack, with a little bit of luck, it would fail on one or two voltages that were not being used. So I did this and sure enough, it failed.

GAM: Where was this?

JD: This was at Cory Hall in Berkeley. He and Dr. Cunningham were in a little building, oh, a block away I suppose, and Joe would show up and run decks of cards through this 602A for hours, and hours, and hours, and hours. This is card machine speed, and slow card machine speed at that. But when I got through, I discovered I had voltages adjusted that would not fail with the test deck. Joe came in, ran his work, and found no errors with his work.

GAM: Beautiful!

JD: Just the way life should be.

GAM: Yes, yes.

JD: I think that's when our fast friendship was formed. This was the electrical engineering department—I can't remember what they called it then. The building was Cory Hall, and the name of the person will come to me in a moment, who was the chief honcho there.

GAM: What year was this, do you remember?

JD: Yes, let's see, we'll figure it out. This had to be 19...

GAM: 1950?

JD: Early 1950s.

GAM: Well, yes, Joe came to the Lab in 1952, so it was prior to that.

JD: That's right, and I'm getting that date by when my first wife decided she'd had enough of me. This is a real movement point. If you think you've learned something about life before that, you've got a surprise coming. So, this had to be 1953 probably. Heavens name, what was the electrical engineering...?

GAM: Well, the only one I knew was Bill Wattenberg.

JD: No, no, no, anyway, his name will come to me eventually. At that time, the only electronic computers I could see anywhere had to do with the Electrical Engineering Department having decided that they needed a computer.

GAM: Yes.

JD: By God, if you're going to be in this thing, you gotta have an electronic computer. And so they built their own and, unfortunately, they didn't understand what kind of solder to use, and I did not feel it was my place to tell them how to build a computer, so I didn't straighten them out. It would have been a great friendly thing to do.

In the meantime, I went to a school in Poughkeepsie where I supposedly learned how to repair one of the keypunches that was new. And, always, when that happened, a class was given a tour of the building in IBM style, that's the way they did it. So, I got upstairs following an instructor, who showed us a room that was absolutely mind boggling. It was full of several hundred IBM 727 tape drives. These are the big things. I don't know who ordered them, but they were ordered with the idea that there were going to be a lot of IBM 701s and 702s and 704s built. The reason I say that I don't know who took that chance and ordered them, is that it wasn't terribly obvious that they were going to be a successful machine. But, that wasn't how it was. Somebody had the right idea and understood that if we, IBM, didn't do it, somebody else would—Philco, or whoever. So, I went back home and, within weeks, I was told to go back to Poughkeepsie again. This time to study the innards of a computer, and this time I enjoyed it.

I had been in radio work for years and years. I had become a radio repairman in 1937, just out of high school. I got involved with ham radio at the same time, and this saved my soul, although it had kind of funny times getting there. When I was drafted into the Army, I was sent to Fort Ord for basic training. There was a young man and I saluted him properly, because he was a corporal—what else would you do, if you are a brand new private? He gave me a test. He said, "Here's a ten question test." Do you use rosin-core or acid-core solder?" And I said, "That's easy, you use rosin-core solder" and he said, "No." And the next question was, "Do you clean your soldering iron with sal amoniac or do you wipe it with a cloth?" and I said, "That's easy also, you wipe it with a cloth" and he said, "No." And I realized then that we are not on the same wave length and I said to him, "These are simply not so." And he was so irritated that he took his ten questions, open on the clip board, and held it up for me to see that it was marked "Radiator Repair." If I had gotten smart with that guy, he would have had me in a motor pool in no time at all, and I would have been lost. But, his answer to that was, what else do you say when you have an answer like that? Oh, spelled O-H. And he looked at his book again and said, "OK, here's another ten questions," which I answered properly and, a day and half later, I was in a radio repair organization, the first one on the West Coast.

GAM: Radio repair, not radiator repair.

JD: Radio repair, not radiator repair, the two don't mix.

So, there had been a guy, a little short fellow, who had said to me, before I allowed myself to be drafted, the very idea of being drafted, for God's sakes, what's the matter with you, "I can make you a Corporal at the end of 12 weeks of basic training. And I said, "No, I'll have to think it over, but I don't think I'm going to do that. I'm not going to join the National Guard, which is what he was part of." So, I let myself be drafted, and short of almost having gotten into a motor pool instead of into a radio repair outfit, I was happy. For 12 weeks, I could yell at the top of my voice and tell the other guys how to march and at the end of the 12 weeks, I was a Sergeant, for God's sake. Just like that, they had the ratings to give out, and they gave several Sergeant's ratings, and I got one of them. And, within another week or two, the Lieutenant and I, a big, tall, lieutenant drove down to Camp Roberts, and I din't know what we were there for. We were in the motor pool down there, and he said, "Just wait here, I'm going to go find whomever we're supposed to see." Later, I saw him coming back through the motor pool, through the cars, and I could see he was with somebody, but I had no clue who it was. Well, it was the corporal who had tried to get me into the National Guard. He looked at me with my sergeant's stripes, and I think at the instant he saw me he realized that I'd done the right thing and he'd better start moving. Actually, he became a lieutenant, in no time at all, by going to OCS, I would never have done that. So, that's how I got caught up in more radio work, and that led to wonderful things like being sent with a Signal Repair Unit to Melbourne, Australia. The military police there needed communication. I was told, "OK, go ahead and install this FM communication set in a military police vehicle." So, three guys came out in that vehicle, I didn't know who they were or what they were, but I put the radio in, there was nothing to that. Then we had to go out and test it. We drove all over Melbourne, it's a city the size of, well I don't know that it was as big as San Francisco, but certainly the size of Oakland then. And finally they said, "Well, it's about time to try this thing out." So, they took the microphone, and I think to myself, "OK, I'm going to witness the testing of this radio." Well, they handed the microphone to me. I had enough sense not to pinch the button on it, but I told them, "I have no idea what you're expecting of me, and I don't know what I'm supposed to say to this radio operator downtown at the police station." "Well, say the same thing you would at home." And so I stop and think, "OK I've listened to, and enjoyed listening to, the Los Angeles Police Department, even up in Sacramento. And finally I said, "Uh, Melbourne Police, this is car ..." "Oh," I said, "What car are we?" and they had to think that through. So, they decided we were going to be car 100. "What car are we?" "Oh we are going to be car 100." "OK, this is car 100, would you give me a time check please?" and let loose of the button. Everything was quiet, it stayed quiet and I'm thinking to myself, "The damn thing isn't working" and they started laughing. These three cops started laughing and they didn't stop. And finally, the radio came alive and said, "It's ten minutes of midnight. Thank you car 100." Big deal! Great roaring laughter. Well, what I didn't picture was another group of people waiting around the police operator, waiting for this first Yank call to come in and, of course, he didn't know what the hell to do with it. What's the Goddamn Yanks doing now? First, they run off with the women and now they are running off with the police cars! Well, life continued to be like this. It was really a wonderful experience. And, when I came home, it was with an Aussie bride, who saw the light and took off eventually. And that was ok too—I long since have forgiven her.

GAM: What year did you come home?

JD: 1945. I was actually on furlough, and was discharged a month before VJ Day. I had been in the Philippines a month before that, and could have been transferred to Japan and would have seen some of the things you have seen. I think it's probably just as well. After I got out of the army, I went to school and then went to work for IBM. I didn't think I'd stay there, but I had too much fun.

GAM: When you got to the Lab it was 1953 or 1954?

JD: Well, it was more like 1955 because the IBM 701 had been in for a while, for two or three years. I never learned to repair it. But, I did have the experience of seeing the failure of that machine because of the sun shinning through the South Windows onto the Williams Tubes.

GAM: Ah, yes, do you remember Ernie, the guy with the peg leg?

JD: Yes.

GAM: I got a Williams Tube from him when we got rid of the IBM 701, and it's now in our collection at the Computer Museum and History Center over at Moffett Field.

JD: Wonderful.

GAM: Well, Cunningham got a 701 with a core memory, and it made all the difference in the world to its reliability.

JD: That's right. He got the 701 that had been out at the Lab, and I can remember it being installed. Ted Ross was the engineer for it and I forget all the trouble we had. I don't know whether Joe Brady had anything to do with the discovery that the sun would shine through that window at a certain time of the year and make it fail. But, it started failing, once a day, and it would go out completely until the sun moved. I don't know who found that out, but that was a neat job as far as I was concerned.

GAM: So, the first machine you worked on, then, was the 704?

JD: Right, exactly.

GAM: Nice tube machine.

JD: That's right, just hundreds of tubes. I was in my glory, and the thing that I did, that nobody else really had the guts to do, was to go around and tap the tubes. Because I'd been doing that for years, I was careful so others didn't see me tapping the tubes. But, I could find out which tubes were causing the machine to fail by simply tapping them.

GAM: Yes.

JD: Eventually, of course Shockley invented the transistor. Is that right?

GAM: Well, Shockley was involved with it, yes, at Bell Labs. Actually, lots of people like to say that, with the hundred million dollars that the Government poured into the research area, maybe it was the Government that invented the transistor.

JD: They invented the source of money anyway.

GAM: Great. Anyway, they certainly were looking for it.

JD: Yes, well anyway, that's another story. But, Ted Ross and I decided we had to get involved with this, so we bought the cheapest transistors we could buy and we made an oscillator. And we turned it on, and it oscillated as it should have, and made the tone, and we invented the Ross Effect, which was holding your fingers on the oscillator, on the transistor, and making the frequency change. You may hear differently, but that is, as far as I'm concerned, the Ross Effect.

GAM: Well, that's great, but by virtue of you having been here when the 704 got uncrated, you were involved with all the machines that IBM delivered to us from then on.

JD: That's right, exactly. Until they started un-delivering them. But that's how I spent my time on the Photostore all during the 70s. And, when the last year of the 70s came, that was when somebody called me up and said I could retire. And I didn't have any trouble at all saying yes; I was ready to go.

GAM: Well, you understand, that, nationally, you have the reputation that if you hadn't been around, the Photostore would have been an abject failure.

JD: Well, it was fun, it really was. The thing that surprised me about it, in circuitry, was that they used Digital Frequency Control, where you change the frequency so that an oscillator will track the bits of stuff on the film. That way, you have a fun plaything for the engineer because he can make adjustments and watch the scope and see how fast the thing is tracking. He can also watch and see the pulses of energy coming from the reader tube, in other words, the photoelectric tube, and make the data come in at the right rate so it can be read and, most important of all, so that it can repair glitches in that signal. And that was the reason that machine almost never sent out a message grossly in error.

GAM: Well, there were two. The other reason was that it had this error correcting code on it which accounted for 30 per cent of the memory capacity, but who cared, given we had more than 1012 bits.

JD: Yes, that's right.

GAM: I used to tell visitors that, when the Lab put the Photostore on line, it doubled the on-line memory of the planet earth.

JD: Yes, that's right.

GAM: Incredible.

JD: And operating memory.

GAM: Yes, and it worked, and it worked beautifully. As you remember things now, what were parts of the Photostore that caused you a lot of trouble? There must have been one weak thing or something like that? I mean, was it the electron beam recorder or the little factory that developed the chips or what?

JD: The little factory that developed the chips. The thing that used to worry me was that one of us would get killed because it was surrounded by high voltage. Well, not high, just a hundred and fifty or two hundred volts AC, and good back up for that, so if you did accidentally get one hand into the stop solution and the other on some piece of metal you could do yourself in. So, that was the single biggest problem, keeping it clean enough so that it wasn't a danger to itself as well as the people working on it.

GAM: Whose idea was it to put the surfactant into the wash water? That made the Photostore really usable here, whereas you remember, at the National Security Agency site, they didn't do that and...

JD: I didn't know that story.

GAM: You went back there and tried to make their thing work, didn't you?

JD: No, I got as far away as LASL, and even that was a kind of funny story.

GAM: Well, tell it.

JD: Well, I was called and told to take a call at Los Alamos. So, I saluted and headed for the airport. The last thing I did before I got on the plane was check in and see what they wanted me to do. They still didn't know, exactly, what they wanted me to do, but finally they said, "Get on the airplane." So, that takes care of that, at least for the moment. So, I got into Albuquerque and I had never been in New Mexico before. Well, that isn't so, but that's another story. Anyway, I got to the airport and rented a car, but before I did that, I called the office, yet again, in Oakland, and they still didn't know what I am supposed to do, but they say, "OK, rent the car and go on up to Los Alamos. So I did, it was a pleasant drive.

GAM: It was a long one.

JD: A long one, and I got up there and made some more telephone calls. The first thing they said was that they wanted me to come back home, and I said, "That's fine." "No" they said, "call so and so at Los Alamos." So I did that, and I never did really work on that machine. Somebody had made an adjustment on the machine so that the recording beam was not adjusted absolutely correctly and, if you were able to adjust the reader so that it didn't care that it was not adjusted correctly, it would actually read in spite of this glitch in the recording thing.

GAM: Yes, ok.

JD: And if you looked at the chip with the microscope, you could see that all you had to do was spread out the reader beam so that it would not get confused about what it was looking at and would be able to follow the thing and send the data to it's electronic circuit so that it would get the right speed and so on. But, at the end of the week, they weren't sure what to do. And so I said, "Let's consider this a meeting, myself, and two or three others, to decide whether I can go home and visit with my wife this weekend. So, we did that and I was on the plane the next morning, on my way back to Oakland. I think that I learned something. I think that the people who repaired the machine down there, or took care of the machine there, felt stabilized and it really was a matter of something I've always enjoyed, namely putting my hand on somebody's back and saying, "OK, it's OK folks, you're all right".

GAM: Calm down!

JD: Calm down, and so that took care of that. I never went out again on a call on that thing.

GAM: Well, all I know is, from the statistics collected here, there and elsewhere, that our machine at Livermore was reliable, compared to the others.

JD: Yes.

GAM: We used to call it the Photosnore because it was down a lot.

JD: There's another subject. I've learned, within the week, how to control my snoring.

GAM: Oh, yes?

JD: Betty is very appreciative of it.

GAM: I'll bet.

JD: Instead of all the pills and Jerry Rice things and all this, I ended up going down to Dom's Surplus and getting a piece of plastic that would raise my head up so I no longer snore. Isn't that nice?

GAM: That's great.

JD: Anyway, I don't even wake myself up anymore, which I used to do.

Back to the Photosnore—I had never heard that by the way.

GAM: What, the name for the Photostore?

JD: Yes.

GAM: Well, I suspect that was probably from the users. I don't know who first used the term. But, there were two things that contributed to the success of the Photostore—one is Jim Dimmick, he kept the store running, and the other was Garret Boer, who gave us a really decent software environment.

JD: Ah, if you'd asked me who I was gonna name for that person, it was Garret—that's right. That is absolutely true.

GAM: He did an absolutely magnificent job on the software.

JD: I guess he's still around, I've seen him recently.

GAM: Oh, yes.

JD: He lives in the neighborhood. And I hope that the good stuff that he came up with is still useful, because it really was a marvelous thing.

GAM: Yes.

JD: Now, the other thing that I appreciated, that went into the Photostore system was the CRTs that hung in various offices, showing what the status of the machine was. If I was out taking care of some other small machine, which I occasionally did, I could look up and see how things were really doing, and I knew I didn't have to wait for a call. I could see that the damn Photostore was down and I would say, "Excuse me folks, your key punch will take care of itself for a little while, and I will go take care of this machine and make it work right."

GAM: Yes, the show channel on the TMDS was a nice thing, a status display everybody could look at. I don't know who all did it, but I do know that Bob Judd was involved with getting it up on the display.

Did you have any direct dealings with Sid?

JD: Well, occasionally, I really liked the guy, I thought he was wonderful, but I was also very careful not to stumble in front of him. He had a tremendous sense of humor, I don't know whether you ever heard the story of Al Fram? Ed Schoonover may have told you about it.

GAM: Yes, but please, you tell it.

JD: I think that Al had gotten the idea, by overhearing Sid talking to somebody, that if some IBM thing doesn't work by a certain time, out it goes. So Al got on the phone and he was reporting all this to the office. I don't know who the manager in Oakland was, but he was gertting this report from Al. And Al was not only telling the story, but he was imitating Sid; completely, talking out of the side of his mouth the way Sid did.

Suddenly, Al became aware that Sid was standing behind him. And, Sid appreciated this tremendously. Al turned and got out of there as fast as he could. But then Schoonover said to Sid, "You've gotta go back and tell him it's OK." Because, by now, Al was ready to hang up his hat and get the hell out of there. Oh, gosh.

GAM: Well, Fram was a very funny person, and knowing how he was makes this story all the more appealing. And yes, Sid went back and said it was OK. Everybody had a good laugh out of it, one of the nicer little things. I remember where it took place; it was in that building where the 701 used to be.

JD: Yes.

GAM: I think that it was unusual that you got to stay at the Lab so long. Ordinarily, IBM rotated the Customer Engineers around, outside of the project, and put them in other places, because of the experience they gained having worked at the Lab, and they wanted to get some of that experience into these other installations.

JD: And I'm sure you've heard that "IBM" means, "I've been moved".

GAM: Oh, I've never heard that! Great.

JD: I had to work very hard to stay put. I didn't work to make the stock go up, or anything like that, I worked because I enjoyed the people who were around, to hear the latest joke going around, from Cralle and any number of other people. I don't know whether Schoonover ever told you about when he and I went to talk to one of the CDC people. We spoke to the Control Data people about the fact that tapes written on the IBM and CDC computers were not interchangeable. This had nothing to do with mass storage. This was the 704, and if the tape heads were turned slightly, you would no longer be able to get good data off the tapes.

So Ed says, "Well, you've got to tell the Control Data guy and I thought to myself, "Dammit, Ed, it is not my role in life to correct them. This is precious data, if they can't fix the Goddamn thing so it'll run, why then..." At any rate, eventually, the guy asked me, "Well, how much force do you have to use to turn the head enough to read the data correctly?" You haven't heard this story?

GAM: No.

JD: I said, "This much." He nearly fell over, because I'm pushing on his shoulder.

GAM: Well, as I remember this whole thing, we found that we could not interchange tapes between the IBM machines and the CDC machines.

JD: That's right.

GAM: And, one of the things was that the inter-record gaps had to be adjusted, because the CDC record gap spec was not as tight as IBM's, so we made the record gap an inch instead of three-quarters of an inch and it made the problem go away.

JD: I didn't know about that.

GAM: Well that's what I'd heard.

JD: Yes, and that's probably so. Because that was an easily-adjusted thing. Do you know that one of those CDC guys works as a salesman at Good Guys?

GAM: Yes, yes, I remember that.

JD: OK, anyway, he's still works there. Occasionally, I see him. If I need to know something that I think he knows... Are you aware that, in the business of making a tape drive or a high fi set, if you make a totally transistorized tape reader, or an ordinary CD reader, or whatever, that people are not as happy with it as when you put vacuum tubes on top of the chassis? Are you aware of this?

GAM: No, really?

JD: That if you walk in and see this beautiful black box with six radio tubes on top of it, it's brand new and you ask the guy, "Come on, tell me what gives. What are you guys doing?" Transistors may be wonderful, but he says, "If we put the tubes up there, they sell much better."

GAM: Really?

JD: Well, it turns out that transistors make the sound crisper, but who recognizes what real sound is nowadays, and when you're young, 18 to 22 or whatever, you're not happy with crisp transistor sound. You think it has to be overloaded somehow.

GAM: Yes.

JD: So you don't argue with them. Now the real problem is you need power tubes.

GAM: Yes.

JD: So where do you go for those? Russia. They still make them.

GAM: They do, huh? Aren't we making such tubes in this country?

JD: Nope, at least not that I was able to find out.

GAM: Oh, that's amazing.

JD: I suppose we do in amateur radio, I'm not sure. But I'm not really in amateur radio anymore—I am with the little two meter hand held things, but I can't get on the air without the lady next door losing her ability to watch the soaps. And that has been something that I have not been able to change, and so I don't get on the air.

GAM: Well really, that's just a matter of one of the major harmonics of your broadcasting.

JD: That's right, that's right, but it's there and she loses the ability to listen to that soap and...

GAM: That's bad.

JD: That's bad, that's right. Although I've had one suggestion about how to handle it: Give her the phone number of the FCC office in Washington, the one with the menu, "If you have a touch tone phone, please touch 1 now."

GAM: Yes.

JD: They say that, by the time she worked her way though that, she'd have forgotten where she was. Well, I haven't done that, and her husband works for the Lab, and he's done a good job out there, and so I just let things slide.

GAM: So, all things considered, you ended your career in late 1979 essentially.

JD: That's right, October 31.

GAM: And you started it at the Laboratory in about 1953.

JD: That's right.

GAM: So that's a generous tour of duty.

JD: That's right, and I enjoyed the people. I loved the people out there.

GAM: It was a great, great Laboratory then.

JD: That's right.

GAM: Did you work on the Data Cell at all?

JD: No, that was Marlton's job. I used to watch him work on it. I tried to take jams out of it, and I was all thumbs, and I think maybe the reason I didn't do well on it is I didn't want to. And they were—what's the word for something that's everywhere?

GAM: Ubiquitous!

JD: Ubiquitous, they were ubiquitous.

The strangest call I ever had was from one of the major corporations in Silicon Valley, up the peninsula a little ways, where somebody had a tape drive and they couldn't make it load. They had a controller for it and everything. They'd done all these good things, and so I was sent over there and drove across the proper bridges and into San Mateo County, found the right person and was taken into the room where the machine was sitting, connected to God knows what. I said, "OK, I guess you're going to have to show me what you're talking about." So, the guy got out a reel tape and loaded it backwards. End of story. I had him sign my slip and I was out the door before I even saw what was going on in the place.

GAM: Oh, wow!

Well, you consider the work on the Photostore was the best stuff you were doing? You worked on all the other machines, the computers?

JD: Well, The 704 series and 7090, the 7094. Do I have that right?

GAM: Well, 704, 709, 7090, the 7094, and the Stretch, the 7030.

JD: I never worked on the Stretch, although I was actually trained for two weeks on the Stretch in Kingston, NY. You know more about the Stretch than I do. While we were there, I suppose Sid had something to do with it, but, at any rate, it was discovered that the Stretch really had stretched some parameters a bit far. So, a major question was, how much you were going to pay for it and so on. And, what happened next, before I went back there for that school, and I had not yet met Garth Huebner and all those people yet. But, they were in the class which Marlton and I attended. Before I had left here, I had talked with Chuck Broughten about methods of communicating so we would understand each other. I think it's OK to tell this story. But, at any rate, he would write a letter to me depicting as part of the letter what was going on back here at the Lab. Chuck and I agreed thatm if there was some message that was supposed to be sent that he would depict a cartoon of something and this would tell us how things were going. Well, anyway, at ten o'clock in the morning or so, a young lady walked in with some mail for us.

GAM: Back there?

JD: Yes, back there, this was in Kingston, not Poughkeepsie. She handed the mail to the instructor and the instructor looked at the letter and handed me my letter and it was from Chuck. I had mixed feelings about just ripping it open at that instant, but I really was curious to see what was in that letter. So, I carefully got it open and looked at the letter. Marlton knew about this arrangement, and I looked at the letter and here was a picture. It was a picture showing a Customer Engineer under water, going down, with bubbles coming up. And maybe the Stretch was also somewhere next to it. But, at any rate, it was obvious we had to rethink where we were. I was on my way back to Livermore the next day and I was fairly happy about that.

GAM: Well, here's what happened actually. Do you remember Von Holdt?

JD: Yes.

GAM: I believe it was he who discovered that the Stretch was not able to produce the speeds that we were expecting.

JD: Oh, OK.

GAM: And Sid took that and tried to cancel the contract, as the Computer Sciences Corporation had cancelled their contract. They were going to buy a bunch of Stretchs, remember? But the Stretch wasn't going to be as fast as had been promised, which was to be 240 times or so faster than the 704. So, IBM said (to Sid), "We think you are not doing the calculations correctly, so we'd like the chance to send out a team of mathematicians and programmers to see if we can improve the algorithms you're running. So, Sully Campbell came out heading this group, and he had about seven people with him. They spent something like two weeks—they had been specially cleared, so they could look at our big codes. And, at the end of two weeks, he said, essentially, "I have to say that there's nothing wrong with what you're doing, you're doing it better than what we thought could be possible, and yet we don't see any increase in speed. They finally decided the speed difficulties were because of mismatches between program architectures and the "look ahead" unit, and things like that, but rather than let Sid cancel the thing, IBM adjusted the pricing in very special ways. They essentially lowered the Stretch CPU to a hundred thousand dollars, but the peripherals maintained their original costs. It still cost a dollar a bit for the memory, and so Sid said, "Well, I'll buy two Stretch CPUs", but they wouldn't sell him two. Sid was very, very clever that way.

Anyway, our official opinion about the Stretch is that it paid for itself, and more, by being able to turn some designs around during the 1962 Christmas Island operation. So, in that sense, it was a success. Its biggest problem was that it didn't really have all the software it needed.

JD: Well, there's a point. We had to learn to write software with many millions of words. To me, that's one of the many remarkable things that is still going on. What was it, that Von Neumann said? "We can do all these things with two thousand words of memory."

GAM: Yes, he said less than two, and he was right. And Edward Teller, when I interviewed him, made that comment that when we got the LARC, we had ordered thirty thousand words of memory on it and Von Neumann said, "That's too much, you don't need that much." Edward said, "I was too smart to disagree with him." But, when I got back to the lab, I called him up and asked him, "Can you come out and visit us?" So, he came out and basically what they said to him was, "Well, Johnny, if you're a Von Neumann, you can get along with Four Thousand words of memory, but all the ordinary people need thirty thousand words and maybe even more. Von Neumann withdrew his objections.

JD: Isn't that nice. And that's right on the nose.

GAM: Well, even he hadn't plumbed the depths of what does it mean to do arithmetic over a huge mesh. And what is this constant movement of data in and out with these inadequate memories. Now, I think that, though it's a story not yet told, the development of very large codes as was done at the Laboratory, for whatever reason, was fundamentally different from the stuff that goes on in colleges and ordinary little businesses. They have big codes, but they don't sweep through the entire data sets all the time; they'll be looking at a data entry or they're proving a number theorem problem or something like that. So, what we did with large-scale problems, was quite different from what was ordinarily done with computers and I think that made quite a difference. Anyway, Sid was not able to buy two Stretchs. But, we had the one, and we made it pay for itself. And, as you know, when it got surplused, talk about unbalance, Lowell Wood bought the most of the main frame, but not the memories, for three thousand dollars, He didn't buy the entire machine, he bought the CPU, the logic unit. Someone else bought the typewriter, you know the golf ball typewriter, for six hundred and fifty bucks whereas Lowell paid only three thousand for the entire computer, except they didn't sell him the memory because those were still usable on the 7094's.

JD: Oh, that's right, 7302 memories.

GAM: Yes, I remember Norman taking me in and showing me, we could peek into the little window there, we found the right cores to watch, and we watched the bubbles coming off the thing because it was being cycled too frequently.

JD: Norman?

GAM: Norman Hardy, a genius.

JD: That's right, that's absolutely true. He was one who used to show up in the middle of the night as we finally got a computer running and would have the first program to drop in to test it out, to see how it worked.

GAM: Yeah?

JD: Remember the 780? The 780 was the large CRT.

GAM: Yeah, the direct view.

JD: The direct view CRT. He had built a one-card program to run this view of a sign. He took the card over to a 011 card punch and he plugged a hole and brought the card back over and dropped the card into the card reader. But, while he was gone, John Delong had gone back behind the 780, pulled out two wires for vertical and horizontal and reversed them. Norm brought his card back and dropped it into the card reader, hit the start key, turned around and instead of doing this, or whatever he expected it to do it was going the other way.

You could see all these wheels turning in Norm's head. They completely described what had been done, what had happened, and it's no different than the story of, I don't know whether Bob Cralle did it, or who did it, who fixed up the display device, so that when Norm would write in numbers...I'm sure you've heard this story.

GAM: Yeah, I was involved with it. It was on the dd80, and the demo was using our light pen, to show handwriting recognition.

JD: Oh, OK, fine.

GAM: That blew him away too.

JD: That's right, and I was in the room when that happened. You could see the wheels turning for that one too. What did the display say? I think it said "That's Norm Hardy."

GAM: It said, "By God, you are Norm Hardy; Hi Norm!".

JD: Ah, dear.

I always loved my Australia experience. I spent three years in downtown Sidney, working on the eighth floor of a ten-story building called the Grace Building. And, as far as I know, we're talking about Grace Shipping Line. I wandered around, had a wonderful time. At lunchtime, I would go down onto the main drag. It was a big wide street called Martin Place, and I would eat at a restaurant down there, Purdium, and then come back to the office, eventually, and turn out a few more bits of work, which was ordering, not computer parts, but radio parts. As an example of one of the more exciting few minutes, I was walking through Martin Place, and heard fast footsteps behind me and decided it might be a good idea to turn around and see what was taking place. What I saw was two Aussie MPs, Military Police, chasing an Aussie soldier who, I suppose, was AWOL. As he ran through the crowd on this lovely spring day, he would dodge behind people and move around and so on. The two soldiers chasing him couldn't make any progress because, as he went through the crowd, the crowd would step out of the soldier's way and into the way of the military police at the same time saying, "Go it, Aussie." In other words, helping him escape. That was just one of many experiences. There was a bookstore I loved to go into. It was D-Y-M-O-C-K. So, I would buy books there that I had no way of getting home, except there was a nice math book that I purchased and, in the math book, was a slip of paper I never looked at it until very recently, and it was signed out to Sergeant D-Y-M-O-C-K. And that was the first clue I ever had that "Dimmick" might be pronounced some other way or spelled some other way. I was on the internet and here was their ad, and I wanted a certain book and so I ordered it from them, and it came about two or three months later. It was the first time I really made the Internet mine. Since then, why, I can make mistakes, but I can't do any wrong.

GAM: You didn't go any further into the way you spell Dimmick?

JD: Well, in Scotland, I'm told there are lots and lots of Dimmicks. I've never really checked up.

GAM: It would be intriguing to follow up on.

JD: One thing I have done is VON. Do you know what VON is? You probably do. Voice On Network. It turns out that, on VON, there are various chat rooms, the usual thing, and there is also a ham radio list with call letters from all over the world. I put my call letters down and the first thing you know, one of these pictures comes to life and some guy's there with his call letters and we have a QSO. QSO being the ham radio equivalent for a chat. It's one of the Q codes. I've talked on that, probably, two dozen times. The closest one is in San Francisco. When I was first getting started on it, and I had some questions. Neither the guy nor I were able to really make everything work the way we wanted it to. We'd gotten to the point where Ma Bell had to be involved in it. I realized that, since he was in San Francisco, it wouldn't cost me too much to make a call to San Francisco, so I unhooked the computer and hooked the telephone back up again and asked for information. I had the guy's name, because if you push the right button, you find out what the guy's address is and everything else, so I called him. He said, "Oh, you're the guy I was just talking to". And I said, "What were we doing wrong?" And, so, we discussed this and figured out what was wrong and the next time I was on the VON, it went much smoother. It wasn't too much later that I found myself talking to a young man, he said he was 17 years old and he said he was in Durban, South Africa. He was telling me he was kind of disturbed about the kind of politics that was going on in South Africa at that time. It was never very smooth, but he told me he thought he would go out to Australia. If you're up here you go "down" to Australia, if you're in South Africa you go "out" to Australia.

GAM: See, there's a sense of handedness we have here.

JD: Yes, that's right, and, so, the last one I'll tell you about is a guy I've been in regular e-mail contact with in Christchurch, New Zealand. Here was a guy, a ham radio operator, who was near Christ Church. I had him describe it to me and I was really pleased. We had something to talk about, not just the weather.

GAM: Or the gear you're using.

JD: Or the gear we're using. So I said, "Oh by the way, a very dear friend of mine did some research during the last few years, before her death, on how Livermore got its name. This, of course, was Janet Newton. It came about because of a British sailor by the name of Robert Livermore jumping ship in 1835 in San Francisco bay. He says, "That's interesting, that's interesting, that's interesting.

GAM: OK, let's hear the story of the dummy.

JD: The story of the dummy. Do you remember a cartoonist, Virgil Partsch?

GAM: Yes, very much.

JD: There was one picture he had that I always loved, and it was of the guy who had his Techtronics scope, I think it was an ad for Techtronics, and he's trying to see a pulse of energy on the scope, and he has reached right on through the computer, and the squiggle is still on it's way, and he hasn't reached it with the scope probe. Well, this really, somehow or other, resonates with all my feelings of the years I have spent with a scope probe watching the shape of a pulse. You put energy into it, you wish it well and so on, but you hope you will see it so you can find out what is wrong with the machine. When we unhooked one of the 704s, or was it the 70l? When we unhooked this one machine, we talked and joked about the amount of time we had spent looking at pulses with a Techtronic scope and it was a sort of conversation amongst us, you remember all the hours you've spent doing this?

GAM: Yes.

JD: Somehow, the words escape me, I guess they came from me. We had visions of actually opening up the machine when it was received in Poughkeepsie and finding a guy in there still holding onto a scope probe.

GAM: Yes, great.

JD: I went off to school on the new machine and, in the meantime,,,, I heard that somebody had actually put a dummy into that machine. I thought it was funny at the time, but I didn't realize how it would be accepted and received at the other end. There was no warning whatsoever, they got the machine back there, opened it up, and here was this body with a scope probe, the whole thing. At least three VP types made a trip out to Livermore to have a long conversation with us to be sure we didn't do that again. Sid, when he finally heard about it, I think I told him years later, laughed. You can imagine him. I don't know that anybody lost his or her job.

GAM: Well, they shouldn't have.

JD: They shouldn't have, but God-O-Mighty, these guys were out for blood. Now one more story.

GAM: OK.

JD: It had to do with the tours around the plant that each class received. They went down through a sort of open area where there wasn't very much activity, and shoved way back in the corner was a 733. The 733, if you'll remember, was a drum unit, about so big, great big drums go around, and on the side of it was a sign where it was from: "Livermore Radiation Lab." I think it was all spelled out, and the fact that it was contaminated. I may even have written on it "contaminated", not realizing that it had a special connotation. And the thing had been shoved way back there so nobody would get close to it and get irradiated. So, we got that sign out of there. Lithium Chloride is what was in it. It was from that first air conditioner in the 701 Building.

GAM: Boy that is a long time ago. That's great.

JD: During my very first class in Endicott, a group of 15 to 18 of us were being toured through 590 Madison in New York City. We were upstairs where T.J. Watson senior's office was, and when we were shown his office, one of my fellow students had gone on in there and sat down with his feet up on Watson's desk. I was behind him, so I could see how he was using this desk. He didn't see me do it, but I went out into the hallway and I started towards the open door in the hallway that would look in on him at that desk. I walked like this, scrape, scrape, scrape, and I went on around the corner. He heard that, and he couldn't get out of the chair because it was leaned way back and his feet were on the table and he couldn't do a thing. How wonderful.

GAM: That's great.

JD: OK, enough of IBM.

GAM: OK, well, we have run on, but it was fun. We'll stop here and let me thank you for a truly interesting chat.




For information about this page, contact us at: comment@computer-history.info