Clive Dawson
Jul 15, 2026
Noah Smith: Let's start by saying as an introduction that we're getting an oral history interview with Clive Dawson. This is July 15, 2026. And Clive, we would like to confirm again with verbal consent that you agree that this becomes public and this recording will be open for public use.
Clive Dawson: Yes, absolutely.
Noah Smith: Okay. We're going to start with just warming up and learning about your background.
Clive Dawson: Sure.
Noah Smith: I know that you've mentioned speaking Spanish and I'm very intrigued to learn about where you came from, your family, where you grew up.
Clive Dawson: I'll give you a brief history and feel free to cut me off if I get into the weeds too much. I was born in Mexico City, Mexico. My family basically were expatriates. We lived in Mexico for many years. My grandfather originally moved to Mexico from Chicago in 1902. The president of Mexico at the time decided that he wanted all of the public buildings and the cathedral in Mexico City decorated with colored lights for the holidays. And he hired an electric firm from Chicago to come and do the job. My grandfather worked for them. So he and his crew came down to Mexico by train and worked on the lighting project, which was challenging because colored lights had not been invented yet. They had a lot of false starts trying to figure out how to do these colored lights without them melting or exploding from the heat. But anyway, they finally got it done and it was a great success.
And then he went back to Chicago and arrived in the winter of 1903-1904, which was the worst winter that Chicago had had in the last 50 years. It was sub-zero weather with 20 inches of snow. He took one look at what he had returned to and decided, "What in the heck am I doing here?" He turned around, went right back to Mexico, set up an electric shop downtown Mexico City. Eventually he married, my mom was born, she worked for the US embassy, met my dad, and the rest is history. I went to the American School in Mexico City, which is the school she had attended. In fact, I had several of her same teachers. I didn't come up to the US until I graduated from high school, at which point I went to college at Stevens Institute of Technology. I had been there the previous year as part of a National Science Foundation summer science training program for high schoolers. I thought that being right across from New York City on a wonderful campus with a fantastic view was great, and I liked the place so I applied.
Noah Smith: The Stevens Institute. Is this in New York City itself?
Clive Dawson: No, it is in Hoboken,
Noah Smith: New Jersey
Clive Dawson: New Jersey on the Hudson River. Just across from Manhattan. One parenthetical story. Stevens was short of dormitory space, so they actually bought an ocean liner that had been in use back in the 40s and 50s as a transatlantic passenger ship. They anchored it on the Hudson adjacent to the campus and I lived aboard a ship for three years from 1968 to 1971. Those were precisely the years when the World Trade Center was under construction. I would look out my porthole every morning and I could see that thing as it rose from nothing to its full height of 110 stories. And of course, 30 years later, I watched it come down in 15 seconds. That was a very traumatic experience for me and for all the Stevens people.
Noah Smith: So, your family, were they affected by the Mexican Revolution?
Clive Dawson: Yes, as a matter of fact, they were. I mentioned that my grandfather started an electric shop in downtown Mexico City. Next door to him was an apartment building where a French family lived who ran a dry cleaning business. They had a little daughter who was about five years old. My grandfather was in his mid-20s at the time. He would see this little girl coming and going from the next door apartment. When the Mexican Revolution started in 1910, the family decided that it was too dangerous for their daughters to live in the middle of all of this. So they sent them to San Antonio and they went to school at Our Lady of the Lake. By the time they returned to Mexico, she was a very nice looking young lady and my grandfather started courting her, even though there was about a 20-year age difference, which I guess at that time wasn't all that uncommon. So they ended up getting married and then my mom and uncle were born. The Mexican president who wanted to do the lighting project was Porfirio Diaz, who is considered by many to be a dictator. It was amazing who you talked to because if you talked to the Americans in Mexico, they thought he was the best thing ever. He was very sympathetic to foreign investment, but history was later revised because he didn't really treat the peasants and the farmers very well. So there are two sides to every story.
Noah Smith: Understood. Clive, do you have Mexican citizenship?
Clive Dawson: I do. As a matter of fact, I had Mexican citizenship by virtue of being born in Mexico. My mother had American citizenship, but as it turns out, one of the US rules is that citizenship doesn't pass down to the child unless the parent has actually lived in the US for a certain amount of time, which my mom never did. Therefore I did not automatically obtain American citizenship. This of course is a very hot topic nowadays with birthright citizenship. I was perfectly happy to keep my Mexican citizenship because at the time when I came up here it was 1967. The draft lottery had been conducted in 1969, my number was too low for comfort, and I thought my home country is fine for me. I have a sentimental attachment to it. So, I stayed Mexican. I came up on a student visa but my graduate school career was extended by several years because I was having so much fun with the DEC-10 in Texas. When I got married in 1978 is when I first got my green card by virtue of the fact that my wife sponsored me. Fast forward to 2001, 9/11 comes along. As I mentioned, this was a trauma for me. I decided then that I had been a guest in this country long enough. So I got my US citizenship at that time and I'm now a US citizen.
Noah Smith: Nice. Great story. So you've already mentioned actually a little bit, but the question is how did you arrive at UT? It was for grad school?
Clive Dawson: I graduated from Stevens in 71 and started at the University of Texas in the fall of 71.
Noah Smith: How did you make that choice? How did you choose to go to UT?
Clive Dawson: I had visited when my parents had come up from Mexico City to San Antonio for spring break and I met them down there. My sister came down from a college in Minnesota, so it was a little mini family reunion in the spring of 1970 in San Antonio. I was starting to think about grad schools. I had my eye on Stanford and MIT as I was firmly convinced that I was going to be studying artificial intelligence. But here I was 70 miles away from Austin and I thought I'm just gonna run up there and take a look at the UT campus. It was spring break, so there was nobody around, but a professor had come in from home to meet me and interview me. Not only was he very impressed with me, but he offered me a teaching assistantship right on the spot.
Noah Smith: Beautiful.
Clive Dawson: And the campus was great. It was close to Mexico. That was essentially what made the decision for me.
Noah Smith: This is a great story. This is 1971. You were looking at Stanford and MIT. You decided to come to UT.
Clive Dawson: Yeah.
Noah Smith: What was it like when you arrived? What neighborhood were you living in? And what was your first experience at UT?
Clive Dawson: I moved into a dormitory. I didn't know anything other than dormitories at that time, having lived my entire time at Stevens in a dorm or the ship. There was this gigantic dorm on the UT campus called Jester. When I first came, they did not have a graduate student floor. My first roommate was a Korean grad student and then there was a guy from Arizona State who was also a grad student. Eventually they had the top floor dedicated to grad students. I stayed at Jester for a total of three years before moving out to an apartment in the campus area.
Noah Smith: Do you remember where you were when you moved off campus? Did you move up north of campus?
Clive Dawson: Yes. My first address was the Villa Arcos apartment on 3301 Speedway. Then the next one was after I had met my wife-to-be. She was a grad student as well in sociology and we met at Jester. We like to say that we met in the back of a police car because she was doing her master's thesis at the time on police training. She had gone down to the Austin Police Department and spoken to the chief, and said, "I'd like permission to attend your police academy because I want to do a thesis on how police are trained." He was very enthusiastic about it and said, "Absolutely. Come on. Just give us a copy of your thesis." There were no women police officers at the time and they welcomed her into their class as one of their own. She went to the firing range, she did all of the practical problems, the classroom stuff, and the physical calisthenics. The only thing they wouldn't let her do was when it came time for the cadets to be assigned to real police officers to experience a shift on the street. They said, "No, you can't do that because it would not look right for a woman not in handcuffs to be in a police car with an officer. The public's going to think it's his girlfriend."
Not to be discouraged, she came back to the dorm and said, "I need somebody to be my companion." And I thought that sounded pretty cool, and I liked her anyway. So we would go down to the police station at 10:00 on a Saturday night. They would assign us an officer and we would ride around until 6:00 in the morning. That turned out to be so much fun and so interesting that we made friends with all the officers. Long after she had finished her thesis, our idea of a Saturday night date was to go down to the cops and ride around until six in the morning. Around 1975, we did what all good couples do which is we rented apartments right next door to each other up on 39th Street. We got married in 1978 and she had a position at St. Edwards University here in Austin. She was a professor of sociology and criminal justice and spent 37 years there, ending up as dean of the School of Social and Behavioral Sciences. She eventually had to retire back in 2013 because she got Alzheimer's disease and she is currently in memory care.
Noah Smith: So let's look a little bit at the institutions and should we talk about the computation center, or should we talk about CS, or both of them together?
Clive Dawson: Well, let's start with CS. The one thing that really pained me about coming to UT rather than to Stanford or MIT was that I no longer had a PDP-10 to play with. Stevens had gotten a DEC-10 around 1969. I worked for the computer center there. They had a very wise director who basically said, "This computer center is going to be run by students." He had a skeleton staff of responsible adults overseeing things but all of the operations and systems programming were done by students. That's how I learned about the DEC-10. I became a systems programmer doing operating system installations and things like that. When I arrived at UT, I discovered they had a CDC 6600, which was several times more powerful than the PDP-10. But then I had to go and use the key punch again to prepare decks of punched cards. It was just agony, like a return to the dark ages. They eventually got an interactive terminal system called Taurus, which let you sit at a dumb terminal and interact with the CDC system. That got rid of the punch cards, but all of the tools and editors were so primitive compared to the software the 10 offered that I was still suffering.
Now, I had to work through three PhD advisors. The first one was Bill Henneman, who had gotten his PhD from the MIT AI lab. Very interesting fellow. He was a high school dropout and he showed up one day at the MIT AI lab and impressed Marvin Minsky and Seymour Papert sufficiently that they admitted him into the AI lab on the spot with no high school or college degree. He eventually left there with a PhD. But he didn't get tenure at UT, so he was gone after two years. I had to go to another one, and then he left. Finally, around 1978-79, I went to my third adviser. When he decided to leave, I said I'm not going to go to a fourth. I've got to get my thesis done. I had finished all the course requirements back around 74. In fact, I arrived with a master's degree because I had gotten the master's simultaneously with a bachelor's at Stevens.
When I arrived at the CS department at UT, I was cocky enough to think that I could pass the qualifiers and just start writing my dissertation. The faculty took a dim view of that. They said, "You may be here with a masters, but you're going to have to pass our qualifiers, and to do that, you're going to have to take our courses." So I had to take two years worth of CS courses, which actually turned out to be great because it rounded me out as a computer scientist in things like numerical analysis and automata theory. During that time, I had been offered a teaching assistantship. I was teaching introductory computers to freshmen and I loved that. I really love teaching. But around spring of 75, I had run out of eligibility to be a teaching assistant. They tried to keep me on as an assistant instructor, and at one point they made me an assistant instructor trying to keep me on.
But finally they said, "Okay, look, you really have to get out of here." By coincidence, it was the spring of 75 when the DEC-10 arrived on the UT campus. As I wrote in 'The Soul of an Old Machine,' I greeted that with the joy of a prisoner being released from the dungeon. I walked over to the Comp Center, went straight into the director's office and essentially said, "Here I am. I know that system backwards and forwards. Hire me." They hired me and I became one of the main systems programmers for the DEC-10 at UT. They installed it over at the HRC, now known as the Harry Ransom Center, on the third floor. I met my boss, Patricia Caroom, and colleagues like Tommy Loomis and Rick Watson.
Backing up to the CS department for a minute, during those first few years I was busy taking courses. Then I passed the qualifiers and started writing my dissertation and made friends with fellow grad students including Rich Cohen and Mabry Tyson. It was a very stimulating environment. Artificial intelligence was a shadow of what it is today, but that's what I got my degree in. I wrote my dissertation on pre-processing and problem-solving systems. I was doing robotic simulation problems with the "blocks world," like stacking blocks, and how to get software to make robust plans to do those types of tasks.
Noah Smith: It's that MIT program that I don't know how to pronounce.
Clive Dawson: Yeah, SHRDLU. There was an MIT grad student there by the name of Terry Winograd, and he was the one that did SHRDLU. That was the blocks world. Everybody started using that plus Shakey at SRI, which was a physical robot, and everyone started using that as a good model for improving problem-solving systems.
Noah Smith: Understood. So, Clive, you actually already hit on the next question. Let's back up a little bit. HRC. So, the question is, how is it that the DEC-10 ended up in HRC? What was the sequence of events for that? And was this surprising?
Clive Dawson: I was vaguely aware the building existed, but I'd never been in it. I don't know the circumstances because by the time I got hired it was a done deal. I think it was just a question of needing space, and they figured out they had some over there.
Noah Smith: Okay. This would be the simplest possibility, right? So, yeah, in a few minutes we'll get to the question of the Linguistics Research Center which was also there in HRC.
Clive Dawson: And right across the hall.
Noah Smith: I wanted to ask you, would you say that there was anything special about the atmosphere being there in HRC?
Clive Dawson: There was certainly very nice synergy between the Linguistics Research Center and the Comp Center. The HRC has been completely remodeled since those days. They basically gutted the third floor; it's now like a huge reading room. I had the impression they actually raised the ceiling. The third and fourth floors are kind of a public library reading room type of place.
Before then, it was just hallways with closed office doors. There was a central core surrounded by four hallways, which is where the DEC-10 machine room was. On the western side was the Linguistics Research Center. Those folks gravitated to the DEC-10 very quickly. There was one researcher there named Dr. Helen-Jo Hewitt who was an amazing woman and a font hacker before that was even a thing. She realized the DEC-10 had some capability. This was back when you had those daisy wheel robotic typewriters you would type the document, pause, roll back the page, pop out the English daisy wheel, pop in the Greek one, and then fill in the Greek letters and any diacritics. One of the projects I did for her was a series of TECO macros that would do all of this automatically. It would go through a document and insert the necessary escape sequences to pause and roll the page back. You would switch out the daisy wheel and it would automatically skip to the spots where foreign characters needed to be. She would just sit back and watch; the only thing she had to do was change daisy wheels.
That became a real challenging series of TECO macros. One of the main activities of the Linguistics Research Center at the time was machine translation. They had a project with Siemens Corporation for German to English translation called METAL. All of this work was being done on the DEC-10 before it eventually moved to the DEC-20 when I took over that.
Noah Smith: I'm glad you said this. I for some reason had gotten the impression that this was still on the CDC. I'm happy to know that it was on the DEC-10.
Clive Dawson: Well, I think they did have some software on the CDC, but they got over to the DEC-10 with all due haste.
Noah Smith: I'm glad to have it confirmed.
Clive Dawson: Plus, it was so convenient for them to have the computer right across the hall. They had terminals in their offices and a line printer right across the hall. They loved having the computer nearby.
Noah Smith: I'll just note real quick that we just learned a few weeks ago that Tommy Loomis actually continued working on METAL with Siemens for many years after.
Clive Dawson: Yes. He actually went on. I believe he went to Germany, in fact.
Noah Smith: So, this is something to follow up on later, but yeah, this is a long story.
Clive Dawson: Right.
Noah Smith: Um, I wanted to ask you about a few other names that are the next questions. So, the name George Culp certainly, but also the name Jim Browne. I think two professors.
Clive Dawson: Jim Browne was definitely a force of nature and he was chairman of the CS department for many years. He didn't like me very much. And the reason for that was because back then, and this is one of the reasons my graduate student career extended for so long, I would go into Bill Henneman’s office and we would sit playing chess as we talked about AI and MIT. Jim Browne would walk by the open door and see us playing chess and he thought that was just such an utter waste of time. I'm sure it probably affected Bill’s career directly and it affected mine indirectly because I should have been working on my dissertation. But after I got my degree all was forgiven. He and I worked closely together because he was the chairman of the Faculty Computer Committee.
And the Faculty Computer Committee was basically responsible for plotting the direction of the Computation Center and their acquisitions. They were responsible for raising the funds to obtain the DEC-10 and DEC-20. Jim Browne was there for many years. A lot of students got their degrees under him. He had an outside company called SES doing work on databases and languages. Students like Forest Baskett and Tom Keller were closely associated with that company.
Noah Smith: Understood. This makes total sense. They had worked with him from early on. From the late 60s. Jim Browne was their main, I think, faculty sponsor.
Clive Dawson: Yeah, he was their mentor, advisor, and so on. He and Woody Bledsoe were probably the two most powerful people in the CS department in terms of policy and direction. George Culp had office space in the DEC-10 complex over at the HRC, so I got to know him pretty well. I wasn't really involved in his computerized education CBE project, but he constantly needed help in getting things to run on the 10. So I would help him with that. I understand that he's still living out there west of Austin, in Johnson City.
Noah Smith: That's good to know.
Clive Dawson: And he would be a good person to try and track down.
Noah Smith: I will make a note of that. Any other names from that era that you think of?
Clive Dawson: Another professor that brought in a lot of money to the CS department was Don Good. When I first arrived, the comp center was in Waggener Hall. I think we had a discussion of this online. Waggener Hall was immediately east of the Comp Center under the steps, between Speedway and Inner Campus Drive, and it housed the Classics Department and Philosophy and I'm thinking even maybe a chunk of Geology in the basement. That's where the CS department was and that's where I met Dr. Angus Pearson who interviewed me. He and his wife were beloved faculty members and I was his teaching assistant for CS 404G, the introductory course.
The CS department eventually moved over to Painter Hall but quickly outgrew the space because it was sharing it with Physics and Astronomy. Astronomy had the telescope up on the roof of Painter. Physics had labs in the basement. CS had the second, third, and fourth floors, but it wasn't enough. Don Good was the head of the CMP, the Certifiable Minicomputer Project. They ended up getting space in the tower. Around 1975, the Perry-Castañeda Library opened across from Jester. Speedway and 21st Street. So the main library, which had been in the main building including the stacks in the tower, moved over to PCL. This left open space in the stacks and the CS department snapped up a floor or two for the CMP. Rich Cohen worked for them. We eventually had to get a cable from their systems down to the Comp Center for the IMP because they were hooked to the ARPANET back then.
The only other thing I'll mention about the space in the tower is another story. I'm an amateur magician and I was told when I arrived that UT had acquired a collection of Harry Houdini's papers. I was very interested in tracking those down. And so I went to the library and I said, "Where can I find this?" And it took a while, I finally got to somebody who said they had all that stuff on the 19th floor of the tower in the stacks. I take the elevator up to the 19th floor, and when the door opens, I am greeted by a room full of cardboard boxes stacked every which way. It was just a pile of junk boxes everywhere. There was nobody there. I walked in and started poking around. I open some boxes and all of a sudden I'm reading original Harry Houdini letters. All of his stuff was sitting there scattered around in cardboard boxes on the 19th floor, open to anyone. I was absolutely horrified. I'm happy to report that the Houdini collection was transferred to the HRC. It is now a curated collection. It's one of their star things and you can go to their reading room, put on the white gloves, and peruse the material to your heart's content, they have exhibits of it and so on. But I just remember getting my grubby hands into those boxes and hopefully I didn't do much damage. CS was scattered in places like Painter Hall and the tower.
Noah Smith: That's the good old UT.
Clive Dawson: Yeah, exactly. So anyway, CS was scattered pretty much in all sorts of places. Painter Hall, the tower. Several people, including Woody Bledsoe, who was actually a joint appointment between Computer Science and Math, had his office over in PMA that then became RLM that then went back to PMA.
He was up on one of the higher floors. That's where his office was. He was a member of my PhD committee. Since he was intimately connected with AI, in particular automatic theorem proving. He's one of the world pioneers. He was a very good friend of mine. He sadly eventually got ALS and passed away in 1995. He came to me for help in trying to get automated speech onto his computer so that he could continue using it as long as he could.
I would go to his house and help him set up various pieces of hardware. I was in touch with him until he passed away. Woody Bledsoe was probably the professor that I was closest to throughout my time at UT. In 1984, when MCC came to town and Bobby Inman became the head, he hired Woody Bledsoe to head up the AI lab at MCC.
Woody Bledsoe turned around and hired me to run the AI computer infrastructure at MCC. The first thing I told him was that I wanted to get a DEC-20 for MCC. That was a little tricky because in 1983 DEC announced the end of their 36-bit computers. We knew the DEC-20 was not going to have any follow on, and yet I convinced Woody that I could keep it running forever.
Noah Smith: So there was a DEC-20 at MCC.
Clive Dawson: Right. We started with the DEC-20 at MCC and basically graduated to Symbolics Lisp machines. We were the largest installation of Symbolics Lisp machines in the country for a while. I ran all of them plus the 20 until I left MCC in the late 90s.
Noah Smith: MCC was, I've heard, the greatest concentration of Lisp machines in the world.
Clive Dawson: I wouldn't be surprised. There were two basic flavors of Lisp machines both arising originally at MIT, built by faculty and grad students. There was a split in philosophy that gave way to the Symbolics Lisp machines and the LMI Lambda machines. We went with Symbolics, and they even put their offices in the building MCC first occupied before they moved to the main installation on Braker Lane.
Noah Smith: I had the impression that your dissertation might have been related to work at CMP, but now it sounds like it's because of Woody Bledsoe.
Clive Dawson: Woody was not directly involved with CMP. That was Don Good. They were working on theorem proving in the sense that they were trying to do formally verified software for hardware used by the defense department, in particular the NSA. They had a box delivered to them by the NSA called the BCR box, which is what was connected to the ARPANET. They ran software on that and used the DEC-10 and later the DEC-20 to run their verification programs to develop software.
Noah Smith: So the next topic is to talk about the dawn of the ARPANET on campus. So maybe we can dive into these together.
Clive Dawson: Let me just mention that the authoritative story on CMP is going to be Rich Cohen. He worked for CMP and was part of that project, so he knows a hundred times more than I do about that. I was going to suggest that maybe you would want to do a joint oral history because he and I feed off each other and refresh each other's memories really well. But we seem to be doing okay so far here. That was CMP, and I think we've covered the CS department in general.
Noah Smith: I would love to talk to Rich. On the outline, we've come through how the DEC-10 arrived and we want to look at how the ARPANET arrived as well and maybe finish up with what the data day-to-day life was like there in the computation center.
Clive Dawson: Yeah. So, you know, I got an office over at the HRC right next to the DEC-10 machine room and staff of about five or six. A few systems programmers and user consultants. We developed into a very close-knit group. You can read about that in the piece I wrote when we shut it down in 1982. We had T-shirts made that said, "DEC-10: We share our time." That was a close group.
And one day, because of the CMP project, we heard that an IMP was going to come to campus and they didn't really have a place for it. So they put it in the machine room with the DEC-10. Mysterious people from BBN in Boston came and hooked it up. They basically said don't touch it and left. For at least several months the IMP was simply a node that had no host computers hooked to it. It was just a midpoint between White Sands in New Mexico and Gunter Air Force Base in Alabama. Those long lines stretched across and met in Texas and there was the IMP.
I remember one memorable occasion when my boss at the time, David Phillips, who had taken over from Patricia Caroom and had been running the PDP-11 over at Texas Student Publications, was managing the DEC-10 installation. He worked for the Comp Center, so they brought him over and they had him manage the DEC-10 installation. He became our boss, a very nice fellow. One day he was wandering through the machine room. I don't know if there was some maintenance going on or what, but he accidentally turned off the power to the IMP. It took about 30 seconds for the phone sitting on top of the IMP to start ringing. It was from Boston, and they basically asked what happened to the southern leg of the ARPANET. He turned it back on, it automatically rebooted, and things were okay. That was not something that ever happened again. Eventually we decided that we might as well use it, so what are we going to hook to it? The obvious thing was the DEC-10, but it turned out the DEC-10 did not have the necessary hardware or software to handle an IMP.
We would have had to get software from CMU or Harvard, and it would have been TENEX which was the alternative operating system developed at BBN. That would have been too big of a disruption for the DEC-10 customers on campus. So they decided instead that the PDP-11 would be a much more convenient thing. We got the necessary software from the University of Illinois and installed it on the PDP-11 over in the communications building, in the basement where the printing presses for the Daily Texan were. David knew all those people, so it worked out.
Because of that, the IMP got moved from the DEC-10 machine room over to the communications building. It wasn't until a few years later that the Comp Center got its own PDP-11/70, which then became the obvious choice for the IMP. That's when they moved it over to the Comp Center. It was also around that time that the CMP project in the tower got its BCR box. They ran a cable from the tower down to the Comp Center. By then I was running the DEC-20 over in Painter Hall, so they ran a cable from the DEC-20 to the Comp Center so that the DEC-20 could get onto the ARPANET. The DEC-20s were easily connected. All you had to do was get an AN20 box, which we did, and the software already existed.
Noah Smith: And just a note that this AN20 is now in Clive's garage.
Clive Dawson: It is. And if you want, I can take my laptop and wander over to my garage and show it to you.
Noah Smith: We'll ask for some photos, for you to email us some photos in the next couple days.
Clive Dawson: When the DEC-20 was finally uninstalled at UT, it went over to the Pickle Research Center in North Campus. They have this huge surplus warehouse. I wandered over there one day and said I'd like to buy one cabinet of the DEC-20. They said, "Yeah, 100 bucks." I borrowed a pickup truck and hauled it to my house. That was a totally irrational and unabashedly sentimental thing to do.
Noah Smith: And totally legendary. So was there only one IMP, or was there a second IMP there as well.
Clive Dawson: There was only one IMP, and it had a total of four or possibly six ports. Originally it was the PDP-11 at TSP, and then when it moved to the Comp Center, the PDP-11 stayed installed, the DEC-20 plugged in, and the hardware in the tower plugged in. So we had three hosts hooked to the IMP at that time.
And then when I left UT in 1985 and went to MCC, we clearly needed to be on the ARPANET. So we actually ran a line from Braker Lane, originally from the Echelon buildings on 183 and then later from the main MCC building via the Pickle Research Center, with a microwave link to the main campus and the IMP.
Noah Smith: I have many times looked at this microwave link. I'm fully aware of it. I was guessing we would get to the microwave link. So let's say the first four internet computers in Texas, right there.
Clive Dawson: Yeah. And you know, I claim responsibility for installing number two, which was the DEC-20 in Painter Hall.
Noah Smith: Sounds like you had a lot to do with the one at MCC as well though.
Clive Dawson: Yes, that was maybe number five. But we're talking about the whole state of Texas. Here's an interesting sideline, in the late 80s, it was clearly going to be the end of the ARPANET. CSNet was arriving, and commercial networks were emulating the ARPANET. The Defense Department decided to let commercial people take over, but they needed to keep the skeleton ARPANET for military installations. Unfortunately, Texas didn't qualify and we were in deep trouble because they were going to take the IMP away from Austin.
I got on the phone with the Defense Communications Agency and AT&T because the long lines went out to White Sands and Gunter Air Force Base. By then another ARPANET link had appeared in Dallas at Collins Radio because they had defense contracts. I got the brilliant idea that if we could pay for a line from Collins Radio down to Austin, we could stay on as a spur of the ARPANET. It took a lot of phone calls to the Defense Department and AT&T, but eventually we managed to get a line at MCC's expense. UT Austin was a beneficiary. They ran a line from Collins Radio down to the Comp Center , which allowed us to stay on the ARPANET, which pretty soon became the Internet. The lines to White Sands and to Gunter disappeared but that's how we managed to stay on for a while longer.
Noah Smith: And this is right about when I first got onto campus. My first Internet experience was probably via this exact configuration. So through the UT IMP to Dallas Collins radio and then onto the wider net.
Clive Dawson: Yep. Right.
Noah Smith: Let's back up a little bit. When you mentioned, I think in an email, that when the IMP was first hooked up in TSP, you were able to dial in. How many people were doing that?
Clive Dawson: Right. That was a closely guarded secret.
Noah Smith: I bet it was.
Clive Dawson: They didn't have that many ports, and priority went to the people publishing the Daily Texan. But at night they had a few dial-up ports through the MICOM, which was the terminal concentrator that served the UT campus. So you could dial up to the MICOM and connect to the PDP-11 at TSP. Once you got on that, you could go out to the ARPANET anywhere you wanted.
There was a group of us in the CS department, the linguistics research center, and the math department who knew about this. We would get on and the best place to connect to was MIT, particularly hosts like MIT-AI. This was a brand new concept to us, you'd telnet to the host and start running commands. You didn't even need a password. You logged in just so that people would know who you were but there was no password required. I thought it was insanity. The main force behind that was Richard Stallman at the MIT AI lab, the legendary free software person who was responsible for the GNU project and free licensing.
He came to Austin to give a lecture at MCC and refused to enter the building because they required identification. Bob Boyer had to host the lecture at his house. So we would get on and play Zork. Zork was developed on MIT-DM using a language called MUDDLE. That was very addicting. I don't want to tell you how many hours we spent playing Adventure and Zork on the MIT machines. I was exploring their version of TECO, which was far more advanced than what DEC supported. It became the basis of Emacs, the legendary editor developed by Stallman. It was essentially a collection of TECO macros. When I saw what they had done with TECO, I decided to take DEC's version, which didn't have any video terminal support, and at least put in support for video terminals to move the cursor.
That became a widely used version. DEC's version was TECO-24. I developed TECO-124. Pretty soon I started getting requests from all over the country. They would send me nine-track mag tapes. I was exporting TECO all over. One guy from the NSA told me I couldn't send it to him at work, so I sent it to his home.
Noah Smith: I want to note for anybody reading this interview in future years that one of the things we would like to do is restore Clive's version of TECO. It was used for Decwar. So, it would be nice to have a working version again.
Clive Dawson: Yeah. Yep.
Noah Smith: And I also want to note that everything we're hearing here is kind of the legends, the real legends from the old days.
Clive Dawson: Yeah.
Noah Smith: How many people do you think, like 25 or 50 people from Austin, were able to get on the Internet.
Clive Dawson: I would say it was 25 or less.
Noah Smith: All right.
Clive Dawson: It grew slowly, and then once the DEC-20 got hooked up the floodgates opened because anyone with an account could telnet anywhere they wanted.
Noah Smith: Okay. That's exactly the next topic on our outline, the migration to the DEC-20 in Painter Hall.
Clive Dawson: Right. Yes, the DEC-20 was intended to be a dedicated research machine. A consortium of departments contributed to the purchase and then contracted with the Comp Center to run it, which was my role.
I finally got my degree. I graduated in November 1980 and immediately got a raise and became the manager of the DEC-20. I had to leave the DEC-10, but I still maintained close ties with those people. My new job was to get the DEC-20 up and running.
Originally five groups contributed to the purchase of the 20: the CS department, ATP (Woody Bledsoe's group), CMP (Don Good), LRC (Linguistics), and I believe ICS (Jim Browne's group). They all got accounts and allocated disk space. It's funny to think about how much disk space that system had. It was maybe a little over a gigabyte total. The disk units were the size of washing machines. All those groups got established, started collaborating on the Internet, and it was a great system. One of the people who worked at CMP, Mike Smith, contributed to it.
The cool thing was that the official DEC colors were white and burnt orange. He brought in this enormous set of Longhorns and we put them on top of the cabinet in the machine room. They stayed for the length of the life of the DEC-20 at Painter Hall.
Noah Smith: We would love to have photos of the DEC-20.
Clive Dawson: I somewhere have a picture of those Longhorns and I salvaged the nameplate. One interesting story about the DEC-20 in Painter Hall. The comp center had an RJE terminal there which was just a line printer and card reader. People would come over to submit decks and get printouts from the CDC. They divided up that space and made a room for the DEC-20 with a raised floor for cabling. One winter, maybe 1981 or 82, it was very cold. The physical plant decided they didn't need the water chiller running because it was cold enough outside, so they turned it off. The DEC-20 did not agree with that at all. The room started overheating. I made frantic calls to the physical plant but they said it would take too long to get it back on. Meanwhile the temperature kept rising and I knew the 20 was going to shut down. So I decided to open the windows and let the cold air in. I sent a message to all users saying, "Save your work often, I've opened the windows to buy us time." About an hour later, it started snowing. I somewhere have the scrap of paper from the console saying, "DEC-20 will shut down in 20 minutes. I've had to close the windows to keep the snow out."
Noah Smith: I hope you can find this piece of paper as well.
Clive Dawson: Anyway, yeah, it's there. I got it somewhere.
Noah Smith: Oh, we would love to have photos of these. That's a great story. All right, where are we at here? We're in Painter. I had a quick question for you about the broader UT comp center. So, I think the CDC Cyber was down there in the machine room kind of mid 70s, late 70s.
Clive Dawson: Yeah, the CDC 6600 gave way to the Cyber, probably in the late 70s.
Noah Smith: Okay. I'm just curious, were you aware of the Cyber being used by the aerospace department? Was that in your awareness at all?
Clive Dawson: I knew about the aerospace engineering department and that they had accounts, but they didn't have any accounts on the DEC machines as far as I knew. I didn't have any direct uh involvement with those folks.
Noah Smith: Okay, yeah. I think my professor and his colleagues were on the Cyber and not on the DEC machines. So, this kind of confirms that. That's good to know.
Clive Dawson: There were a couple of interesting departments that I did know of. Robert Schneider in Petroleum Engineering was an avid DEC-10 user. He was the one who made up the T-shirts that said "Dr. Fonken never played DECWAR” which were worn at the going away party. Dr. Fonken was the vice president for research and it was his decision to shut down the DEC-10. We thought that he was an evil man and, you know, how could he possibly do this, but you know I guess one of his good traits was common sense and financial sense and he could see the numbers and we couldn't.
Noah Smith: Right.
Clive Dawson: So it was a no-brainer as far as he was concerned. The other department was zoology. Dick Richardson was also a character and an avid DEC-10 user.
Noah Smith: I knew Dick Richardson!
Clive Dawson: He passed away sadly. He was a character. I don't know if he ever played Decwar, but he was definitely an avid DEC-10 user.
Noah Smith: I can imagine Dick playing Decwar. Dick Richardson was a character. Oh, this is beautiful.
Clive Dawson: Yeah. He was over in Patterson.
Noah Smith: I had a job in Patterson for a few years. That's when I worked with Dick Richardson a lot, on computers of course. I was doing computer support in Patterson.
Clive Dawson: Yeah.
Noah Smith: So, beautiful. Two other names. So these are from the the older generation. I'm just curious if you had any interaction or knowledge of Al Matson in chemistry or David Young.
Clive Dawson: David Young actually taught one of my numerical analysis courses. I knew him because he was a member of the Faculty Computer Committee. I would occasionally have to make presentations about the status of the DEC-20. But he wasn't an adviser of mine.
The chemistry person who represented Chemistry on the committee was Dr. Lagowski. The irony is that after I retired, I helped senior citizens with computer issues. I'll go and set up their Macs for them or, you know, fix their email or their cable TV or whatever. And I generally will go to retirement homes to do this. One day I got a call from a woman living at the Querencia, and she was the widow of Dr. Lagowski. She and I had a lot of great conversations about him.
Noah Smith: Very nice. Okay. Wonderful. We're moving on in our outline. Next up is to talk about research programs like CMP and LRC. Let me ask you this. Are there any other major research programs that we should have the names of?
Clive Dawson: The authority on CMP is Rich Cohen. LRC was directed by Dr. Winfred Lehmann, who passed away. His successor was Jonathan Slocum, who got his PhD around the same time I did. He was director of the LRC for many years. He lives on the Mueller site now, and I saw him about three months ago.
He's doing pretty well. Slocum would be a good person to contact for LRC details. All the users on the DEC-20 were identified by initials of the research project, like CMP.COHEN or LRC.SLOCUM. I was CC.DAWSON for Comp Center and Woody Bledsoe was ATP.BLEDSOE. Those were the three biggies.
Noah Smith: Three three big research programs.
Clive Dawson: You know the other one that I'm thinking of, I forgot about Cognitive Science. CGS. Lauri Karttunen and his wife Frances had a presence on the DEC-20. It was kind of a cross between linguistics and AI.
Noah Smith: It's good to know the name.
Clive Dawson: Yeah.
Clive Dawson: They got some funding and contributed to the DEC-20. We set up accounts for them and they were good users.
Noah Smith: So the next questions are kind of about let's say official policy on using the resources on the DEC-10 and the DEC-20. Were there written policies? Were they enforced?
Clive Dawson: As far as Comp Center policies, the DEC-10 was just another part of the stable of computers. The main one was the CDC 6600 and then the Cyber. The DEC-10 became another one. There was a group in engineering, the Advanced Graphics Lab run by Marg Knox, that had all sorts of interesting graphics and plotting hardware. All of these had pretty much the same policies. You would apply for an account and computer time was charged for. If you were taking a class or were a RA, your professor would have an allocation of time.
The DEC-20 was similar, except you had to be one of the projects that had sponsored the 20. The DEC-10 was essentially wide open to anybody on campus as long as they had an account to pay for it.
Noah Smith: if you were using it for recreational purposes, you were still being charged for the time.
Clive Dawson: Recreational use was frowned upon unofficially, but it was dirt cheap. You could play Decwar for six hours and the bill would be less than 10 dollars. I was involved in the accounting procedures. When the DEC-10 first got installed, there were no ways to communicate between it and the CDC system. The first type of communications were in order to get DEC-10 accounting info into the centralized software on the CDC for billing. I was in charge of collecting that data. The DEC-10 kept track of CPU and connect time. I had to compile that data and write it onto a magnetic tape to send to the CDC. That was a challenge because the DEC-10 had nine-track tape drives and the CDC had seven-track until they finally upgraded. It was a jury-rigged system.
Another legend, people would steal accounts and passwords mostly for playing games. I became aware of someone making a career of shoulder surfing at the undergraduate library to see people type their passwords. He was stealing time and it became a bother, so I decided to catch him. It wasn't easy because he was dialing in at night to the MICOM and then going to the DEC-10 to play Decwar. Okay, so back then the phone company was not forthcoming when it came to “well, we had a call at like, you know, 10:30 at night. Uh, can you tell us where that call was made from?” “Nope, we can't.” It took me literally months of laying the groundwork and establishing relationships with Southwestern Bell, which back then was the company that was Austin's telephones, right? And they had their offices at Guadalupe and 18th Street. They had a big office building and central office hardware there on Guadalupe and 18th.
I met with Mr. Vance and eventually he agreed to try and get me some data, but it wasn't good enough to catch him in the act because I would get reports of calls from three weeks earlier. I did get a list of phone numbers, and the weird part was they were all from branches of Franklin Savings at all hours of the night. It was a financial institution known as Franklin Savings who was one of the big savings and loans places in Austin at the time. These numbers were coming from all these different branches of Franklin savings at all hours of the day and night of the night you know 10:00, midnight, 2:00 the morning, 3:00 in the morning.
Noah Smith: Security guard.
Clive Dawson: He was a security guard. He would be assigned to different branches and spent his time dialing in to play Decwar. I never confronted him, but I got a name and linked him to his legitimate account. I wrote a piece of software tailor-made to catch him. It would look at the account number coming in and if it was him, it would branch to a special program.
A note would appear saying, "We have a new version of Decwar to test, but it requires a hardwired connection. Go to a terminal facility." I waited until he logged in from the undergraduate library and ran over there. But they had a hundred terminals and I didn't know which one. By the time I figured it out, nobody was there. I lost enthusiasm because I didn't know what I'd do if I actually confronted him. I was obsessed over it, and years later I read 'The Cuckoo's Egg' by Cliff Stoll.
Noah Smith: I was thinking of that.
Clive Dawson: Okay. And I read that book and I said, "This is me.”
Noah Smith: Yes.
Clive Dawson: This is exactly what I did to catch my hacker, you know. Cliff Stoll had much bigger stakes with international spies from Germany, and I was dealing with a Decwar player on campus, but all the stuff he did with the phone company was just like my story.
Noah Smith: This is beautiful. Thank you very much for sharing that.
Clive Dawson: Gaming was minor compared to the serious work like theorem proving and mathematical simulations which used hours of CPU time. One user on Decwar was only using a slice of the comp center account, which by definition didn't pay for anything. The only thing charged to users was the IO to their terminal. Compared to serious numerical computing, it didn't use much. It was just part of, we considered it, part of the infrastructure. So the only thing that got charged to users was the IO to their terminal, right? It would query the main database showing the status of the universe and send the output to the terminals and that was minor.
I was doing serious computing with my dissertation so I knew what that was like. My stuff would run for hours trying to plan a complex block world device or something. I mean this is stuff that can now be done in like a couple of seconds, if that, and back in the 70s it would take hours of Lisp programs to, you know, grind out those solutions. I shake my head when I realize what baby steps we were taking back then compared to what has happened.
Noah Smith: I'd say we're in the last third of the outline. We're already talking about Decwar. So, what time do we have? We've been, oh, we're already getting close to two hours.
Clive Dawson: If you send me follow-up stuff, I can do some research in my files. I've got boxes of comp center papers in my garage. I'm sure there are tidbits there like the correspondence between me and the Defense Department to get the Collins link done.
Noah Smith: So this is a complete interruption now that this topic has come up. You know we're working with Briscoe to archive this type of material. So this is a topic for the future.
Clive Dawson: Yes, they're just going to get eaten by silverfish if they stay in my garage. I also have a complete set of the TOPS-20 notebooks, which was the bible of DEC-20 software. It was about 25 notebooks of documentation. DEC would send out quarterly updates that you'd have to add to the binders. I have a full set from the 80s and they take up a lot of space. If anybody wants them, they're there.
Noah Smith: Well, we'll leave a note right here for our friend Shannon Costello, the archivist at Briscoe, because she is looking for real material to bring into Briscoe.
Clive Dawson: All right. Okay.
Noah Smith: So, we'll leave a note right here. Shannon we will be talking to you.
Clive Dawson: Yeah.
Noah Smith: All right, let's finish up. Let's finish up with just a quick look at Decwar. So my first question was what was the institutional attitude? I think we've covered that this was accepted.
Clive Dawson: You have to be aware of the environment that existed among the DEC-10 staff. We had serious work to do for hundreds of users, but we were addicted to this stuff and were there eighteen hours a day. Nobody told me to develop TECO-124, it wasn't a necessary piece of software until it got popular. We would work there and play there.
We all got along really well. One day this kid Bob Hysick shows up. He's not in the roster of comp center employees, and my only explanation is that he might have been a work-study student. He started as a computer operator handing out printer output and worked his way up through user services. He was really good and started programming MACRO-10. We had a version of Star Trek developed for the CDC that people would go over to play. Dave Matuszek and Dan Reynolds were big names there. Bob Hysick had the idea to make it a multi-user thing. And, you know, I remember talking to him about it. I told him I didn't have time, but I explained the two-segment memory management on the DEC-10. High segment was pure code, low segment was data. I said you could use the high segment for a shared database, but you'd need a locking system so people wouldn't step on each other. This is elementary now, but back then using it for a multiplayer game was a new idea. I had never heard of it before.
Bob went out and figured it all out. I probably helped him debug things. He got it working, people noticed, and it grew like gangbusters. It started proliferating to other sites like my TECO-124. Bob would write a tape with all the Decwar stuff, include my TECO because you needed it, and off it went.
Noah Smith: So it's got the same spirit as the MIT AI lab. Did you consciously feel that?
Clive Dawson: It was sort of like the MIT AI lab. I had my work half busy tracking down hackers, and my role-model half thinking wouldn't it be nice if computers could be like MIT everywhere? Two other people who helped Bob the most were Tommy Loomis and Rick Watson. Tommy was a better programmer than I was. I could hold my own, but my strong point was taking other people's code and improving it, TECO for example, whereas Tommy was really good at starting projects from scratch and getting them organized properly and everything. I bet he helped Bob a lot. Rick Watson has unfortunately passed away. You should talk to Bob to find out where he got the most help. And I bet you those two guys, you need to talk to. Bob will know where he got the most help from, so I would encourage you to talk to him about that.
Noah Smith: I hope that we talk to Bob next, that we get an oral history interview with Bob next.
Clive Dawson: Yeah.
Noah Smith: Tommy Loomis, I think is a great idea. Rich Cohen as well, right? These are all interviews that need to happen. All right. Let's ask about the tape and sending out the tapes. So, we have records that there was a $50 fee, the mailing fee for the tape.
Clive Dawson: $50.
Noah Smith: We know that some of them ended up in Australia
Clive Dawson: I never heard about that. Actually, no. I take that back. I think my only requirement was that they had to send me a tape. I would write it and send it back, but I didn't charge for postage or making the tape.
Noah Smith: So, this was one of my questions, who did that? And it sounds like it was you.
Clive Dawson: I did it for TECO, and I'm sure Bob did it for Decwar. The height of the Decwar distribution was 1980-82. By then I had already left the HRC for Painter Hall, so I wasn't as aware of how big it got or what money was rolling in. I don't know if Bob had an arrangement with the comp center. I guess he must have.
Noah Smith: Well, this is a great place to wind it up actually for now. The final questions are about what we can call the aftermath. Were you aware at all of how CompuServe took Decwar?
Clive Dawson: No, I was not. I remember from Usenet boards like alt.sys.pdp10 there was a guy named Harris Newman. He would get on every so often and was like “I'm really trying to track down Decwar” and you know, he was obsessed with it. I mean, he was really trying and I remember I replied to him and said he had to find Bob Hysick. All I remember was that Bob was up in the Midwest somewhere. Usenet, for example alt.sys.pd10, was one of the main places and that was mostly MIT centric and Stanford.
Noah Smith: I was on there too.
Clive Dawson: Nobody had tracked down where he ended up. Harris was looking for the sources, but I never pursued it myself. I've had experience with SNOBOL, which was developed at Bell Labs. One of the researchers was an adjunct professor at Stevens. So, SNOBOL was developed at Bell Labs and one of the researchers at Bell Labs was an adjunct professor at Stevens.
Noah Smith: New Jersey.
Clive Dawson: Yes, because they were all in New Jersey and he would come over to the campus once or twice a week to teach a course on compilers or His name was James Gimpel. He decided to write a native version of SNOBOL for the 10. Originally SNOBOL was a macro-based language, but that was inefficient. Gimpel decided he had to do native code. And the way he did it was really amazing. I would have loved to have been a part of this, but unfortunately I had already left and come to UT.
Around 1972, he taught a course where the students built SNOBOL for the 10 as their term project. He came up with a really nice system called SITBOL. When I arrived at UT and was using the DEC-10, what do I see but SITBOL? I knew Gimpel had an agreement that Stevens would retain the license. You couldn't distribute it without permission. I go, wait, what is this doing here? Because I knew that Gimpel had reached an agreement with Stevens that he could do that and develop it but that Stevens was going to retain the license for that software.
I called my buddies at Stevens and asked if they knew DEC was distributing SITBOL. It was a major scandal. DEC had to make concessions to Stevens. That's how SITBOL got on the DEC-10s. I'm guessing CompuServe and Decwar was a similar story, an attractive product that would attract customers. I mean, back then all those distribution tapes were out there, you know.
Noah Smith: Information wants to be free.
Clive Dawson: Exactly. So that's how SITBOL got on to the DEC-10s. Anybody who wanted it could use it. I would have been upset if CompuServe was charging for it.
Noah Smith: And my understanding is, and I've looked at what we have today, the letter that came from UT, there's nothing about commercial usage. So, things were different back then. Like we talked about with the AI lab and Richard Stallman.
Clive Dawson: Oh, yes,
Noah Smith: Things were different.
Clive Dawson: They were. Yes, they were.
Noah Smith: I think we've covered everything and and more than I had dreamed
Clive Dawson: I certainly enjoyed it and it brought back a lot of good memories. I'm prepared to revise and extend my remarks as they say in Congress.
Noah Smith: To wrap up this session. Is there anything that you can think of that was important that we haven't hit on, that you'd like to. And you know we can come back to it later.
Clive Dawson: I will make it a project to go into the garage and dig out those boxes. I'm sure that will bring back stories and memories that should be on record. We can talk to the archivist about that too.
Noah Smith: Please do take some photos, and we'll share those photos with Shannon. Give her an idea of just how dangerous this could be for her future, haha.
Clive Dawson: Yeah. The notebooks take up at least five big boxes and I have an equal number of boxes of my files from that era.
Noah Smith: I'm guessing that anything that was official UT, like your communications with Collins, I guess that that was probably MCC, but I would say UT adjacent.
Clive Dawson: A good source of information is a complete set of computation center newsletters. They were published monthly and had topical news about new computers and operating system versions. It was distributed hard copy to all users. Everything from the announcement of the DEC-10 to its demise will be in there. I wouldn't be surprised if Decwar found a mention in the newsletter at some point. I have a bunch of newsletters, but not the full set. We can discuss by email and explore this with Shannon at Briscoe.
Noah Smith: We'll discuss by email and get an idea of what's there and and explore this with Shannon at Brisco. You know, we're picturing doing an exhibit at Brisco.
Clive Dawson: Okay.
Noah Smith: And what Brisco likes to show is real historical material.
Clive Dawson: I attended the celebration of life for Bill Bard, who was an assistant director. We should mention Charlie Warlick, the director of the Comp Center. His philosophy trickled down to all of us and allowed something like Decwar to be possible. There are organizations that are run in such a way that things like Decwar would have never been possible.
When I first walked into Charlie Warlick's office, he had a big blackboard going the length of the wall. And on the very top edge of this blackboard, he had a saying. And I copied it down. I was so impressed. And every place that I ever worked after that, including MCC, including AMD and Samsung, every place that I went, where I was normally a manager of some IT group or whatever, I would write that saying down on my blackboard or my whiteboard as it were.
The saying was, "There they go. I had better run and catch them, for I am their leader." It is profound and inspiring and contributed to the environment that allowed Decwar to flourish. He was perfectly happy in standing back and watching his troops run the show.
Noah Smith: I'm glad you said this. I haven't learned enough about him, but everything I've seen is admirable.
Clive Dawson: I met John Otken who pointed me to the people who made replicas of the PDP-10 console with a Raspberry Pi. Apparently it can run TOPS-10. I'm seriously believing that I'm going to get one of these.
Noah Smith: It can not only run TOPS-10, it can run Decwar. I've got one sitting right here.
Clive Dawson: Oh, you bought one of these things.
Noah Smith: This is the end of it. Yeah, I built it myself.
Clive Dawson: Oh, you got the kit.
Noah Smith: Yeah. The first thing I did when I learned about it is say “I'm going to run Decwar on that.”
Clive Dawson: Excellent. Okay. Well, that does it. I'm getting one.
Noah Smith: All right, Clive.
Clive Dawson: All right. Well, this was a pleasure and I hope that it contributes to the stories and legends of the old days.
Noah Smith: Oh, this is fantastic. This is a treasure trove and I can't thank you enough. And looking forward to talking the next time.
Clive Dawson: We'll be in touch. All right. Well, take care.
Noah Smith: You too.
 |
| Clive and Richard Denney. |
 |
| The evening of The Soul Of An Old Machine. |
 |
| Another from the same evening. |
