The HLF Portraits: Fernando J. Corbato
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 66 | |
| Author | ||
| Contributors | ||
| License | No Open Access License: German copyright law applies. This film may be used for your own use but it may not be distributed via the internet or passed on to external parties. | |
| Identifiers | 10.5446/40204 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
The HLF Portraits17 / 66
2
9
14
16
17
18
20
23
25
26
27
29
30
34
35
36
38
40
43
46
50
53
57
58
00:00
Turing testBitProcess (computing)Physical lawFamilyInheritance (object-oriented programming)QuicksortEvent horizonVideo gameMathematicsStaff (military)MereologyRational number40 (number)Extension (kinesiology)Musical ensemble3 (number)Term (mathematics)Degree (graph theory)GradientAreaBlock (periodic table)Formal grammarPosition operatorStudent's t-testLength of stayPoint (geometry)Natural numberCatastrophismMultiplication signRight angle1 (number)Pivot elementMeeting/Interview
09:34
Goodness of fitRight angleHTTP cookieFood energyTelecommunicationWave packetMultiplication signCurveComputer programmingSpring (hydrology)Musical ensemblePhysicalismUniformer RaumDirection (geometry)Order (biology)FrequencyNormal (geometry)Factory (trading post)Digital electronicsScheduling (computing)Optical disc driveSocial classMereologyMetropolitan area networkMathematicsInheritance (object-oriented programming)Sound effectFilm editingWordPoint (geometry)CASE <Informatik>Row (database)Service (economics)Universe (mathematics)Term (mathematics)RoutingOnline helpMeeting/Interview
19:08
Right angleQuicksortFamilyVideo gameLevel (video gaming)Group actionMultiplication signPattern languageLaptopCategory of beingElectric generatorCodeGastropod shellMoment (mathematics)MereologySet (mathematics)Decision theoryNeuroinformatikSoftware developerStorage area networkGraph coloringPhysical systemWave packetModule (mathematics)HeuristicDiagramDigital electronicsSocial classIntegrated development environmentDisk read-and-write headPhysicalismMainframe computerComputer programmingAxiom of choiceBell and HowellExpected valueDirection (geometry)Process (computing)Sheaf (mathematics)Term (mathematics)Physical lawObject (grammar)Revision controlBitWater vaporPower (physics)Nuclear spaceOnline helpMeeting/Interview
28:42
NeuroinformatikPunched cardOffice suiteDirection (geometry)QuicksortMusical ensembleShift operatorMoving averageMathematicsRight anglePhysical systemVideo gamePhysicalismLevel (video gaming)HypothesisGroup actionDivision (mathematics)SubsetUniverse (mathematics)Degree (graph theory)Punched tapeVirtual machineNumbering schemeOrbitBitCategory of beingProcess (computing)Magnetic-core memoryElectronic visual displayCalculationInterior (topology)Multiplication signoutputCore dumpSemiconductor memoryTube (container)Shape (magazine)Meeting/Interview
38:16
Online helpPhysical systemMultiplication signFrustrationDistancePhysicalismBuildingAssociative propertyProjective planeUniform boundedness principleStaff (military)NeuroinformatikTask (computing)PeripheralQuicksortTranslation (relic)Formal languageModemSpecial unitary groupHydraulic motorRadical (chemistry)Disk read-and-write headRemote procedure callMereologyWordComputer programmingSemiconductor memoryEvoluteShooting methodArithmetic meanStapeldateiCuboidMachine codeMainframe computerGroup actionInteractive televisionOrbitDecision theoryWave packetDifferent (Kate Ryan album)Process (computing)Magnetic-core memoryDirection (geometry)Meeting/Interview
47:50
Service (economics)BuildingClient (computing)Group actionVirtual machineSquare numberRadical (chemistry)SpacetimeElectronic visual displayProjective planeResultantFrequencyObservational studyComputerInformationstheorieSmoothingVector potentialComputer programmingBitCrash (computing)Dean numberGraph coloringNeuroinformatikSeries (mathematics)Core dumpPhysical systemFrustrationPlanningMetropolitan area networkMultiplication sign2 (number)Right angleWave packetCASE <Informatik>Moment (mathematics)Kritischer Punkt <Mathematik>Logistic distributionOrder (biology)Web 2.0Bell and HowellSimilarity (geometry)Complex (psychology)Meeting/Interview
57:24
Group actionMereologyProgrammer (hardware)Multiplication signComputer programmingSelf-organizationProjective planeExpected valueNeuroinformatikSet (mathematics)Graphics tabletCodeQuicksortFlow separationPhysical systemDirection (geometry)Power (physics)Category of beingReplication (computing)Software developerLevel (video gaming)Object (grammar)Right angleBitRevision controlMeeting/Interview
01:05:06
Internet forumBit rateComputer animation
Transcript: English(auto-generated)
00:17
Professor, you were born in Oakland, California in the 20s, I think?
00:23
1926. 1926. Who are your parents at this point, and why are they there? My parents were both graduate students at Berkeley, University of California. I was born in Oakland, which is adjacent to Berkeley, which is where the particular hospital happened to be.
00:55
After my father got his doctorate, he took a position at UCLA.
01:03
Which is where you really grew up in Los Angeles? Which is where I really grew up. Before we go there, where are they from, your parents, and where did they meet? Well, I met at Berkeley. They met at Berkeley. My father was born in Spain, and my mother was born in San Francisco.
01:35
I have not thought about that for a while. I think that is correct.
01:42
She was born in 1906. No, no, sorry. She was born in— Well, again, the exact date, but she was born in that area, in the Bay Area.
02:02
My grandfather and his wife and my mother were forced to basically witness the San Francisco earthquake and fire, which I believe was in 1906.
02:30
So my mother was born in 1903, and it was pretty horrendous.
02:42
People had to basically live outdoors until things calmed down. Oh, it was catastrophic. Catastrophic. Were they both in the same graduate department? They were both in the Department of Spanish, yes. Both in the Department of Spanish?
03:01
Yeah. Was she of Spanish heritage? No, no. My mother was—she was getting—I think she got a master's degree. My father got a doctorate. And my father took his first job at UCLA, first and last job.
03:24
First and last job. A little bit like you, when you find an academic home, you stay. UCLA was just expanding into its new campus in the western part of Los Angeles, and he was part of that first crowd of people that staffed it up.
03:48
Do you happen to know whether he had intended to return to Spain, or was he pretty committed to an American life? I think he was basically committed to an American life.
04:04
Everything else I know is sort of folklore. This is usually true about one's parents. Right. Are you living in Westwood as a family? In West—yes, actually West—well, Westwood between Santa Monica Boulevard and Peak in Olympic Boulevard.
04:32
I know it very well. I'm from Los Angeles too, so I know exactly where you grew up. How many children will there be in the family at all?
04:40
I was born in 1926. My brother was born six years later. So there are two of you? There are two of us. What is the nature of your schooling at this point? Are the schools good? Are your parents already worrying about your education and thinking that they want certain things for you?
05:02
What is that intellectual childhood like? Well, the public schools in Los Angeles were pretty good. So that was sort of the norm, and that's where I went for both grammar school and later junior high.
05:32
And junior high. I went to the Selby Avenue Public School, which is a few blocks from where we lived.
05:45
And then I went to—for the seventh, eighth, and ninth grades, I went to Emerson Junior High.
06:03
Your parents are humanists, not scientists, which is not necessarily always typical of Turing Award winners in terms of their childhood. When are you beginning to develop an interest in science or technology? Was it in your childhood or was it much later?
06:26
I guess I noticed I was good at mathematics, and I wasn't particularly attracted to a humanist curriculum.
06:56
But to some extent, my career was shaped by the later events, which I guess the—
07:18
About the time of—as I was about to graduate from junior high and go to high school, I haven't worked out the exact timetable, but—
07:37
Yeah, in general.
07:41
But the pivotal event occurred, namely that we had Pearl Harbor. Very pivotal. And that shook things up a lot.
08:00
For example, my parents had, I think, either rented or leased a small beach cottage down near Malibu on the coast about, I would estimate, five or ten miles away.
08:21
And it was down—we were down there one day, and I was on the beach lolling in the sand. And I must have had a portable radio on, because we suddenly heard about Pearl Harbor.
08:41
And that was a shock to everyone. Yes. And the full import of it didn't really register right away. You were about sixteen at this time. I mean roughly, forty-one.
09:03
Forty-one, I would have been fifteen. Fifteen, okay. Fifteen. And so one of the immediate consequences was the U.S. suddenly imposed gas rationing.
09:26
And the reason wasn't because we were short of gasoline. It was because we had no way to get the rubber for the tires that were in very short supply.
09:42
And so the way of stopping people from driving too much was to put gas rationing in. Why? So that was one immediate consequence. And one of the other side effects of that was this beach cottage that my parents had leased or rented.
10:03
Couldn't be accessed. It was enough of a trip. We didn't want to take it anymore. Are you in high school at this time? I was about to, either about to or just entering university high school. University, the great high school, yes.
10:22
Yes, in the western part of L.A. there. And so everything, the war began to shape everything. In particular, the high school began to offer an A and Z period.
10:43
It had the normal one through five or something, but a beginning A was before and Z was after. The normal scheduled times. And the reason for that was so that young people could work in the factories and support, help the war effort.
11:10
I took it another way. I saw it as an opportunity to accelerate my education. So I took both A and Z.
11:21
Yes. And the normal one to five or something, whatever. So a normal period in high school would have been four years. Three. Three. Three. It was a three-three. Three-three. And you were racing to make it less time. Make it less time. And I basically was able to almost lop off a year.
11:47
So I was almost finished with high school in two years rather than three. In order to actually pull it off, I had to go to summer school the next summer. And I took a couple math courses, which were easy.
12:04
Right. And the courses, well, there's a general curriculum, of course, but are you expected to major in a particular direction? I don't recall I had a major at that time. Okay. But it allowed me to finish up at UCLA, at Uni High.
12:26
At Uni High. And basically, I was able to enter UCLA in the, I forget the year, but it would have been basically,
12:42
I was able to, and I guess I probably took a major in, I don't know if I had to pick a major, but I was nominally aiming at a physics major, I think.
13:02
And... You were about seventeen? That would, yes, I was. And the reason I know that is that by the time I was eighteen, I was scheduled to be drafted.
13:23
Yes, of course. And that didn't, I suddenly, somewhere along in the spring of that first year at UCLA, someone told me about the Eddy program.
13:42
Okay, tell me about that. And the Eddy program was a Navy program which offered twelve routes of schooling in order to train people to be electronic technicians, to service and maintain the sudden influx of sophisticated electronic gear,
14:09
such as radar, sonar, and the like, that was in the fleets. Which is like, for you, a kind of beginning... So that was a very attractive path.
14:22
And so what I did was I aborted my UCLA experience and enlisted in the Navy somewhere about May of 19... I'm not sure of the year. Yes, but anyway, it's going to be 42 or 43, something like that.
14:45
Yes. And you're now going to control your participation in the war rather than be randomly drafted. Right, and get a decent... first of all, get a training in electronics,
15:01
which seemed like an attractive thought, and basically have a year-long education in a different... basically as a technician. But it was far more desirable than just being a crud in some sense.
15:24
And it's also an opportunity, in a way, to learn way ahead of the curve about the technology of the time. I didn't think of it that way, but... Oh, young people never do, but in retrospect... It gave me a leg up because I had a tremendous exposure to, at that time,
15:52
sophisticated electronics. And you're finding that you're not so bad at this. Exactly. I was always... I was used to being kind of at the top of the class.
16:07
In any case. Even in high school, and not the very top, but among the top quarter or so.
16:24
Is this going to take you away as you think about your future, and I remember how young you are at this point, but to, if you will, feel a little more applied notion of science and technology, rather than the physics you and the math you had originally started with,
16:43
or is this not really shaping you so precisely? I guess I didn't know where I was going exactly. The Navy, of course, was total preoccupation for almost a year. Right.
17:00
It was a tremendous exposure. I mean, at first, the program started with one month, or maybe it was four weeks, at a so-called pre-radio school, which is a manly high school in the south part of Chicago.
17:27
And that was an exposure to kind of getting you used to being in the Navy. Yes. Or in the service. From there, my next stop was three months of what I guess they called,
17:50
remember what they called it, three months of further training at the Hotel Del Monte
18:02
in Monterey, California, which had been taken over by the Navy. I didn't appreciate how famous it was until afterwards. Are you surrounded by very clever young men who had all been tapped for this,
18:24
or are you finding yourself alone in your interest and abilities? No, I was with basically a class of people. And so, from Del Monte, I spent three months, I guess that was called pre-radio.
18:48
The title, it was kind of just arbitrary, and that exposed us to very rudimentary explanations of how circuits worked
19:04
and how electronic terms weren't in use yet very much, but how basically one could analyze a circuit in a heuristic way without any precision,
19:26
but just in sort of an understanding of what was going on. So it's a ground work for you and a lifetime really to learn this process. Right, and after pre-radio, I was assigned, my next stop was six months of further training
19:51
at what is called Treasure Island, which is an artificial island that was built for the 1939 fair,
20:02
San Francisco World Fair off of Yerba Buena Island halfway between San Francisco and Oakland. And it was on that island that the Navy had basically a barracks-like environment
20:25
where we both had our sleeping quarters and went somewhere on the island for classes, further classes.
20:42
And we had, in particular, we had books which described contemporary electronic equipment, and we would sort of walk through these various circuits, various diagrams of the equipment
21:04
with heuristic explanations of why things worked the way they did. The thing I remember most was that we couldn't keep our notebooks.
21:21
We were forced to turn them in every night because they were all classified. Because you have such a rich life, I want to proceed. Are you in the Navy throughout the war? I was. Well, I finished up my time at Treasure Island, and it was approximately a year after I was in the Navy.
21:53
And my first assignment was to be on the commissioning crew of a ship, a destroyer tender,
22:02
being built up at Tacoma, Washington. And I was part of the commissioning crew, and the ship wasn't finished yet, but it was being finished at that moment. And so my next stop was going to Tacoma, Washington, where I did things like help.
22:31
I think we helped load the ship with supplies and various, I don't have any firm recollections exactly of that,
22:45
but what I do remember was that it was, I think it was in August, August, I think my notes say. Again, I don't think we need the precise time, more what you're doing and thinking.
23:04
No, the key thing was that, unbeknownst to us, of course, Japan surrendered. And the reason, of course, is we dropped two nuclear bombs on them. And after the shell shock of that occurrence, the Japanese were persuaded they had no chance.
23:31
And so suddenly the war was over. Certainly the one in the Pacific, yes. Right. And our ship was just about to be commissioned and or had been commissioned.
23:45
And so we went on a, I think we went up and down the coast, the West Coast, and the ship just sort of shaking it down.
24:06
The ship being a destroyer tender was supposed, was a repair ship, and so it had a lot of interesting aspects about it. One of the things they belatedly found, discovered that they had gone up and down the West Coast
24:31
and discovered that the ship was not riding level in the water. And the reason was somebody had painted the bow markings incorrectly by ten feet.
24:46
Right. So the ship was basically tilted. And the reason I recall that was that the way they corrected that was they pulled into a,
25:01
in San Francisco Bay there was a particular island which was a munitions depot. And we loaded ammo for two days straight to ride the ship. I've got to get you out of the Navy and into your next stage of education.
25:20
Okay. So how do I get you to what I believe is Caltech? How are you returning both the civilian life and to education outside? Okay. Well, after the formal conclusion of the war, I spent about six to nine months in the Navy
25:52
as we kind of tapered down. Meanwhile, the ship had migrated through the Panama Canal over to the East Coast,
26:08
which I didn't get to experience because I was sent to a special school for mothballing and equipment somewhere in the East which I forget where it was.
26:23
Anyway, when I returned to my ship we were, I believe we were in Portland, Maine when I was finally released from the Navy. And I forget exactly how I did it.
26:42
Again, it's more how do you then think of the next stage of your life that I want to pursue? So the next stage of my life was to go back to my family home in West Los Angeles and contemplate where I would go to college.
27:04
And I decided I'd start over and go to Caltech. I had had a childhood friend who'd gone there and been in the V-12 program and I guess I knew it had a high reputation.
27:25
And now I had the GI Bill in my pocket basically, which paid either a lot or all of the tuition. And so I didn't feel guilty.
27:41
Was it the drain on your family? It wasn't the drain on my family. Exactly. Do you know what you're going to specialize in yet? I think I picked physics because I don't think we had to pick a major right away. But physics was my choice because everybody picked physics.
28:04
It was kind of the challenge of the day. So I majored in physics and had four very pleasant years at Caltech. Are you finding that you're pretty good at this?
28:22
I was always in the… They had an honors section, the so-called A, and had two of them, an A and a B section. And I was in the A section. Are you being encouraged by particular professors or just the benefit of a broad Caltech education?
28:43
Just the benefit of… There were a few people that stood out, some positively, some negatively, but no particular mentor. You graduated in physics. I graduated in physics. Somebody is telling you you're good enough to go to graduate school?
29:04
Or are you not yet sure what the next step is? I guess I don't remember deciding, but I obviously did expect to go to graduate school.
29:22
And I applied to several schools, got admitted by MIT with an offer of an assistantship. A paid assistantship, which part-time.
29:41
Right. And that seemed like a good bet. And so that's how I ended up picking MIT, which was… It becomes your intellectual home essentially for the rest of your life. Right. I didn't realize I was doing that. No, but again, well, it doesn't. But there you are.
30:02
You're at MIT. Have you been asked to choose a direction there, or can you again… Well, in graduate school, they expect some sense of what you want to do there. So what do you think you're doing? Are you there in physics? Are you there in some other… in mathematics?
30:21
I was in physics, but two critical… First of all, to get back to MIT, I decided I'd drive my car cross-country. Okay. And I did it all by myself, solo.
30:41
It was packing… it was a coup, a 1936 Ford coupe, and which… And this was way after 1936, so it was not a new car. Not a new car. It was pretty… it had been modified and had a lot of peculiar features in it.
31:04
But I drove it solo back to MIT. And the thing that sort of gives me pause now is how many risks I took.
31:25
I ended up… I don't recall staying in any motels or hotels. I just slept by the road at night, which was, in retrospect, a very dangerous thing to do. Right. But I drove it in about… it must have taken me two or three days.
31:46
At least more… actually, it must have been more. But I got to Caltech… to MIT, and…
32:00
So, what was the question? Well, it's essentially how your intellectual direction proceeds from that. Okay. This Professor Morse, who was not only the graduate admissions officer, but he also had… he was quite entrepreneurial.
32:21
And he convinced the ONR to fund about ten or twelve fellowships, or research assistantships, which he could dole out to key people. And he had offered me one of those assistantships.
32:43
And so this got me exposed to some of the early punch card equipment, which was kind of dull. But in particular, it got me exposed to working at the Whirlwind I computer.
33:05
Okay. There was never a two, but… The Whirlwind I was kind of a precursor to the SAGE system, the air defense system that MIT built through its…
33:21
So it spun off MITRE Corporation. And I never was associated with it, but I did work with… I got to work with Whirlwind using the unclassified third shift from midnight to four or something,
33:46
which led me to some pretty horrible hours. Did it dazzle you with what it could do? I mean, in retrospect, of course, it couldn't do very much, but there you are experiencing this. Whirlwind was a pretty potent computer for its time.
34:02
It could do a lot. And it was still a vacuum-tube computer, but it managed to do…
34:23
You could do significant calculations on it. Yes. And it had cathode ray tube displays and punch paper tape input and output. It did not use punch cards particularly.
34:45
But it was a very fast computer for its day. It had some home-designed memory tubes, which were a little problematic.
35:09
Unbeknownst to myself, a year or two later when I was still working with Whirlwind, over a weekend they changed the cathode ray tube memory scheme to the first core memory scheme.
35:30
Not only did the machine suddenly get faster, but it became incredibly reliable.
35:42
So that was a real eye-opener. So I saw some of the very early technology there being developed and prototyped on that machine. In this process, are you beginning to shape ideas of the direction of your research,
36:08
maybe the dissertation? Are you becoming more fascinated with certain issues and problems? I got into a group with Professor John Slater, who was interested in mathematically predicting
36:30
the properties of chemical, of molecular molecules and the like. I ended up drifting around a bit, but eventually latched onto a topic of computing molecular orbitals
36:50
using the Whirlwind computer and sort of ground out a thesis, but I wasn't very proud of it. I basically got my degree.
37:07
But one consequence of this, though, was that of Phil Morse, who had persuaded IBM
37:21
to donate the use of a 704 computer to MIT in exchange for which MIT would administer the use of that computer with some 40-odd New England colleges and universities who would have time on it.
37:53
Ostensibly, it was one shift for MIT, one shift for the New England colleges.
38:05
But it was all smeared together, so that was an artificial division in the first place. One of my memories of that was, I forget the exact year, but the year that Sputnik was launched.
38:37
That's 1957, I think. 1957-ish, and I was at the center as a research associate, and the U.S. was basically shocked
38:58
that the Russians had been able to pull that off.
39:01
And the Smithsonian Astrophysics Observatory up in Cambridge, basically near Harvard, came down and were using time on the computer to calculate the orbits and things like that,
39:24
and everyone was still getting over the shock of the Russians having done something we didn't even think about. The Department of Defense got beefed up a lot, and one of the ways it got beefed up
39:41
was in trying to create interactive computing. And the group that ended up doing that was called ARPA,
40:05
and ARPA recruited a person by JCR Licklider to come down to be the head of ARPA. And Licklider saw it as an opportunity to dole out funds to nurture timesharing all around the country.
40:36
Can you define timesharing? Because, of course, it becomes central to your work.
40:45
Timesharing was a notion of, I think the first person I heard describe it in any detail was John McCarthy, who was a professor in the EE department at MIT.
41:05
And John's notion was he invented the LISP language also, which is another major accomplishment on his part.
41:21
But one of the things he was advocating was a large computer – MIT, in his words, should get a large computer with a million words of memory.
41:41
Of course, John's private reasons for wanting that was that he could run LISP programs in the million words. Fair enough. So anyway, McCarthy was advocating this, and one of the other professors associated with the Computation Center
42:07
was Professor Herb Teager, and he was going to develop a timesharing system. And he began, he sketched out, or had in his head, a very ambitious timesharing system
42:28
that was going to do everything under the sun, including language translation and all sorts of other sort of exotic and peripheral tasks as part of his research project.
42:46
And so his ability to create a timesharing system kept retreating into the distance. And so I got kind of frustrated with that and decided to do a very simple-minded timesharing system
43:07
using the IBM 704, which, with the help of a couple of other staff people,
43:26
I was able to basically code up myself in machine language a very small timesharing system
43:40
which would coexist with the other users' programs in the memory of the then commonly used so-called Fortran monitor system, which is the batch system.
44:00
And these programs could coexist in the same computer, and as long as programs weren't too big, they could all run together. So that means that people can, I mean, the essential problem is that it's waiting to have access to this because it can't process many different programs going on, and you're solving this problem, you and others.
44:28
The timesharing means that things can happen simultaneously, access to the computer. Am I right? Not quite. Actually, it was a very tricky evolution.
44:45
I think, I don't recall, Morse perceived to persuade somebody back in the government that by getting another bank of 32,000 words of memory, we could have both an A and a B.
45:18
We could have a computer which had two banks of memory, one of which would be for the regular users
45:25
and the other would be for the timesharing users. I see. And so that made it easier, and IBM helped too by, I guess, there was a large,
45:53
IBM built a large box for communicating with remote terminals, dial-up terminals, called a 7750,
46:10
which somehow we managed to, I forget who we coaxed into getting us one of those, but we did have that. So suddenly we had an extra bank of core memory and a 7750 box.
46:26
And now we had the ability to dial-up from remote terminals using the modems of the day, which are big, somewhat bulky modems.
46:46
And you could have computers, terminals all around the campus or even at home if you could spare the expense. It must have felt like a radical advance. So suddenly we had the ability to remote access a computer which was miles away from a terminal.
47:13
Well, this was all going well. Meanwhile, down in Washington, Licklider had seen a modest timesharing system on a PDP-1 or something.
47:31
So he had experienced firsthand the joy of being able to interact with a computer directly.
47:43
So he was down in Washington in ARPA, and he proceeded to approach MIT to see if there was any interest in starting a project to develop timesharing. And that's where you come in. Well, I became a participant.
48:02
The key person to pick up the ball was Bob Funnell, who is a professor of electrical engineering. And had already made important contributions to information theory and so forth.
48:25
He'd been taught at the behest of the dean of engineering. He had been put in a two-week crash course to learn a little bit of programming. But in any case, he saw the opportunity to respond to Licklider's invite to form a timesharing.
48:49
A timesharing project is particularly attractive. So he proceeded to write the proposal himself.
49:02
Basically, he proposed starting Project MAC. And Project MAC became the center of timesharing at MIT. A major center as a result of that, because of Licklider.
49:24
So Project MAC got the money from Licklider, from ARPA. And it was off to what was then about a $3 million a year budget. Which was real money at that time.
49:41
At that time, it would be more like six today. And of course, they needed space to do all this. So Fano was able to lease a couple floors of a new complex over at Technology Square.
50:07
Which is a building that had just been built as an R&D center to lease out to clients by MIT.
50:24
MIT had actually, I guess, encouraged the building of it. But it was only a, at least a, it wasn't quite a landlord, but it was a... And luckily, a service bureau who had opted for two floors of the building,
50:50
folded before it really got going. And suddenly there were two floors available. And so this Project MAC moved in on that, and had space to grow and grow into this building.
51:05
I didn't want to be melodramatic, but I'm very interested in your kind of eureka moment. I mean, you spoke a little bit about the frustration of certain things not yet happening the way you want. What are you out to solve, and how do you do it?
51:22
Well, initially, Fano elected to put at the core of Project MAC a copy of the computer system at MIT,
51:42
at the Computation Center. And we used to refer to them by their panel colors. One machine was red and the other was blue. And so that went along well.
52:07
And shortly after, and after we were, the project was running smoothly. Well, a couple of things happened. Fano had a, we had an invited about, I would estimate,
52:26
between one to two hundred prominent researchers around the country to come visit MIT and see timesharing in action. So that was a so-called summer study period.
52:41
And people came and so on, were encouraged. Were you involved in the demonstrating of the... Well, we had the terminals around and people could use them. I see. And at that time, they were pretty clunky typewriters,
53:07
selector typewriters at best, and sometimes teletypes. And the cathode ray tube displays had not yet shown up,
53:26
except in, I think there was one or two very special terminals that were cathode ray tubes. But by and large, it was typewriter in and typewriter out.
53:45
So where were we? So after the summer study, and the dust had settled a bit, Fano encouraged us to start thinking about a new machine specially designed for timesharing.
54:06
Wow. So there was a group of maybe somewhere between four and six of us that went around and visited the various vendors, potential vendors,
54:27
to see who would be willing, interested in building a significantly large timesharing system. And I remember we visited CDC and among other places.
54:45
I think we were down, I don't remember if we went to Bendix, but we traveled around probably, in other words, four to six different potential vendors. And one of them was GE Computer Department out in Phoenix.
55:10
And they in particular had a, got very interested in our ideas
55:27
and proposed to build a machine especially for us. They had kind of a wild man of a designer, John Kaller, who had bit off,
55:41
he didn't really understand programming very well, but he bit off more than he could chew. And in any case, out of that came the planning for the Moltik system. Ah, yes.
56:01
Critical point, yes. And the Moltik system was a tremendously ambitious, overwhelmingly ambitious system,
56:20
which gradually got scoped out. What's his name, Bill Laps?
56:46
In any case. Ed David. Ed David down at Bill Laps. So we wrote a series of papers and Ed David down at Bill Laps got extremely interested and decided to join us in a joint venture basically,
57:07
where we would all order the same equipment and try to work on the system together. It was all kind of idealistic and it didn't, but because of the logistics,
57:23
Bill Laps being a couple hundred miles away, basically the center of the action became around MIT. And the GE people, they basically didn't have, they really didn't have research-level programmers available.
57:53
So they were part of the effort and they didn't contribute a lot.
58:07
So Fano decided that we needed to write a set of papers describing what we planned to do. Your wild ambitions. My wild ambitions. So we wrote these papers, which in hindsight were incredibly,
58:26
and we presented them at one of the computer conferences, and it was an incredibly ambitious agenda. Because most people were guessing you didn't have enough knowledge to achieve that yet?
58:43
We didn't really foresee how intricate the problems would be. Things that were kind of designed, desirata, were taken as expectations of we could pull it off.
59:10
So we started on this very, very ambitious programming project. Undaunted. I mean, you felt you could do it.
59:25
We thought we had a chance, and we started with all the machinery. We self-taught ourselves about how to organize programs and things. But it took a long time to get really our feet on the ground.
59:46
Yes. Much longer than we had anticipated. This is not unknown in research. And we went through several...
01:00:02
of the system, trying to put it together. Eventually, Ed David down at Bell Labs had sort of sold the idea of his group working on Multics as kind of the next big thing,
01:00:30
which was expected to come in and supply the labs with computing power. So he'd been a bit naive to over-promise.
01:00:42
Over-promise, exactly. And so, basically, they had to pull out. The rest of the lab at Murray Hill basically
01:01:00
pulled out of it, and Ed David basically had to drop out of the arrangements. Right. But your team is… Our team is still going. Still going. Right. It's turning out to be a very hard slug. This is a stretch. We've gradually got the thing up to speed.
01:01:30
The initial version or two were so cumbersome that they didn't work very well. People had to do drastic slashing of expectations and streamlining modules.
01:01:47
We basically had to rewrite the system a couple of times to get it to come up to any reasonable performance levels. But you did get there? Eventually, we got there.
01:02:04
Just because we don't have much more time, I'd love to hear, what are the consequences of having gotten there for the development of computer technology? I think we ended up basically putting out a set of desirable properties for a large computer, one that might not be entirely attainable right away.
01:02:43
We established a pattern, at least within our own group. You write up what you plan to do before you do it. Up until then, programming had often been sort of the art of the possible. Someone sitting down with a coding pad and programming what came easily.
01:03:04
Here we were setting goals and objectives and saying, we're going to program to meet that. That was kind of a new phenomenon. I think the biggest impact that Multics probably had was by people reading the papers and seeing
01:03:26
out of all these ambitious goals, being inspired to maybe pick off a part of them or something. Although people didn't try to replicate Multics, in fact it was written off as a failure by many people.
01:03:50
Very interesting. Because it was slow, it was late and so forth. We basically set the directions of designing before you program.
01:04:09
That eventually became the habit of researching computers. Your approach is what people picked up. I think so, I think so. It was not a, and I guess the fact that we got it up and working
01:04:34
eventually with a testimonial to the doggedness of the other groups, but it was a major challenge to get it to that stage.
01:04:54
We'll end with the concept of doggedness, which is probably as important to research as any other question. Thank you so much.
Recommendations
Series of 66 media