Apollo Guidance Computer History Project

First conference

July 27, 2001

David Mindell's introduction

DAVID MINDELL: Welcome to all of you, and thank you very much for coming. My name is David Mindell. I'm an Associate Professor here at MIT in the History of Engineering and Manufacturing. What we're doing today is the first of what we hope is a series of sort of group oral histories of the Apollo Guidance Computer Project, which we think is very interesting and important, and hope you do as well; we have a suspicion you do.

What we want to do today just briefly to get started, is give you a little background about what we're doing here, how we got here, and how it's all set up. But for the most part, I will tell you a little bit about where the project is funded from and why. Then we'd like to talk a little bit about ways that we might proceed with the project. But for the most part, number four on the agenda is by far our most interest, which is to really talk about the history, and begin really teasing out what the major issues are, and get some of your own recollections and ideas.

As you can see, we are video and audio recording the entire proceedings. This is not a single event which we will make something out of, but rather, hopefully, one of many events, including individual interviews with some of you and other people, more group sessions, and we'll talk a little bit about, hopefully, some on-line Web collaborative discussion, which is one of the points of the project.

This project is jointly funded by the Sloan Foundation, and the Dibner Institute for the History of Science and Technology, which is about two doors down the street.

There are basically two major points. First of all, we're historians of science and technology. We write about everything from ancient science an technology up through the modern era. But when you really start looking into the period of say, from post World War II and up to the present, the enterprise becomes very, very large and very complicated, actually quite quickly. So you can go look at projects in the 30's that seem to have this kind of intimate scale to them, and then you go to the 50's and you look at projects, and they're just these enormous things. So part of the idea is that we as historians need to begin thinking about writing history in a different way. We are partly doing research in how do you write history of large, complicated, modern scientific technical enterprises, and especially, of course you have the advantage that it's relatively recent, the participants are still around.

The disadvantage, in some way, is that there are these piles and piles and piles of documents. You go back to the Civil War, pre-Civil War period, and you're limited by your sources, and most historians, in some way, are used to being limited by their sources. What you have is a small fragment of what actually happened. Whereas, you go into the Draper Library, or go to NASA and you have so many, too many sources, that you can't possibly read them all in a life time, and you need the guidance of the people who are there to point you partly to what's important and what's not important.

A second major thrust of the project is what is the World Wide Web good for as far as doing history? And it's related obviously, to the first. One of the ideas is given the need to have participants interacting and given the world wide scope of the Web, perhaps we can sort of leverage the world wide Web as a way to get participant's stories into the historical record in a way that's more efficient, more interesting, less expensive, larger scale than simply a story on a tape recorder traveling around to individuals.

That's the background for the overall Sloan/Dibner Project on the History of Recent Science and Technology. You'll see that the Web address is http://hrst.mit.edu. Within that project, there are five sub-projects which we're all doing in parallel. So bio-informatics, the merging of computer science and genetic engineering and biology; material science, a huge topic, of course, but obviously a very important one, especially post World War II. Molecular evolution, a slightly more focused project within debates over both evolutionary theory and molecular biology; and what now they're calling physics of scales, which is basically some fundamental debates about particle physics during the 70's and 80's. Over here we're one of these five Apollo Guidance. So there are five different principle investigators on this project; what you're seeing here is but one of the five. Taken together, for historians it's a large project. It is currently planned to go three years. We're about one year into it, given all the planning we've been doing. My sense is that those three years are even going to go a little bit longer.

When I was invited to participate in this project, I thought, actually not very hard. It was very clear to me that the interesting thing to do would be the Apollo Guidance computer. And I'll give you a little bit of explanation about why I and my group are coming to this particular project at this time.

I'm just finishing a very large book on the history of control systems, and the early history of feedback theory, and essentially the contribution of control systems, control theory, and signal processing to the development of digital computing. So I'm interested in a history of computing that is not so much a history of mathematical geniuses and sort of AI and logic machines, but a history of computing that is much more about engineers building computers to control particular machines. Now the book is very long and very detailed. It covers a lot of things; naval fire control, Gordon Brown and Harold Hazen and their work here during the 1930's. It goes from 1916 to 1945, and it ends in 1945; essentially with Norbert Wiener and cybernetics. Draper Laboratories (the Instrumentation Lab) is a big part of it as well. I have a Mark 14 gun sight up in my office that I was very happy to buy off e-bay last year. It wasn't hard to bid for. There weren't a whole lot of people after it. But I was thrilled to get it. It was essentially in brand new condition. The whole story of the Mark 14 is in there. Draper's collaborations with Gordon Brown, the Instrumentation Lab's work on fire control during the second world war; even earlier work on gyroscope flight instruments; especially Draper's collaborations with Sperry which are very interesting.

As some of you may or may not know, when Draper collaborated with Sperry on the Mark 14, the contracting officer at Sperry was a young aviation lawyer named James Webb. That was the first interaction between Sperry and Webb, and when Webb was at NASA in the 50's he says, "We need one of those? Let's go back to that guy. They did a good job for us during the war." So there are some very interesting earlier connections there.

So the book ends in 1945-1948, roughly, and I have long thought that really the logical end of the story is the Apollo Project. But of course, to do that whole span would have taken me a thousand pages instead of 500.

So we're skipping a little bit in the middle, but this project for me, in terms of my own historical interest, is very much the next logical chapter by my interest in the history of computing the history of control systems, and also the history of manufacturing. One of my real interests is the relationship between manufacturing issues and design issues in the history of control systems and computers. This is obviously an interesting project to look at those questions.

In that sense, we're continuing a kind of longer trajectory of historical research going on at MIT. You'll hear a little bit more, when we introduce ourselves, from Slava Gerovitch, who is working on a somewhat parallel story in the Soviet Union and is joining our project partly to work on the American half of it, and partly to begin looking at what the Soviets are doing at the time, and have some comparative perspective.

So I'm going to turn it over to Sandy, but in sum, we're very excited that you're here. A lot of the little conversations that we heard arising spontaneously are very much the kind of things we'd like to continue, so we're going to try to keep our talking to a minimum, except just what we need to do to introduce you to the overall sense of the site.

We had one meeting, when was this, about last January, with our advisory board, which is a few participants and a few historians, some of whom are here today, and talked a little bit about overall strategy. We'll say more about that, and we would very much like your input on that as well; who to talk to, what the critical issues are, how we should handle -- Obviously this whole thing is imbedded in a much larger project than NASA, which is imbedded in a much larger set of historical developments.

Just one thing that Sandy will cover is that what we'd really like to do is establish a good conversation here, and then sort of segue that onto a Web-based conversation. And that part of it is experimental. To do that, we don't know how to make it work best. We don't know how to really get it going. We have a bunch of ideas. Each of these project is going to do it in a slightly different way. But we decided that the best way to start was to get people face to face, start that conversation.

What we hope to do with probably a small portion of the video but mostly the transcripts, is edit this conversation into a Web-based conversation that a reader can follow. There is some editing required to turn something that's listenable into something that's readable, but really begin to experiment with how to generate conversations about historical topics in engineering on the World Wide Web; both for other scholars who may want to write scholarly books about the topic, and also for the public.

The next thing on our agenda is to talk about how we want to write this history. Because one way to think about it is simply as the history of the computer, which is a thing, or 75 things. But of course, it's continuous with the history of the Instrumentation Laboratory, the history of the Apollo project the history of computing at MIT, a history of the part of the project at Grumman. We're trying not to focus exclusively on the computer, but also about what are the different techniques that contributed to these things coming together at this point.

One of the reasons we study and teach the history of technology at MIT is because one of the things that MIT does is teach engineering. And one of the uses of this project we hope, ultimately, of our project, the historical project, is to put together a set of teaching materials to be used in at least the aero-astro curriculum, if not elsewhere. We've been talking to Ed Crowley, who is the department head over there, and I do this in my other teaching in Course VI (?) here, where we look at engineering projects that have been done in the past, sometimes successes, and sometimes failures. MIT has always struggled with how do you teach engineering students how engineering really gets done. That's not an easy thing to teach. And one of the ways that we approach it is by teaching about historical projects: how things get learned, how mistakes get made, how they get fixed, how lessons get embodied. These kinds of things I think are very valuable, and very much the kind of thing we're trying to draw out for later use as a teaching tool.

I'd like to go around and have people just introduce themselves. I mean, we know your names and who you are, but for the record, a few sentences about what your role was, how you got into it, even where you went since. One of the things we're very interested in is how this project diffused a set of skills, and experiences and technologies into the computer industry, the New England economy, whatever the relevant unit is. We're hoping to try to trace some of the impact of the project by what people went on to do afterwards. So I'll turn it over to Sandy Brown.

Sandy Brown's introduction

site last updated 12-08-2002 by Alexander Brown