Issue No.03 - July-September (2005 vol.27)
Published by the IEEE Computer Society
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MAHC.2005.38
About six years ago, I attended a scholarly history conference that was trying to pay its bills by opening its doors to the general public. The combination of historians and computer scientists produced some interesting discussions during the talks but the two groups tended to keep apart in the breaks between sessions. To be honest, the historians tended to gather in little klatches after the presentations and talk about the ideas in language that was probably unfamiliar to the computer scientists. At one point, I felt a need to bridge the gap between the two groups, so I sought out an attendee I didn't know, introduced myself, and asked what brought my new acquaintance to the conference.
"I'm a hobbyist," he said.
Sensing that he might be shy, I was about to launch into a lengthy sermon on the importance of the amateur in the practice of history. It is a speech I've given often and I have polished it to a high shine. However, my conversation partner interrupted me before I could get into the second sentence.
"I've rebuilt a CDC 6400 in my basement," he added quietly.
For a moment, I was caught off guard and was speechless, a rare state for me. My mind raced through all the old stories I had heard about CDC mainframes: the size and weight of the machines, the amount of electricity they consume, and the heat that flows from the central processor. I would have made a glib remark about his hobby dimming the lights of his neighborhood but I decided to hold my peace.
Listening for once, I heard the story of what it takes to rebuild a working machine when the social infrastructure—the manufacturer and its support staff—is gone. Some parts of the CDC 6400 were scrounged intact from old users. Some elements had to be repaired. A few small pieces had to be built from scratch. His copy of the operating system had to be patched so that it matched his assembled configuration. We were called back to a talk before I could ask, What does one actually do with a working Control Data mainframe?
The act of building a machine designed by others is a fundamental act of history. This is true for both reconstructions, like the Manchester Baby described in this issue by Christopher Burton, or creations of uncompleted machines, like the Babbage Difference Engine built by Doron Swade. In some cases, such as that of the Babbage Difference Engine, the technical work proves that something was feasible. More often, the work of reconstructing an old machine brings us back into the mind of the designer. We learn the nature of the problems that the designer faced, the tradeoffs in each possible solution, and the forces that drove the original workers to their final design choices. You have learned history by walking in the shoes of those who have gone before.
This issue, which presents several successful reconstruction projects, would not have been possible without the work of Doron Swade. Doron did a tremendous piece of technical history when he built a working version of Babbage's second Difference Engine at the time of the Babbage Bicentennial in 1991. It was a project that taught us much about Babbage as an engineer and the way in which brass and steel could be turned into a computing machine. His work to recruit the articles for this issue has brought us a portrait of computer reconstruction work. It may not be quite so dramatic as an operating Difference Engine, or a CDC 6400 in the basement, but I'm sure that you will find it every bit as valuable.