[inline]

Walter M. Gibbs' Homepage

Walter M. Gibbs
Engineer
5888 Harrison St.
Redwood City, CA

Hi there, wayward netizen! My name is Walter M. Gibbs, and I was an engineer for the ENCOM corporation up until I retired to lecture and do research in 1989. I live with my wife Joyce in Redwood CIty.

It was back in the early fifties when I began work for the Eckert-Mauchly Computer Corporation, (Actually Remington-Rand) specifically on the mercury delay-line memory technologies for the UNIVAC computers. As a young man (I was just 25 at the time), I was obviously intimidated to work for such innovators. However, after trouncing Howard Aiken in a game of chess, J. Presper Eckert offered me a job. John Mauchly became something of a mentor to me.

After nearly 20 years working on, or programming UNIVAC computers, I went into business with a friend of mine, forming the ENCOM Corporation in 1969. In its early years, ENCOM worked mostly on military contracts (we did some work on the ARPANET project) and as pioneers in encoding and data storage, we set up some of the earliest hard disk arrays for military simulation and artificial intelligence experimentation -- long before anyone had these preposterous notions about "cyberspace".

For almost 10 years I worked at the helm of the ENCOM corporation, in the business office -- not behind a terminal. As soon as I could (1979) I set up a board of directors (perhaps a mistake) and got back into research and development, working on ENCOM's new high-powered laser arrays and input/output subsystems. Dr. Laura Bradley and myself performed some breakthrough research on digitizing matter with a laser beam, with matter actually suspended in the laser beam. The matter in the beam could then be played back out according to a prerecorded model. This was done with a special frequency modulation that I'm still not able to discuss publicly.

My I/O subsystem work was much more mundane -- my main project was called DUMONT, an I/O controller daemon that interfaced with the system's base architecture at the lowest level, to provide immediate terminal or printer based interface between programs and the users operating them, even from system to system. DUMONT was integral to the propagation of the ARPANET, and one of the first multi-threaded pieces of software ever written, able to handle up to 16(!) user requests at once!

Today, the DUMONT program lives on as a highly-modified (read: rewritten) user-interface and requestor daemon. I have since set aside principal work, and Kevin Lenzo has taken over the project and re-dubbed it "infobot" in order to avoid confusion and/or outcry from the general computing community.

I also had a small part in a generally overlooked chapter of computing history that took place in September, 1982. There was a documentary made about the near-disaster called "Tron", and I am told that you can still find it at some videocassette rental shops. I still hold out hope that Kevin Flynn will be found alive and well.

Anyway, I am almost completely retired now, and spend most of my time working in my garden or writing AI routines in FORTH. I also maintain a small "Blog" of interesting Uniform Resource Locators at this address. At any rate, you can find more information about myself and my projects here.


Walter Gibbs: wmg@tastytronic.net last modified 11.01.2000