3. From ArcMac to the Media Lab
At one point in my undergrad days, probably in 1984, my friend Art took me to visit the MIT Architecture Machine Group (ArcMac). I remember a dark (and cozy) room with a variety of screens showing colorful 2D graphics. I had done some UROP (Undergraduate Research Opportunities Program) work at the MIT AI Lab. I had been interested in AI ever since reading James P. Hogan’s novel “The Two Faces of Tomorow”. But seeing paint-program color graphics revealed a new application of software to me.
That summer I got a job at the Xerox Research Centre of Canada (XRCC), a sister lab to the better-known Xerox PARC. I did not have a car, so I (fondly) remember my two-hour commute each way: bus, subway, subway, bus, walk.
My task for that summer at XRCC was supposed to be developing 3D graphics for displaying molecules for the chemists who worked at the lab. They had just received a new DEC VAX 11/750, which took center stage in their new computer room. Unfortunately, the graphics terminal they had ordered had not arrived. But I did see some photos of a 3D robot arm and martini glass from the vendor of the missing graphics terminal, Raster Technologies. These were the first 3D renderings which made an impression on me. They were created by Alan Barr at, I believe, Caltech.
I wrote some 2D graphics and plotting code for the chemists that summer, in FORTRAN 77 on a DEC VT220 terminal, which could display four shades of gray. Well, green, actually.
But I was definitely bitten by the 3D graphics bug, and on my return to MIT, I headed to ArcMac and got a UROP job at the lab. They had just received a Symbolics 3600 workstation, were about to relocate to their new home, and change their name to the MIT Media Lab. My journey in programming was about to become more exciting.
Design & Code
The image above was generated by an L-system, a grammar for generating self-similar shapes. I thought CL would lend itself well to grammars, and procedural modeling has always been an interest of mine, so it was one of the first things I have coded up in my explorations.
I recall being in the audience at Siggraph 1988 when Przemyslaw Prusinkiewicz presented his paper on generating botanical shapes (i.e. plants) using L-systems. Yet this is the first time I have actually implemented them myself.
