The Early Development of Programming Languages (1976) [pdf]

Hey, that's my old boss at Imagen, Luis! He was one of Knuth's fair-haired boys. (Hails from Argentina.)

Luis and Les Earnest spun off Imagen as a start-up from the TeX project at Stanford around 1980, and I joined them soon after that as #4 or #5--memory lost in the mists of time. We were productizing the laser printer prototype Luis and Kok Chen built for the TeX project, using Andy Bechtolsteim's Sun-1 68K-based VME card as the processor. (It was fun hanging around with Knuth socially a bit, since he was on our board.)

The first printers were LBP-10 "wet process" lasers from Canon, which Canon used to explore the laser market, and Imagen got the first one the US, I believe. We worked very closely with Canon, as they prepared to launch the first dry process laser engine, the LBP-CX, which was the heart of our next generation of products, and also the basis for the first Apple LaserWriter.

Don't tell Steve (Jobs), but we got the first LBP-CXs in the US. The Canon folk told us to keep that a secret, as they had to pretend Apple was getting the very first crack at their printers.

Of course, Apple + Adobe blew us away in the end. We bet on a memory-conserving banding image processor with display-list processing on the fly while the paper was moving, while Apple/Adobe said, screw it, we're going to put 1MB of RAM in the printer, and take as long as we need to build the page before printing. Guess one should never bet against RAM getting cheaper over time, but that wasn't obvious at all at that point in history. (I think Apple had decided that reducing the cost of the printer wasn't that urgent.)

It's rare to see a paper by Knuth not in TeX.