Stack Computers [Guest editor's introduction]

Those of us who have worked with stack computers for several years have found an unending source of amazement-and some consternation-in how few computer professionals there are who seem aware that there is an entirely different approach to computer architecture from the familiar von Neumann model. Sometimes when such phrases as "zero-address machine" have dropped from my lips, and I see that the phrase had about as much meaning for the listeners as "zero-bit word" might have, I have wished that knowledge of stack machines were more widely spread. Fortunately Jack Grimes, the technical editor of Computer, had the same desire and suggested that an issue be devoted to the subject. This is just about the right time to treat the subject of stack machines, what with the rumors rampant about impending LSI stack machines from Intel and Zilog. Another rumor has it that Hewlett-Packard is coming out with an LSI stack machine which will be not unlike the HP3000. I wouldn't even be surprised if Amdahl made an interesting announcement in this area. Microdata has announced plans to actively market a system based on their 32/S. Several other high-level language machines from other vendors will probably be announced in the next couple of years. Although stack machines and high-level language machines are not equivalent, I believe that a stack architecture is best for most of the more common programming languages. I suspect that the new machines to be introduced by some of the semiconductor manufacturers will be called "Pascal machines" by their manufacturers rather than stack machines. Pascal is an Algol-like high-level language developed by Niklaus Wirth with a number of desirable features which the original Algol didn't have. Although Pascal has been around for a number of years, its popularity seems to have started to spread rapidly in the last year or so. One of the reasons for its new popularity is the easy availability of a Pascal compiler written in Pascal. Several articles have appeared describing the ease of bootstrapping up to a compiler for a new machine using this compiler. The compiler generates code for a "hypothetical stack machine." A program must be written to emulate or interpret the code. The idea of building a machine very close to the hypothetical stack machine has obviously occurred to a number of people. I was first introducted to the idea of a stack machine as a student of F. L. Bauer in Germany in 1961. By that time, he and K. Samelson had already patented a specialized stack machine. I remained intrigued by the idea, but had no opportunity to work with a stack machine until 1970, when I participated in the implementation of an on-line system based on the B5500. By 1974 we had 510 terminals attached to two B5700's, which operated 24 hours a day, 7 days a week with query response times of 1 to 3 seconds and data base update times of 6 to 12 seconds. Each B5700 had from six to 15 programs running at the same time with a total main memory of only 192K bytes of 4-microsecond core. Most of the on-line system had been implemented by no more than four programmers. Since then, I have seen no reason to alter my feeling that most computers should-be based on stacks.