Bakin%HI-MULTICS@sri-unix.UUCP (02/23/84)
From: Jerry Bakin <Bakin @ HI-MULTICS> I haven't seen a discussion of the STAR in here lately, how about it? Are users of the STAR happy with its performance? Does it have a relational data store? How easy is software development on it? We are thinking of getting one to use in a software development environment to produce software automatically for other targets. We would be using the nice graphics (are they available at a subroutine level?) to create all sorts of software related diagrams, (Transition Diagrams, Control Diagrams, Data Flow Diagrams, Petri Nets, etc.) and run those through analysis programs, to generate code for other systems. Is this feasible to ask of a STAR? I wouldn't think twice about asking Multics, but a STAR is sooo much smaller... Jerry. <Bakin @ Hi-Multics>
darrelj@sdcrdcf.UUCP (Darrel VanBuer) (03/03/84)
You can't program a Star. It's a dedicated super word processor.
The same hardware running Interlisp software is called a Dandelion, and is
as fast as a 780 running Interlisp. Great for developing lisp software.
--
Darrel J. Van Buer, PhD
System Development Corp.
2500 Colorado Ave
Santa Monica, CA 90406
(213)820-4111 x5449
...{allegra,burdvax,cbosgd,hplabs,ihnp4,sdccsu3,trw-unix}!sdcrdcf!darrelj
VANBUER@USC-ECL.ARPAguy@rlgvax.UUCP (Guy Harris) (03/06/84)
> You can't program a Star. It's a dedicated super word processor. > The same hardware running Interlisp software is called a Dandelion, and is > as fast as a 780 running Interlisp. Great for developing lisp software. Running an OS whose kernel whose code size in bytes is almost as big as our 4.1c here? Yup, "super" is the word for it. It's a bit more than a "word processor"; it has a mini database system (it's not really relational, but imagine a "relational dbms" with *one* relation and you're not too far off), electronic mail, etc.. There is a development system that runs on the hardware, also; someone told me that the Star applications programs can run on the same machine as the "Mesa Development Environment" under the Pilot OS. So *if* you can pry the Mesa Development Environment out of Xerox, yes, you can program a Star. The Star consists of a Dandelion microprogrammable engine running microcode to give it the "Mesa processor" instruction set and running the Pilot OS with the Star applications software on top of it. The "Mesa processor" is discussed in the Proceedings of the Symposium on Architectural Support for Programming Languages and Operating Systems (ACM SIGARCH Computer Architecture News, Volume 10, Number 2, March 1982 and ACM SIGPLAN Notices, Volume 17, Number 4, April 1982, ACM Order Number 556811) and the OS is discussed in an article in CACM called "Pilot: An Operating System for a Personal Computer" (Communications of the ACM, Volume 23, Number 2, February 1980) and in a paper in the Proceedings of the Eighth Symposium on Operating Systems Principles (ACM SIGOPS Operating Systems Review, Volume 15, Number 5, December 1981). The Mesa language (I've heard it described as "industrial strength Pascal"; it's got the usual sort of abstract data type thingies, as well as a "fork" and "join" primitive for process creation and a *very* PL/Iish signal mechanism) is described in the Mesa Language Manual, Xerox Palo Alto Research Center paper CSL-79-3. Whether you consider the Star a success or failure (they haven't sold many, but then how many of *you* are willing to buy a deskside "office automation" computer that costs $15K, as the Star originally did?), the Mesa processor, the Pilot OS, the Mesa language, and the Star applications software are worth reading about. There are several clever ideas in all of them. Guy Harris {seismo,ihnp4,allegra}!rlgvax!guy is discussed in the