marc@noe.UUCP (Marc de Groot) (12/29/89)
Mitch Bradley writes: >Furthermore, Forth programmers are anarchists at heart (otherwise they >wouldn't buck the establishment and continue to use Forth despite all >the pressure to use C). Absolutely right. Now, how about if the Un*x Forth people continue the anarchist trend and write their own standard? >It's not C programmers that establish the standard; it's the marketplace. > >If Forth comes up with its own standard, who will use it? Successful >Forth systems for the Macintosh use the Macintosh window system. >Successful Forth systems for the Amiga use the Amiga window system. > >In the Macintosh marketplace, anything that doesn't conform to the >Mac user interface is despised. This would be true in the PC marketplace >too, if the PC had a user interface standard (one that came bundled with >ever PC and which most applications used). Fine. Let's come up with a standard that uses the Un*x system calls and C library. Seems like the most natural way to go. >The ANSI team is composed of people and companies who are willing to put >their money and time where their mouths are. By ANSI rules, meetings >must be held at various locations. This requires travel. It probably >ends up costing about $2000 a year, or more, to travel to all the meetings. That's terrible. It locks out people like *me* who can't get away from work (and who would rather put the $2k/year towards buying a house). We have Usenet, and we can design a standard without going anywhere. >Regardless of how you say it "ought" to be, getting a group of opinionated >Forth programmers to agree on anything is really, really, hard. Starting with a group of Un*x Forth programmers will make the job a wee bit easier. >Anybody who does not believe me should try to drive the establishment of >a standard way of doing something in the Forth community. I started >trying to do this for file system interfaces back in 1983, and I have >kept at it ever since. I don't need to try. Thank you for telling me what happens when you stick your finger in the light socket. I can reason by inference that this will happen for me. >Did anybody choose to adopt my 1983 proposal (which was based upon >C "standard I/O" principles and thus a know complete and useable set)? >No. No vendors. Well, I suggest we start the discussion about a Un*x Forth standard with your proposal. I believe that the way the industry is, if we come out with something that works, the industry will gradually shy away from the cackling hens of XJ314 and look at the non-vaporware we present. Besides, I want some- thing NOW. Here's some ideas: - an interpreter coded in C - 16 or 32 bits depending on pointer size on your machine - Set up to work on everything from Version 7 on - Tweaked to use null-terminated strings - Handles strings and floating point with the same routines as Un*x C - Some attempt made to be compatible with IBM PC and other environments where the C library closely matches that of Un*x ^M -- Marc de Groot (KG6KF) |"...few people know what to do with a computer. Noe Systems, San Francisco | They decide that running an operating system UUCP: uunet!hoptoad!noe!marc | is a substitute for doing productive work." Internet: marc@kg6kf.AMPR.ORG | -Chuck Moore
stever@tree.uucp (Steve Rudek) (01/04/90)
In article <744@noe.UUCP>, marc@noe.UUCP (Marc de Groot) writes: > Mitch Bradley writes: > >Furthermore, Forth programmers are anarchists at heart (otherwise they > >wouldn't buck the establishment and continue to use Forth despite all > >the pressure to use C). > > Absolutely right. Now, how about if the Un*x Forth people continue > the anarchist trend and write their own standard? ... > Starting with a group of Un*x Forth programmers will make the job a wee > bit easier. ... > Here's some ideas: What! No curses support! :-( I think you've made an excellent suggestion, Marc. I'm going to be with Unix for the long term and obviously a majority of the rest of you folks are intending the same. I think that Forth and Unix have a lot to lend each other. If Mitch would be willing to share his ideas for memory allocation, files, etc. maybe we can see about getting them to work under the main Forths which are available for Unix. I propose (in keeping with my previous posting) that we stone to death anyone who appears to get too caught up in making things "perfect". Let's go for "good enough" as a first cut and save revisions for next year. I apologise that at first, at least, I probably won't be much help in implementing the standards but I will help as best I can in whatever way I can. -- {pacbell!sactoh0! OR ucdavis!csusac!}tree!stever