Minutes of the Special GEnie Forth RT Conference with Bill Muench, president of Ontologic, and coauthor of eForth. Bill discussed eForth, the .asm Forth kernel for the 90's. Entire contents of this transcript copyright (c) 1990 GEnie Forth RoundTable. The contents may be freely copied and distributed in whole or in part provided origination credit is included. Date: 12/13/90 Time: 22:22EST Attendees: [[Bee] FIGGUEST] [[Dennis] D.RUFFER] [[Len] NMORGENSTERN] [GARY-S] [F.TEMPLE] [A.SQUASSABIA] Minutes: The GEnie Forth RoundTable is pleased to welcome this ... evening's special guest, Bill Muench. Bill advises he was born ... into a career as an ontologist early in 1948. He has few ... other interests: hiking, bicycling, swimming, zen, tai chi, .. acoustics of bats, birds and dolphins has taken him to ... remote tropical islands and Alaska, dreams, time dialation, .. the concept of chaos, what is memory and how forgetting and ... mistakes help creativity, etc. Bill needed an extension of ... his perceptions. Forth and computers have allowed him to better ... inteface with the world, more and more clues in pursuit of ... ontology. He works as a consultant and builds custom Forth ... systems. His company is called Ontologic. Tonight Bill is ... here to discuss the .asm based Forth kernel of the 90's, ... eForth, co-authored with Dr. C. H. Ting. .. .. .. .. .. ga Bill <[Bee] FIGGUEST> hello <[Bee] FIGGUEST> eFORTH is derived from my commercial version bFORTH. Both are <[Bee] FIGGUEST> ANS FORTH standard subsets. I designed eFORTH for ease of <[Bee] FIGGUEST> implementing new systems, to easily fit in 8k bytes, to have a <[Bee] FIGGUEST> minimum of machine code and some debugging tools. My debug <[Bee] FIGGUEST> tools are TX!, the transmit primitive, .S, DUMP and WORDS. I <[Bee] FIGGUEST> use a simple meta compiler rather than MASM, which means I <[Bee] FIGGUEST> first write an assembler. That is how I 'get to know' the <[Bee] FIGGUEST> processor. After the new system is running, I optimize it. <[Bee] FIGGUEST> First loading the assembler, then new code words. <[Bee] FIGGUEST> When working, they are added to the kernel, repeat til done. <[Bee] FIGGUEST> I use my host file server, bHOST, to upload source text files <[Bee] FIGGUEST> to the target system. MASM was chosen by Ting as a vehicle to <[Bee] FIGGUEST> communicate eFORTH. It is commonly available, and on a widely <[Bee] FIGGUEST> used platform, the PC. But, any assembler will do. Possible <[Bee] FIGGUEST> alternatives are spreadsheets or wordprocessors with macros. <[Bee] FIGGUEST> <[Len] NMORGENSTERN> You have certainly succeded in your goals!... <[Len] NMORGENSTERN> Do you know how many CPU's that eForth has... <[Len] NMORGENSTERN> been programmed for? <[Bee] FIGGUEST> I know of only 3 at this time <[Bee] FIGGUEST> 8086 8051 and 68000 ga <[Frank] F.SERGEANT> how about an address (& phone#?) where you can be reached? <[Bee] FIGGUEST> while I was thinking , also the 1802 has been done on eFORTH <[Bee] FIGGUEST> my address <[Bee] FIGGUEST> 3901 Branciforte Drive <[Bee] FIGGUEST> Santa Cruz, CA 95065 <[Bee] FIGGUEST> phone <[Bee] FIGGUEST> 408 425 8740 <[Bee] FIGGUEST> <[Len] NMORGENSTERN> Has eForth failed or presented any unexpected problems... <[Len] NMORGENSTERN> on any CPU on which it has been tried? <[Bee] FIGGUEST> there have been difficulties <[Bee] FIGGUEST> 24 bit 56000 and some problems with the 68000 <[Bee] FIGGUEST> its clearing the concepts ga <[Frank] F.SERGEANT> I really like your UM+ and how you do all the math in it .. <[Frank] F.SERGEANT> I don't have much more about eforth, but I'm curious about <[Frank] F.SERGEANT> your thoughts about the RTX & Harris's plans & VME's plans, etc? <[Bee] FIGGUEST> one step back to Len the real problem is been MASM not eFORTH <[Bee] FIGGUEST> <[Bee] FIGGUEST> the UM+ will be <[Bee] FIGGUEST> a machine instruction in a processor that John Rible is <[Bee] FIGGUEST> now desigining looks promicing <[Bee] FIGGUEST> what about Harris and VME Frank ?? frank - care to follow ? <[Frank] F.SERGEANT> any encouraging news? <[Bee] FIGGUEST> At this time no bad news <[Bee] FIGGUEST> a local paper or maybe it is not so local <[Bee] FIGGUEST> EE Times printed a correction about the <[Bee] FIGGUEST> death of the RTX ga Jack Woehr has announced availability of the eForth 8051 on the RCFB,.. where are the others to be found and why not consolidate... the pool of ports to some place in common (like GEnie) <[Bee] FIGGUEST> Richard Haskell just finished the 68000 version <[Bee] FIGGUEST> and presented a paper at FORML <[Bee] FIGGUEST> I think it is just slow at happening but will be on GEnie <[Len] NMORGENSTERN> Have 31 kernel words been enough in every case... <[Len] NMORGENSTERN> that has been tried? Will it be enough for the future? <[Bee] FIGGUEST> that is enough for eFORTH but that is a very <[Bee] FIGGUEST> primitive system what more is needed ga Are you attempting to maintain some control... over what gets called eForth, and by what means ? <[Bee] FIGGUEST> yes Ting and I hope to go over <[Bee] FIGGUEST> each new implementation before <[Bee] FIGGUEST> uploading or publishing them ga please elaborate a bit <[Bee] FIGGUEST> I would like to make sure the colon definitions <[Bee] FIGGUEST> are not changed and the core of code words keeps the same <[Bee] FIGGUEST> I would like to see versions built on other <[Bee] FIGGUEST> assemblers and even some 'C' versions <[Bee] FIGGUEST> Gary S told me of a version compiled on word star not eFORTH <[Bee] FIGGUEST> but another forth <[Bee] FIGGUEST> Richard Haskell may be doing this now <[Bee] FIGGUEST> I would like almost exact copies ga <[Len] NMORGENSTERN> I am curious about a "Word Star" version. Can you ... <[Len] NMORGENSTERN> say more? is here. <[Bee] FIGGUEST> Gary maybe you should but <[Bee] FIGGUEST> I think the editor can replace sequences and create <[Bee] FIGGUEST> the forth tokens. Gary what did you see ? one of the Timex Forths produced by HawgWild was.. written entirely in WordStar non-document with macros and then compiled you can do the same thing with any similar text editor, Word perfect, MS Word... It's been a while, but I can assure you it wasn't that tough. <[Len] NMORGENSTERN> (interjection) That's amazing! <[Dennis] D.RUFFER> Have you had anyone looking into porting it to <[Dennis] D.RUFFER> Windows or OS/2? <[Bee] FIGGUEST> It seems to run quite well on windows <[Bee] FIGGUEST> I was just at John Rible's and it ran quite fast on the 386 <[Dennis] D.RUFFER> John has it running? Glad to hear it!... <[Dennis] D.RUFFER> Is he running it as a WIN app, or a DOS app? <[Bee] FIGGUEST> I do not know enough about windows to answer. <[Bee] FIGGUEST> he is planning to use it to test the designs of his new chip <[Len] NMORGENSTERN> My previous question was really academic... <[Len] NMORGENSTERN> I was interested in the 31 commands as a kind of... <[Len] NMORGENSTERN> "abstract data type" Has there been any other work... <[Len] NMORGENSTERN> on defining an absolute theoretical minimum word set? <[Bee] FIGGUEST> some others have had <[Bee] FIGGUEST> a smaller code word set <[Bee] FIGGUEST> they built AND OR XOR all around NAND but this <[Bee] FIGGUEST> does not seem too useful <[Bee] FIGGUEST> and makes the colon defintions of those other <[Bee] FIGGUEST> words more difficult <[Bee] FIGGUEST> than the code versions ga I maintain the reason figForth and F83 enjoyed their great success... had a lot to do with portability of code across platforms.. I'm gald to see you trying to maintain the core kernel integrity.. is that the reason ? <[Bee] FIGGUEST> reason? <[Bee] FIGGUEST> it Is simplicity is here. what do you ultimately see as eForth's ... real mission, as a teaching tool, or as a programming tool ... or ????... <[Bee] FIGGUEST> most definitely a teaching or learnimg tool <[Bee] FIGGUEST> it is too slow <[Bee] FIGGUEST> as a real programming tool I use it to bring up a new system <[Bee] FIGGUEST> the optimize. I have given it to perple who say <[Bee] FIGGUEST> it is the first time they understood the guts of forth <[Dennis] D.RUFFER> do you have any other tools that can load on top <[Dennis] D.RUFFER> of generic eForth? <[Bee] FIGGUEST> yes <[Bee] FIGGUEST> I have assemblers I use my host program to upload <[Bee] FIGGUEST> anything . I have bench marks case statement <[Bee] FIGGUEST> block tools and even a 260 byte long <[Bee] FIGGUEST> full screen block editor ga can these be made available here in the library ? <[Bee] FIGGUEST> possibly <[Bee] FIGGUEST> when we get some more systems <[Bee] FIGGUEST> I suspect the tools will be also sent <[Bee] FIGGUEST> I hope some one does the word processor version of eFORTH closing comments please bill <[Bee] FIGGUEST> thank you <[Bee] FIGGUEST> I hope any one using or reading eFORTH will post <[Bee] FIGGUEST> comments There is a lot to learn <[Bee] FIGGUEST> about a minimal processor <[Bee] FIGGUEST> thank You again Category 1, Topic 36 is set up... specifically for discussion of eForth.. Bill, on behalf of the GEnie Forth ... RoundTable, Thank you for an interesting evening.

This conference is officially closed. <[Len] NMORGENSTERN> Thank you Bill. Congratulations on eForth... <[Len] NMORGENSTERN> It is a real tour-de-force. === End of Steno notes. ===