Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site ucbvax.BERKELEY.EDU Path: utzoo!watmath!clyde!burl!ulysses!ucbvax!umforth From: UMFORTH@WEIZMANN.BITNET (F.I.G.I.L.) Newsgroups: net.lang.forth Subject: Figil Mailing Message-ID: <8512061124.AA25532@ucbvax.berkeley.edu> Date: Wed, 4-Dec-85 10:54:00 EST Article-I.D.: ucbvax.8512061124.AA25532 Posted: Wed Dec 4 10:54:00 1985 Date-Received: Sat, 7-Dec-85 20:12:00 EST Sender: daemon@ucbvax.BERKELEY.EDU Organization: The ARPA Internet Lines: 473 Date: Wed, 4 Dec 85 00:14 EST From: SECRIST%OAK.SAINET.MFENET@LLL-MFE.ARPA Subject: FIGIL letters and Yet Another ET-FIG Newsletter Organization: Science Applications Int'l. Corp., Oak Ridge, Tenn. Geographic-Location: 36 01' 42" N, 84 14' 14" W X-VMS-Mail-To: ARPA%"UMFORTH%WEIZMANN.BITNET@WISCVM.Arpa" > From: F.I.G.I.L. > Subject: Figil letters > To: Richard Secrist > > Date: 11 Nov 85 21:24:58 GMT > From: ihnp4!stolaf!umn-cs!cray!hrp@ucbvax.Berkeley.Edu (Hal Peterson) > Organization: Cray Research Inc., Mendota Heights, MN > Subject: Re: SPECIAL FIGIL POSTING > > In the newsletter contained in the above posting, Richard Secrist > refers to the Novix NC4000 as: > > ... the first chip > > which directly implements a high-level computer language in silicon. > Though I'm not certain, I believe a group at MIT developed a chip which > directly implemented SCHEME (yet another LISP dialect) in silicon, and > they did it in 1980, give or take a couple of years. I know that Guy > Steele, currently of Tartan Labs, was in on the project, and I think > that Gerry Sussman, an MIT professor, was there too. What I'm not sure > of is whether it was a direct or microcoded implementation. Anybody > out there know? Anybody care? > -- > Hal Peterson / Cray Research / 1440 Northland Dr. / Mendota Hts, MN 55120 > UUCP: ihnp4!cray!hrp phone: (612) 681-3085 Hal - Although to my knowledge this is the first actual direct implementation of a higher level language as a chip's machine language per se, this was probably a bit of a brash statement without further research. Over the years I have heard of a number of microcoded attempts at a language-engine. The only one that come to mind at the moment was a Pascal engine that the Western Design Center crreated a number of years ago during the big Pascal craze. I think the project wasn't a success in the marketplace, or maybe worse... can someone fill us in ? I for one would like to hear of such efforts if you run into some facts. It looks like I could use some enlightenment. Richard SECRIST%OAK.SAInet.MFEnet@LLL-MFE.Arpa ----------------------------------------------------------------------------- > Date: November 26, 1985 > From: Tamir (UMFORTH@WEIZMANN.BITNET) > Subject: Forth on an Apple II or // > > I'd like to hear from Apple user's out there as to what system they are > running on the Apple 2 family. I use Micro-Motion's Forth-79. It has > several distinct disadvantages, some of which were dealt with I think > by the new release MasterForth (Forth-83). There was an interesting > implementation which I was considering called SKYFORTH, priced very > competitively and with standard Forth conventions and extensions for > the Apple, + full use of a standard DOS (3.3 or Diversi-Dos > compatible!). For me this was a big advantage (though I realize some > Forther's are wedded to the idea of standard Forth Screens). Anyone > have an practical experience with SkyForth??? Remarks on other > implementations are welcome as well. > > Thanks. > Tamir Weiner (moderator of FIGIL) Tamir - [ Micromotion (F83; DOS) ] I too have munged for many hours on MM F79. I took the MM upgrade to F83, and although it's nice and comes with printed source listings I was turned off VERY quickly when I found out they went back on their no-license-fee policy of their F79 years. It's like $500/image now, public domain or not. No thanks ! One of the things FORTH needs is people donating applications and such to the public domain and this doesn't help a bit. Nice implementation though - fast, DOS interface, graphics, etc.. [ F83-X (F83; DOS) ] An alternative from the Orange County FIG is that Wil Baden has ported Laxen and Perry's F83 to the Apple ][. I haven't had any real time to play with it yet, but the manual says that it is a Forth-83 standard system which includes a TurdStar-like full screen editor, assembler, debugger, LO-RES graphics demos and other utilities. You can compile minimal systems and I think the source is included, It's available for $25 from Offette Enterprises (check their ad in the last couple of FORTH DIMENSIONS). [ MVP (F79; direct access) ] I've got MVP. Sigh. It's a pity someone wouldn't put some more work behind (or should I say 'on top of') this system - editors and the like - because otherwise it's a really portable environment. The basis is there for something great - public domain and all - but not worth the money anymore. (By the by, I got an object with the expert system package. Could anybody net-mail me the source code files ?) [ GraFORTH/TransFORTH (Mutt;DOS) ] As a FORTH people throw asparagus all over this product from the apparantly defunct Insoft company. It's not really fig, nor 79... but GraFORTH is one helluva a graphics package... very impressive, or so I thought. Really nice machine support - 3D HIRES graphic shapes that are TRIVIAL to make. Things can rotate and fly around, and multiplee things can do this. Not really fast, mind you, but slick. [ LMI Z80 FORTH (F83; CP/M-80, ProDOS) ] Strangely enough I bought his for use on my Apple under CP/M. I want their other products - this thing is nicely done. Excellent doc. A full glossary of words WITH examples. A nice TurdStar clone editor, Memory maps, hints... a real PRO job. I can only imagine what their ProDOS version is like. This is what we buy when we have a job to do in FORTH... as does everyone in our FIG chapter who doesn't use polyFORTH for their professional projects. Recommended. Also available for a host of other micros including 8088s,68Ks, and 8080s, to name a few. [ UniFORTH (F83, ProDOS) ] UniFORTH comes in two flavours - the public-domain sampler which is alleged to be out there in BBS land someplace (also available for $35 from Unified Software Systems) or their full 'Professional' system. Comes with a ton of utilties and a full-screen editor and the source to all of it. We've got this product on on VAX and it is even supposed to support text-file I/O. If I can ever get ahold of the sampler for the Apple (or CP/M) I'd probably have nice things to say about it. (Help me find it, folks !) It's also available for about everything... from the VAXen and PDP-11, to the 68K, IBM-PC, and soon even thee Commodore-64. This would be neat because then I could have the same F83 implementations on all of my machines... -=***=- So - more commentary ? I'd like to hear more about SkyFORTH, other FORTHs for the Apple - particularly under ProDOS - and some opinions from you CP/M folks. Oh - my 'tell me about your public domin FORTHs' survey was largely a bust. I have some information I gathered though that I will post Real Soon Now. Richard SECRIST%OAK.SAInet.MFEnet@LLL-MFE.Arpa And now for Yet Another ET-FIG newsletter... -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- cut here .-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- Excerpts from the ET-FIG News Posting #4 in a series -------------------------------------------------------------------------- Volume 1, Number 4 ** East Tennessee FORTH Interest Group ** 09-Nov-84 -------------------------------------------------------------------------- * Software Review: C64-Forth for the Commodore 64 * by Norman E. Smith There IS at least one good Forth-79 available for the Commodore-64. C64-Forth from Performance Micro Products, of 770 Dedham Street-S2, Canton, Mass. 02021. The recent update of the product includes a full screen editor, a 6502 macro assembler, and extentions for strings, point math, and graphics; it also supports the Commodore-64 disk file system. C64-Forth supports up to four Commodore-1541 disk drives. Properly interfaced IEEE-488 disks and printers are also supported. Words that emulate most of the Commodore Basic I/O functions are included. Screens are stored as normal BASIC sequential files; one screen per file. Any of the standard Commodore file types may be read or written using the supplied words. Some of the I/O word extentions are OPEN, GET , PRINT , and CLOSE. The user need not worry about the details of programming the Commodore's I/O. One other note relating to I/O; I could not get C64-Forth to operate with an 80-column video adapter. This is not a big minus, but limits some business applications. A sample of the built in graphics words include ARC to draw arcs (circles, etc.), BCL and BSET to clear and set bits of the high resolution graphics screen, FILSCRN to fill the screen with a specified color, and many more. A sprite editor is included in the user's manual as an example of using some of the graphics words. Most of the graphics routines are coded in Forth assembler. They are not separately loadable though. The floating point routines are patterned after the routines in Commodore Basic and are coded in Forth assembler for speed, and are included in the baseline Forth. Most of the math words the integer words preceeded by 'FP'. For example, integer multiply is '*'; while a floating point multiply is 'FP*'. The compliment of string words generally follow those available in Commodore-64 BASIC. Normal Forth words are prefixed with a '$'; so $! stores a string at the address on the stack. The string words are included in source screens. A M6502 macro assembler is part of C64-Forth. It includes the same structured constructs as normal Forth. All addressing modes are supported. It is included for those words that are just not fast enough in high level Forth. It also provides a convient way to call the ROM routines available in the Commodore-64. The assembler is included in source form. C64-Forth has a full screen editor that provides basic editing functions to edit source screens. The main feature it lacks is the ability to cut and paste lines. Changes can be discarded, updated in memory, flushed, or flushed and loaded. The editor is included as the last portion of the Forth dictionary so it can be deleted prior to building applications for some additional memory savings. This editor is more than adaquate for editing Forth programs. The documenation is excellent. It provides a small but relatively complete tutorial, not only on using standard Forth-79 words, but each of the extentions as well. The Forth tutorial does not replace "Starting Forth", but is very complete and clearly written. C64-Forth is an excellent product for developing turnkey applications. C-64 Forth includes both SAVESYSTEM and SAVETURNKEY words. SAVESYSTEM saves a new version of Forth as currently compiled. This keeps you from having to re-compile an application every time you load Forth during development. Debugged portions of a program can be saved to cut session startup time. The SAVETURNKEY saves the final Forth application, breaking the link with the outter interperter. Once this link is broken, Performance Micro Products charges no royality on programs developed using C64-Forth. There are a couple of negative points though. The disk drive on the Commodore-64 is VERY slow. You don't notice this much editing screens, but compiling a large program requires patience ! Another problem is that the most recent version uses more memory. In terms of the slow disk, this brings up about the only complaint I have with the implementaion. The original version was about 15k, and took long enough to load. The update is 26k -- which takes FOREVER to load. For this reason, I use the updated version when I require all the extra features, but use the original version most of the time. In summary, about the only thing missing is a meta-compiler. If you have a Commodore-64, C64-Forth by Performance Micro Products is an excellent package. [nes] * FORTH TO PLAY MAJOR ROLE IN SPACE MANUFACTURING FACILITY * FORTH is to play a critical role in making the first orbital manufacturing facilities a reality. Recently FORTH, Inc. completed porting its polyFORTH product to the Motorola 68000 and Intel SMP 8085 systems selected for use in the Electrophoresis Operations in Space (EOS) experiment. The EOS project uses a continuous flow electrophoresis device in the weightlessness of space to seperate biological components such as cells, enzymes, and hormones from mixtures sent into orbit. These materials can then be used to produce pharmaceutical products of a purity never before possible. It is believed that such processing abilities could lead to new discoveries on a par with the Salk polio vaccine. The McDonnell Douglas Astronautics company is primary contractor for the 5,000 pound device, which is to be installed in the space shuttle's cargo bay and activated on orbit. FORTH, Inc. wrote the operating system software and implemented a majority of the low-level I/O routines for the space factory. McDonnell Douglas engineers are now developing real-time tasks in the polyFORTH environment to control the entire electrophoresis process as well as monitor the overall status of the factory module. Robert Wood, lead EOS software engineer for McDonnell Douglas, further explains his choice of FORTH for the project: "We are very impressed with polyFORTH. The polyFORTH multi-programmer [environment] simplifies real-time control development. One or more asynchronous tasks control each factory subsystem, reducing complexity and minimizing the effect of performance [degradation] or functional changes. FORTH's unique interpreter allows us to develop code quickly and debug it on the actual flight hardware at normal operating speeds. By using FORTH, we are saving time and money, two critical factors in profitable space operations. We have committed to using FORTH throughout the EOS project." Elizabeth Rather, president of FORTH, Inc. is pleased to be participating in the program, saying that the EOS facility is "a classical example of [the] real-time applications for which polyFORTH was designed". The first model of the EOS device was flown for the first time on the fourth shuttle flight in June 1982. If the remaining flight tests go equally well, the Forth-based production unit could be in Earth orbit by 1987. [rcs] -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- * ET-FIG NEWSLETTER STAFF * Editor: Richard Secrist Staff Writers: Norman Smith, Richard Secrist Publishing: Joseph Minarick Treasurer: Steven A. Wallace * SAIC VAX/PDP-11 FIG-FORTH UPDATED * Richard Secrist and myself implemented the PDP-11 Fig-Forth on a VAX about 9 months ago. It was a "lunch time" project spread out over a couple of months. It currently includes the assembler, editor, and string package from the "PDP-11 Forth User's Guide" by John S. James. The manual was purchased along with the FIG listings and is necessary to create a usable Forth system because it includes the editor, assembler, and hints for integrating Forth with PDP-11 file systems. The editor is line oriented. It is strange until you get used to the way it operates, then seems quite natural. I deceided to implement the editor that is included in the User's Guide for one very good reason; it is a full line editor is just 4 screens! Since we entered it with the Bootstrap Editor (ET-FIG Newsletter, Vol. 1, Number 2) the size of the actual source code was a big consideration! By way of comparison, the standard "Forth, Inc." type line editor published in Forth Dimensions Volume 3, No. 3 is 18 screens long ! The only deficiency with it is that I have never been paticularly fond of its lack of a "delete line" function. You moved to the start of the line to be deleted, then deleted 64 characters. Yuck. So recently I finally got around to adding a Delete Line command. Screen 18, listed below, is the Forth source code for DL (Delete Line). [ Note: I'll have to key this in from hardcopy later for the FIGIL folk. Look for it RSN. -- rcs ] Usage is simple. The DL word takes the line number to be deleted from the stack, calculates the address of the start of the line within the block being edited, then moves 64 spaces to the line. For example, type: 10 DL to delete line 10. The editor numbers lines from 0 to 15. The update is very minor, but adds to the overall usefullness of our Fig implementation quite a bit. If you have access to one of the several copies we have distributed, you will have to modify the load screen (number 1) to include DL with your editor. Screen 18 should be loaded after screen 8. Also, be sure to add a note to your documentation that there is now a DL command in the editor. We have several other updates in the works, although it takes a while to get to them since this is still a "lunchtime" project for us. See Richard Secrist or Norman Smith if you are interested in a distribution copy of our VAX/PDP-11 Forth. Manuals are available directly from the Forth Interest Group. [nes] * ET-FIG NEWSLETTER WRITERS GUIDELINES * [ I don't know how appropriate this is for FIGIL, but I left it in with the hopes that some other FIGish organization might find it usable in some way. -- rcs ] The ET-FIG Newsletter solicits articles from members of our chapter of the Forth Interest Group. We want the newsletter to be a quality publication with a consistently high technical content. This is best acheived through member participation. If you have an idea for an article, the following paragraphs will help get you started. The first three issues have been written by Richard and myself. We will eventually run out of ideas. We try to put together one main article of 1 to 3 pages along with one or two article of half to a full page, as well as FIG related news. A typed page is 55 lines of 65 characters. This translates to about two or three hand-written single spaces pages. Small Forth words make good articles. The Bootstrap Editor is an example. It is useful and small (1 screen). We will be adding reviews of both Forth books and Forth implementations as regular monthly features. The membership database shows a tremendious variety of Forths, so there is no shortage of possible reviews from within our group. Articles should of course be Forth related. If Forth words are involved, they should be generic as possible. We would appreciate machine readable manuscripts, and are capable of processing the following formats : o VAX/PDP-11: text files - 800/1600 bpi tape, RL02 disk pack, or RX01 floppy, o Apple ][,][+,//e: a variety of Apple file types including Applewriter, Magic Window, or generic ASCII text files, o CP/M-80: Wordstar, Perfect Writer, Roff, or ascii text file; 5.25" single sided formats - Kaypro, Morrow, Osborne, Xerox 820, Lobo Max80, or Apple o Commodore-64: Easy Script, Word Writer, Speed Script, or Quick Brown Fox. o MS-DOS: IBM-PC DSDD format floppies, including Wordstar and text files Please include pertinent data, such as filename, which editor was used to create the text, etc.. A hardcopy listing is also helpful. Even handwritten articles are fine, as long as they are readable. Manuscripts should be double spaced with your name at the top of each page. (You DO want credit, don't you ?!) The deadline for articles is two weeks before each monthly meeting. We try to mail the newsletter mailed about a week prior to meetings. Spelling corrections will be made. Editing will be kept to a minimum unless otherwise requested. We are more interested in the content than the grammer. Main articles are the real challenge. So far the main article has been related to group meetings and activities. We realize that this is not always possible. If you have an idea for a main article, give it a shot! Ideally our main article should be 2 to 4 pages long. If you need more space, by all means use it! Neat Forth words are welcomed by all readers, as are your experiences in getting that new Forth up and running. Don't feel left out if you are a beginner. Your experiences are needed to help the other beginners out (and we do have some, so don't be shy !). Everyone had to learn Forth once (or twice) ! Experienced members can relate and may be able to answer your questions as part of group discussions. Even if you just have a good idea for an article and don't want to write one, let us know! We may be able to find someone to turn your input into an article. Book reviews should be Forth related. They should be a half to a full page in length. (At least that is what we think now. We will know for sure after we have a few examples.) The reviews should inform the newsletter reader of several things. The first is style. Is the book readable and informative ? Does the book properly address its intended audience ? What is the intended audience ? Did you like the book ? We want to start running book reviews in the next issue. Are there any volunteers please ? Forth reviews should comment on the overall useability of the implementation being reviewed. Rate the documentation. Does the documentation answer your questions ? Is it indexed ? Are reference cards for the editor and/or Forth included ? How easy is getting to the point of doing useful work ? What is included: a screen editor, a de-compiler, an assembler, etc., and describe any extensions for your machine or environment. The suggested length is one to three pages. A three page review will probably be used as a main article. This issue contains our first Forth review, C64-Forth by Performance Micro Products. Hopefully, it will be both informative to readers and serve as an example for future Forth reviews. There are more than enough versions of Forth in our local group that there should be no shortage of reviews. If multiple articles are received, the ones that are closest to the current meeting topics will be printed first; with others published in future issues. It would be great to have enough articles on hand for a couple of issues ! Because this is a volunteer effort, trying to write all of these article as well as put everything together, copy it, and mail it is a big task. Your submission can help us get done earlier and get the ET-FIG news letter out to you sooner. Remember, the lengths for articles/reviews are just guidelines. We want to publish a variety of information within our self-imposed restriction of a ten page maximum issue size (it works out to one ounce and hence one stamp). If you have an article that takes 5 or 6 or even 10 pages to do it justice, don't let our article size guidelines stop you. We will figure out a way to get in the newsletter ! [nes] ---------- END OF FIGIL DIGEST MAILING ---------------------------- Acknowledge-To: F.I.G.I.L.