============================================================================ From: STAR::"@KL.SRI.COM,@CUNYVM.CUNY.EDU:JAKOB@FRMEU51.BITNET" 14-JAN-1988 02:16 To: SCINT::XRJJM Subj: A summary: "Forth in the VaxWorld" Received: from KL.SRI.COM by STAR.STANFORD.EDU via SMTP with TCP; Wed, 13 Jan 88 18:14:10-PDT Received: from CUNYVM.CUNY.EDU by KL.SRI.COM with TCP; Tue 12 Jan 88 09:46:35-PST Received: from FRMEU51.BITNET by CUNYVM.CUNY.EDU ; Tue, 12 Jan 88 12:36:47 EST Date: 12 JAN 88 18:26- From: JAKOB%FRMEU51.BITNET@CUNYVM.CUNY.EDU To: INFO-VAX@KL.SRI.COM Subject: A summary: "Forth in the VaxWorld" Dear Net, I have received quite a lot of answers to my Forth Language Inquiry. Thanks a lot to everybody who has sent me a note. After a all, I have found a Forth of which I think will fit my needs. (The Forth of Al Woodhull) Here is now the promised Forth summary for the list: It exists a Forth-BBS, the: North Coast Forth Board - (612) 483-6711. You can find there a lot of Forth software and the forth language itself, e.g the two DECUS fig-forth mentioned later. There is even a Forth-List on Earn: FIGIL (Forth Interest Group International List) is now running off a LISTSERV as FIGI-L@SCFVM. To subscribe, send (Jnet) $ SEND LISTSERV@SCFVM SUB FIGI-L "Your full name" There are a two Fig-Forth versions on the DECUS tapes VAX85D and VAX86A, however, even the author of one them doesn't recommend them for real use. They have lot of bugs and an odd file interface. For experementing, they might be useful. If you still would like a copy contact: Billy Barron Bitnet : BILLY@NTSUVAX or AC02@NTSUVAX VAX Programmer/Operator TEXNET : NTVAXB::BILLY or NTVAXB::AC02 North Texas State Univ. Internet : billy%ntvaxb.decnet@utadnx.cc.utexas.edu or Archie Warnock Span: CHAMP::WARNOCK Al Woodhull has written an improved Fig-Forth, but you have to hurry up, he has only time to make a copy until end of january! He writes: I have a Forth based on the F.I.G. VAX version with a much improved file system and a facade based on Vinge's work that makes it a tolerable, but not complete, Forth-83 look-alike. It's good enough to allow porting work between VAX and Laxen-Perry's F83. I can be reached at AWOODHUL@HAMPVMS.BITNET Al Woodhull On the DECUS VAX85C tape you can find the STOIC language, a Forth derviate. The implemention is recommended, there are many examples included. Archie Warnock writes: There is also a commercial Forth system available for the VAX under VMS. It is a full 32-bit version, complete with floating point math support, a full screen editor, text file support, tape handling, multitasking, etc. We use it as a platform for developing an image processing system. I highly recommend it. It is available from Unified Software Systems P. O. Box 21294 Columbus, Ohio 43221 USA Their price is quite reasonable, especially in the VAX world of over-priced software. I'd be glad to try and answer any further questions you might have. Sincerely, Archie Warnock Span: CHAMP::WARNOCK There is also a the Kitt-Peak Public Domain Forth. This Forth has a reasonable file interface, floatingpoint, and fortran subroutine calls implemented. Unfortunately it does NOT follow the slithest standard. EVERY forth word has got a nice new name, so it takes some time to get started. You have to get the manual. Without it you are really lost. Steve Grandi writes: We distribute this Forth (although we haven't touched it in years and you won't get any help if things go wrong; but the price is right). It is far too big to send through BITnet/EARN but we could probably send it through SPAN. Here is the "standard" memo (it is very out of date...we do have 6250 bpi tapes): Steve Grandi ************************************************* The National Optical Astronomy Observatories (operated by the Association of Universities for Research in Astronomy, Inc. under contract with the National Science Foundation) is making available its Forth system for general distribu- tion. This system is a complete Forth system for either the PDP-11 or the VAX-11 computers, under various operating systems. Included in this Forth package is floating point, trig functions and the like, a "portable" file system, mag tape handling and much more. Effort has been made to allow one to develop code that will transport easily be- tween different hardware systems. The supported operating systems are PDP-11 UNIX, VAX-11 UNIX, VAX/VMS, RSX-11M and a stand-alone version for the PDP-11 (PDP, VAX, RSX and VMS are trademarks of Digital Equipment Corporation, UNIX is a trademark of Bell Labora- tories). To obtain a copy of the system send a 2400 foot magnetic tape to: Central Computer Services National Optical Astronomy Observatories 950 N. Cherry Ave. P. O. Box 26732 Tucson, AZ 85726 There is no charge for the system. The system is in the public domain. You must specify which system you want: 1) PDP-11 UNIX ("tar" tape format) 2) VAX-11 UNIX ("tar" tape format) 3) Documentation source, UNIX "troff -ms" format ("tar" tape format) 4) PDP-11 RSX-11M ("Files-11" tape format) 5) VAX/VMS ("Files-11" tape format) 6) PDP-11 Stand-alone system (distributed as RSX-11M source files in "Files-11" tape format) Unless otherwise specified, a 9-track, 1600 bpi tape will be created. The only other option is 9-track, 800 bpi (we don't have 6250 yet). Sorry, but we are not able to handle other tape formats (Backup/Restore, FLX, etc.). The first three items may be sent on a single magnetic tape. All other systems require a single tape per system. Please do not order the stand-alone system without checking with us first as there are certain hardware limitations and restric- tions. Please do not order more systems than you need, as it simply increases our tape copying time. All systems will be sent with a single copy of the documentation. We are trying to limit our distribution to one copy per company. Please feel free to redistribute the system as required within your company. Current hardware/software limitations: 1) The VAX-11 UNIX runs under 4.3bsd. With a few hours work it could run under either 32/V or System III/V, however we do not have either of these systems in-house and do not plan to do it. 2) The PDP-11 UNIX runs under Version 7. It does not require separate I&D space. Use of Forth's "?ter" word requires the addition of a new (5 line) tty ioctl function to return the number of characters on the input queue. Use of the mag-tape words requires the addition of the Berkeley ioctl tape control (we have drivers for the TS11 and TM11). 3) The VAX/VMS system runs under 4.5. There is a funda- mental problem with control-C that we are living with. 4) The RSX-11M system runs under 3.2. We have no plans to run 4.0 and have no idea if the system will run instantly under 4.0. 5) The stand alone system has only run on an 11/23, 11/24 and 11/44. It requires an RL02 disk drive. No floppy support is provided. The FP-11 floating point is the only format supported - the 11/03 floating point is not supported. The system is maintained as an RSX-11M stand-alone task. Steve Grandi, National Optical Astronomy Observatories, Tucson AZ, 602-325-9228 UUCP: $arizona,decvax,hao,ihnp4!noao!grandi or uunet!noao.arizona.edu!grandi Internet: grandi@noao.arizona.edu SPAN/HEPNET: 5356::GRANDI or DRACO::GRANDI I would like to thank John Lundin specially, his contribution was really great! Here it is: There are said to be a few standard Forth's that work well on VAXen, I've never actually seen one. DO NOT try to use the two FIG forths off the spring '85 and '86 Decus VAX symposium tapes for anything important! I contributed one of them. Neither is suitable for software development. Mine has bugs, Roger's probably does. Both still use the oddball "one file per screen" system from the FIG distribution. Also, if you play with them BE SURE to comment out the control-c/y trapping, it can lock you up. Tell me what bugs you find. If you have compatability mode the RSX FIG version is usable, it includes a few necessary extensions, but this is hardly a solution, just easy to get. The best bet VMS version I have heard of ('86 Rochester Conference "BOF") is from McMasters in Canada. They were considering distributing it for a copy charge (they were selling it before), no idea if they carried through. Anyway, it is (was) supported. If I find the address I'll pass it on. I think McMaster's is a BITNET site, I've never gotten around to checking. There were a couple more mentioned, if I can find my notes I'll forward them. I have two versions I work on sporadically myself.. Sebok's 79 standard native mode Unix version (uses naked op sys calls and some odd underpinnings, so not too easy to translate) and my own "goody" version - token threaded, sharable, never gotten past the implementation study and core code stage. Neither is fit for outside viewing. Maybe after Christmas... There is also an infamous version being developed by University of Rochester. They got funding (rumor has it some came from DEC) and apparently came up with a pre-beta version. I left a tape reel with them two and a half years ago for the beta version and it still hasn't arrived. "Just a while longer, we're cleaning up the code". You might try them, they can only say no. Contact Larry Forsley at the Laser Energetics Lab and let me know how you make out. There is a reasonable version written in C, which will run on nearly anything. It is by Mitch Bradley, is nearly '83 standard, and works well but slowly. He is better known for his shareware Forth for the Atari ST520 (he wrote an Emacs in Forth for it, no less. I haven't yet tried it under VAX C, but it works under Wollongong's VMS C compiler. There are other C versions out there. I have one other C version but it's far inferior and incomplete. I have heard of others in Fortran (used in a government project to get around the language requirement), and ADA (similarly, but for an AI project. This one is proprietary though, no chance of getting it), there is even supposed to be one in Algol! If you can consider variants, and if you need something fast, I can recommend STOIC. It is reasonaly well documented, comes with a good set of examples (a calculator, a dialer, a full screen editor) and is free. The examples give information on making system calls and programming. This is available from the VAX85C Decus symposium tape under SAOSTOIC, possibly the Decus library. Another variant, which is a proper forth but an odd dialect, is Kitt Peak Forth, available in a VAX/VMS flavor. I seem to recall that parts of it are written in Fortran, and it is a descendant of "Varian Forth" (?sp), in turn a variant of one of Chuck Moore's early variants for astronomy work. If you use this one don't expect a lot of similarity to F83. I have the manual for it somewhere and will try and dig it up. It is (was) available from Kitt Peak for a copy charge. * * * * * ----------------------------------------------------------------------- John Lundin, Jr. LUNDIN@URVAX.bitnet University of Richmond If anybody has problems of getting one of the Forths or has maybe a general forth question, just feel free to write me. Ps: If somebody needs a vt220 forth full-screen editor (written in Forth), I could make you a copy. But you have to adapt it to your forth! Thanks a lot to everybody who has helped my search! Jakob Magun Jakob Magun SIO/DASOP Bitnet/Earn: "Jakob@frmeu51" Observatoire de Meudon Span: (28.54) Mesioa::jakob 92190 Meudon France