23-May-84 19:00:26-EDT,9663;000000000000 Return-Path: Received: from CUCS20 by CU20D with DECnet; 23 May 84 19:00:16 EDT Date: Wed 23 May 84 18:59:04-EDT From: Frank da Cruz Subject: Info-Kermit Digest V1 #1 To: Info-Kermit: ; Reply-To: Info-Kermit@COLUMBIA-20 Info-Kermit Digest Wednesday, 23 May 1984 Volume 1 : Number 1 Today's topics: Info about Info-Kermit New Release of LCTERM for Rainbow MS DOS New Implementation of KERMIT for MUMPS on PDP-11 New Implementation of KERMIT for IBM PC under UCSD p-System Hayes internal modem question UCSD Pascal Kermit for Apple? KERMIT Enhancement -- Changing Filenames ------------------------------------------------------------------------------- Date: Wed 23 May 84 12:04:05-EDT From: Frank da Cruz To: Info-Kermit Subject: Info About Info-Kermit Because of the need to reduce the load on the Columbia University Computer Science Department DEC-20 (COLUMBIA-20) during a period of heavy use and hardware & software development, Info-Kermit will be run as a digest for the next couple months. This is the first issue. Another policy that ARPAnet users should be aware of is the recent restriction on anonymous FTP at COLUMBIA-20, imposed for the same reasons. Anonymous FTP logins are not allowed between the hours of 6:00am and 6:00pm eastern time, weekdays, but are permitted outside these hours. A reminder about how to get KERMIT. For ARPAnet/Internet sites, connect to host COLUMBIA-20, login as user ANONYMOUS with any non-null password. For CCnet (DECnet) sites, use NFT to host CU20B; depending upon the arrangements between Columbia and your site, you may be able to access the files specifying user ANONYMOUS; otherwise contact your system manager. In both cases, all the KERMIT files are in PS:, which you can also refer to using the logical device name KER:. The file KER:00README.TXT contains information on what files are in the KERMIT distribution area and how to find them. One file worth checking from time to time is KER:CURRENT.DOC, a brief tabular listing of each existing version of KERMIT, showing the version number and date of the latest release of each version. KERMIT network distribution is also available to users of BITNET through a server called KERMSRV set up at host CUVMA, an IBM system at Columbia. To get started with KERMSRV, type the following command on your own BITnet host: SMSG RSCS MSG CUVMA KERMSRV HELP For those who cannot obtain KERMIT files over these networks, there remain the DECUS, SHARE, STUG and other user group tapes, various PC or micro floppy distribution services, and Columbia itself will send out tapes for a distribution fee (our tape service is described in KER:FLYER.DOC). Archives of Info-Kermit are in KER:MAIL.*. The messages up to March 8, 1984 are in KER:MAIL.83A. Those from March 8 to May 18 (the last "direct mail" activity) are in KER:MAIL.83B. The current mail (starting with this issue of the digest) are in KER:MAIL.TXT. - Frank ----------------------- Date: Wed 23 May 84 12:04:05-EDT From: Frank da Cruz To: Info-Kermit Subject: New LCTERM for Rainbow MS DOS Version 2.14 of Larry Campbell's LCTERM multi-protocol (including KERMIT and XMODEM) communication program for the DEC Rainbow under MS DOS 2.05 or later is now available. There are many fixes since the last release, and enhancements including a session script facility. The program is available via anonymous FTP from COLUMBIA-20 (ARPANET, only outside of peak hours), or NFT from CU20B (DECNET/CCNET), as KER:RBLCTERM.*. RBLCTERM.EXE is the executable program, in MS-DOS binary format. RBLCTERM.C is the source (actually this file contains the concatenation of various C and ASM86 source files). Installation instructions are in KER:RBLCTERM.HLP. Documentation is in KER:RBLCTERM.DOC. Various other files are also included, including a .BWR file that lists some bugs & restrictions. Since there is no "printable" object module, such as the .HEX or .H86 files of CP/M, or the .FIX file provided with MS DOS (IBM PC) Kermit, you should read the .HLP file before attempting to download LCTERM for the first time. By the way, there will also be a version of Columbia's MS DOS Kermit for the Rainbow, which will appear as part of the next MS DOS Kermit release within the next few weeks. ----------------------- Date: Wed 23 May 84 12:04:05-EDT From: Frank da Cruz To: Info-Kermit Subject: New Implementation of KERMIT for MUMPS-11 Kermit-M is a version of Kermit that is written in 1982 ANSI Standard MUMPS, by David Rossiter of the New York State College of Veterinary Medicine, Veterinary Computing Facility. This version is designed for PDP-11 computers running InterSystem's M/11 operating system (version 5), but instructions are provided for converting to other machines or MUMPS implementations. The files may be found in the KERMIT distribution area as KER:MPKERMIT.*. Some of these files have very long "lines" (over 200 characters), which are apparently unavoidable in the MUMPS language. Thanks to Kate MacGregor of Cornell University for contributing this Kermit implementation through BITNET. ----------------------- Date: Wed 23 May 84 12:04:05-EDT From: Frank da Cruz To: Info-Kermit Subject: New Implementation of KERMIT for IBM PC under UCSD p-System This implementation of KERMIT for the IBM PC p-System was submitted by Kate MacGregor and Steve Pacenka of Cornell University Computing Services. It requires version IV.x of the UCSD p-System, and is also intended to be transferrable to other computers that run the same level of the p-System. It was developed on the IBM PC using NCI release C1F. The files are in KER:UCIBMPC.*. The Pascal source files are concatenated together into the file KER:UCIBMPC.PAS; there are also .DOC and .HLP files. The other UCSD Pascal Kermit from Cornell (the one for the Terak) has been reorganized along similar lines, as KER:UCTERAK.*. ----------------------- Date: Tue 22 May 84 12:04:05-EDT From: Alan Crosswell Subject: Hayes internal modem question. To: info-ibmpc@USC-ISIB.ARPA, info-kermit@CUCS20 I would appreciate it if somebody who has the Hayes internal modem or who looked at one and decided against it would let me know what your experience has been with it. I have heard a rumor that it is unreliable. Is this true? Is it worth the price difference to get the internal modem instead of an async card and external one? The professor who is looking is absolutely convinced that he will never want to attach to an RS232 line but will always use a telephone from his home. Alan Crosswell User Services Columbia University Center for Computing Activities [Editor's note: I don't know if the Hayes modem is good or bad, but there are several good reasons for avoiding internal modems in general: . They take up a slot that might be otherwise put to better use . They can't be easily used on other PCs . They can't be used at all on different kinds of micros . They tend to confuse and/or complicate communication software, like KERMIT People interested in using KERMIT- or MODEM-like programs for transferring files should avoid internal modems.] ----------------------- Date: Tue 22 May 84 18:06:05-PDT From: Bruce Tanner Subject: Apple Pascal Kermit? To: INFO-KERMIT@COLUMBIA-20.ARPA I see from VERSIONS.DOC that there are several UCSD p-system Kermits around, but none for the Apple. Is there one available? Thanks, -Bruce [Not to my knowledge. But see the above announcement for a UCSD Pascal Kermit for the IBM PC. There are now at least three versions -- IBM PC, Terak, HP-98x6 -- to use as a basis. Any volunteers? - Frank] ----------------------- Date: Wed 23 May 84 15:20:05-EDT From: Dave Weaver Subject: KERMIT Enhancment To: cc.fdc@COLUMBIA-20.ARPA I would like to see all flavors of KERMIT offer an option to the RECEIVE or GET commands to output to a file spec other than the same one as you are receiving. Not only would this help with file name length problems, but it would be a valuable asset on VAX systems with DECnet, because one could "RECEIVE" a file to a different node. Something like: Kermit-32> get FILE.EXT NODE"user password"::FILE.EXT and: Kermit-80> get TOPS20_LONG_FILE_NAME.EXTENSION FILE.EXT The latter is because TOPS20 supports long file names and it would be nice to be able to copy them without first copying them to some shorter named file. In the former example I could be talking to a 20 from a VAX which is a DECnet node, but I would like to copy the file to another node other than the one I am running KERMIT from. -Dave [Heartily endorsed. In fact, the Kermit Protocol Manual recommends this; see section 9.1 of the 5th edition. Not many implementations support all these options; KERMIT-20, for instance lets you send FOO (AS) BAR, but not GET FOO (AS) BAR. This should be fixed. But note that there's always the problem of delimitation -- some systems (ITS, VM/CMS) have filenames with imbedded spaces. FTP takes care of this by having you type the source and destination filespecs on separate lines. - Frank] ------- 26-May-84 15:45:42-EDT,6067;000000000000 Return-Path: Received: from CUCS20 by CU20D with DECnet; 26 May 84 15:45:34 EDT Date: Fri 25 May 84 20:28:43-EDT From: Frank da Cruz Subject: Info-Kermit Digest V1 #2 To: Info-Kermit: ; Reply-To: Info-Kermit@COLUMBIA-20 Info-Kermit Digest Friday, 25 May 1984 Volume 1 : Number 2 Today's Topics: Info-Kermit Digest Format Kermit at DECUS Kermit on Cyber Machines (2 messages) Kermit on Epson QX10? Kermit Enhancement - renaming files in get and send ---------------------------------------------------------------------- To: Info-Kermit Date: Friday, 25 May 1984 8:04pm-EST From: Frank da Cruz To: Info-Kermit Subject: Info-Kermit Digest Format There were a few complaints about the format of the first issue of the new Info-Kermit digest. Various digest digesters complained of indigestion. This issue should be better -- the dashed separators are the "right length", there are some asterisks at the end, tabs removed, etc. Let me know of any further problems. ------------------------------ Date: Friday, 25 May 1984 8:04pm-EST From: Frank da Cruz To: Info-Kermit Subject: Kermit at DECUS The new KERMIT releases that were announced in the last issue (UCSD, MUMPS), this issue (Cyber 170), and the next issue (most likely VAX/VMS, TOPS-10, and Professional-350) will be available at Spring DECUS. Brian Nelson will probably be bringing new releases of his PDP-11 Kermit for RSX, RSTS, and maybe RT-11. Attempts will be made to bring all these last-minute releases together into a coherent collection for the various SIG tapes (particularly VMS and RSX), and to make them available on whatever machines are provided by DEC for people to make their own copies. If you need any of these new releases and you're going to DECUS, you might want to bring along a tape (a 2400' reel -- the entire KERMIT distrubution won't fit on anything smaller), as well as some floppies for the Rainbow and Pro-350 versions. This will save you the trouble of FTP'ing a large amount of data over the net, or sending for a tape if you don't have FTP access, as well as avoiding cumbersome bootstrapping procedures for first-time users of Rainbow and Pro Kermit. I'm not going to DECUS myself, but I understand a KERMIT session has been set up for Friday (when most people have left). ------------------------------ Date: Fri, 25 May 84 13:24:38 cdt From: knutson@ut-ngp.ARPA To: cc.fdc@columbia-20.ARPA Subject: New Cyber Kermit You can get the source for the new Cyber Kermit via anonymous ftp to ut-ngp. The files you need are 170kermit.for and 170azlib.asm. These two files should replace the current 170kermit.for. The following are a list of changes made: Version 2.0 - File name packets send uppercase file names, Error packets now handled correctly, modified character tables for CDC 63 and 64 character sets, merged Ric Anderson's NOS/BE code, added push and ! commands, added read delay for performance tuning, changed binary data-mode to ignore NEL characters, OVCAP or SEGLOAD version can be produced, reduced field length. Note that this version now truly supports NOS/BE (aside from differences in front-ends). Ric Anderson of the University of Arizona at Tuscon deserves credit for this. I'll hopefully have a true NOS version in a couple of months. Keep up the good work, Jim Knutson ARPA: knutson@ut-ngp UUCP: {ihnp4,seismo,kpno,ctvax}!ut-sally!ut-ngp!knutson Phone: (512) 471-3241 [Ed Note -- The new files are in KER:170*.* on COLUMBIA-20, available via anonymous FTP in the usual manner. Note that a good deal of the Fortran code of the previous release has been replaced by assembler. Also, there doesn't seem to be any new documentation beyond what's been said above, but then maybe none is necessary.] ------------------------------ Date: Fri, 25 May 84 10:11:04 EST From: decvax!mulga!stephenw.murdu@Berkeley (Stephen Withers) To: cc.fdc@columbia-20.ARPA Subject: Kermit on Cyber Machines You might be interested to know that one of my colleagues has modified Kermit-170 (Cyber) to run on the 815 - we'll send you a copy when it's been more thoroughly tested. Our 815 runs NOS 2.2, by the way, and Darryl Chivers (the one doing the job) also plans a NOS/BE version for the 730, but that's further off. - Steve Withers, University of Melbourne [Let's hope the different Cyber versions will get back together one day... - Ed.] ------------------------------ Date: Wed, 23 May 84 To: info-kermit@columbia-20 Subject: Kermit on QX10 From: fenchel@wisc-rsch.arpa Has anyone brought kermit up on the Epson QX10? Bob Fenchel [Not to my knowledge. Anyone else heard anything? - Ed.] ------------------------------ Date: Thu, 24 May 84 10:31:20 pdt From: Ken Poulton To: info-kermit@csnet-relay.arpa Subject: Kermit Enhancement - renaming files in get and send I agree with the need. I found kermit-20's "send .. (as) .." so handy that I implemented it for Tools and Unix Kermits. Doing the same for Get would be nice, too. [Ed note -- Ken is the author of the Software Tools Kermit, written in Ratfor.] A note on syntax: I chose to allow Send to take any number of files (possibly wildcarded); any filename followed with "-as" is sent with the word following the "-as" as the name. Obviously, you don't want to mix wildcards with "-as", but the "-as" needn't restrict you to one file at a time. In my opinion, files with spaces in their names are a pathological case: try to provide a mechanism to deal with them, but don't restrict your syntax (one file per line) because of them. ------------------------------ End of Info-Kermit Digest ************************* ------- 30-May-84 19:14:06-EDT,8442;000000000000 Return-Path: Received: from CUCS20 by CU20D with DECnet; 30 May 84 19:13:53 EDT Date: Wed 30 May 84 19:06:46-EDT From: Frank da Cruz Subject: Info-Kermit Digest V1 #3 To: Info-Kermit: ; Reply-To: Info-Kermit@COLUMBIA-20 Info-Kermit Digest Wednesday, 30 May 1984 Volume 1 : Number 3 Today's Topics: KERMIT for TRS-80 I and III with TRSDOS New Release of DEC-20 KERMIT DEC Pro-350 & IBM PC Kermit Queries KERMIT for Commodore-64 Underway Fix for Kermit-80 V3.9 ---------------------------------------------------------------------- Date: Mon, 28 May 84 16:55:55 CDT From: Stan Hanks Subject: TRS-80 Kermit To: Frank da Cruz Cc: Stan Barber Frank, Stan Barber has finally completed the TRS80 version of Kermit. Included is all of the source, plus a BASIC program that, when executed, will generate a correct binary. The manual entry for the user's guide is also present. This implementation of the Kermit is written in Z80 Assembler, compatible with M80 and EDAS from Mysosys. It runs under all DOS's available for the TRS80 Model I or Model III (and the Model 4 running in Model III mode). It has been checked out under the following DOS's: TRSDOS 2.3 (Model I), TRSDOS 1.3 (Model III), NEWDOS/80 V 2.0, LDOS 5.1.3, DOS+ 3.5 and VTOS 3.0 (Model I). Stan Hanks [Ed. Note: The files have been renamed for grouping together in the KERMIT distribution as KER:TRS*.*. The file KER:TRSMIT.HLP contains the correspondence between the real names and distribution names. The file KER:TRSMAKE.BAS is a Basic program that generates the executable binary, and KER:TRSMIT.DOC is the KERMIT User Guide manual chapter for this version. All available from COLUMBIA-20 or CU20B in the normal manner.] ------------------------------ Date: Wed 30 May 84 16:47:28-EDT From: Frank da Cruz To: Info-Kermit Subject: New Release of DEC-20 KERMIT There is a new release of KERMIT-20 for the DECSYSTEM-20, 4(222), which fixes a couple bugs in the previous release, 4(215), and should replace that version (or any earlier version). The changes since edit 215 include: Fix FILE NAMING NORMAL-FORM (it was broken in the previous release). . Fix problem with two consecutive ~ chars in data when doing compression. . Don't create empty debugging log files. . Change SET DEBUGGING LOG to LOG DEBUGGING, like other KERMITs. . Add SET FLOW-CONTROL to allow explicit enable/disable of XON/XOFF. . Add SET EXPUNGE to allow enable/disable of automatic expunge on DELETE. . Make LOCAL prefix optional on LOCAL commands like LOCAL DELETE, etc. . Add LOCAL TYPE, LOCAL RUN. . Change LOCAL/REMOTE DISK to LOCAL/REMOTE SPACE, like other KERMITs. . Add code to believe remote TTY line speed under TOPS-20 6.0. . Miscellaneous minor bugs fixed. The new release is in KER:20KERMIT.* on COLUMBIA-20 or CU20B. The file suffixes include: .EXE - The executable program .MAC - The MACRO-20 program source file .UPD - The program update history .DOC - A new KERMIT User Guide chapter for KERMIT-20 .MSS - The Scribe source file for the manual chapter .INI - A sample KERMIT.INI file ------------------------------ Date: 28 May 84 19:02:01 EDT From: Chris Koenigsberg To: To: info-kermit-request@CUCS20 Subject: DEC PRO 350 and IBM PC KERMIT Hello, I am a staff technologist here at CMU's School of Urban and Public Affairs, where they have given us 50 DEC PRO 350 personal computers to play with. I wonder if anyone has successfully assembled a Kermit to run on these beasts. Thanks for all the good work you have done. Chris Koenigsberg Urban Systems Institute CMU SUPA MMC204 (412)578-2175 [Ed. Note -- At least two versions of KERMIT for the Pro-350 have been done; one by Bob Denny of DECUS C fame, and the other by Stevens Institute of Technology. The Stevens version will be released this week or next week, and will be presented at DECUS.] p.s. Is there a new version of the Kermit-86 for the IBM running PC-DOS?? Around here everyone uses version 1.19 or 1.19A....1.19 gets hung up a lot when running things like Emacs on the host, and 1.19A is the "brain-damaged" version which they once thought would eliminate the overloading of the TOPS20 front end by drastically cutting the speed with which packets are sent during uploads. (They were wrong, the front end still gets overloaded, and lots of people are using this brain-damaged very slow upload for absolutely no reason!!) [Ed. Note - Columbia is working on a new release of MS DOS KERMIT, far advanced over the current release, which is 1.20. It should be ready within a couple weeks (I've been saying this for months, but this time I really mean it...). The new version has much improved terminal emulation, particularly in the character insert/delete area. The special CMU customization for slowing things down, pausing between packets, etc, has never proven to be necessary at any other site that I know about, and is probably an artifact of some system modifications.] ------------------------------ Date: 30 May 84 16:43:26 EDT From: Eric Subject: C64 Kermit To: cc.fdc@COLUMBIA-20.ARPA Hi Frank, Many of us C64 owners are distraught at the lack of a good public domain communication/file transfer program like Kermit. We are also distraught at the lack of attention Columbia has decided to give to the project of writing Kermit for this machine. Therefore, I have decided to start the project myself. At present the VT52 emulation is complete, and was written by Frank Prindle (Prindle@NADC) and myself. The remainder of Kermit will be written by me, and later some will be done with the help of Brian Beattie (Beattie@mitre-gateway). If anyone up there has already started to write anything, or formulate some ideas as how to approach the project on the C64, we would like to hear from them. I will keep you posted as to our progress. By the way, we will be writing it in Commodores' Macro Assembler and maintaining a working copy in Cross format. Due to various time restrictions, I don't expect the project to be near completion till the end of the summer. Eric Lavitsky (Lavitsky@Rutgers) [Ed. Note - Good for you! I wish we had been able to do it. We announced our intention to do C64 Kermit because one of our departments had decided to require its students to buy C64s. We even went so far as to buy one ourselves to do the work on. But the project never bubbled up high enough on our priority list to actually get started, what with the MS DOS, CP/M, DEC-20, IBM VM/CMS, Unix, Macintosh, and other implementations that were more important to larger numbers of Columbia users. Good luck!] ------------------------------ Date: 29 May 1984 0242-PDT From: Charles Carvalho Subject: Fix for Kermit-80 V3.9 To: CC.FDC at COLUMBIA-20 Kermit-80 v3.9 will always prefix all &'s in the data with #'s. This should only be done if 8th-bit prefixing has been requested. This problem will only be seen when the other Kermit does not request (or permit) 8th-bit quoting, since Kermit-80 always agrees to use 8th-bit quoting. To fix it, replace the following three lines between gtch4a: and gtch4b:. lxi h,qbchr ;[jd] point to 8-bit quote char cmp m ;[jd] is it our quote character? jz gtch4b ;[jd] yes, have to quote it... with: lda quot8 ; Are we doing 8th-bit quoting? ora a jz gtch4c ; if not, skip test and restore char. lda qbchr ; get 8th-bit quote character cmp d ; same as current character? jz gtch4b ; yes, have to quote it... gtch4c: mov a,d ; no. get character back again. [Ed. Note - This fix will appear in the next release of CP/M-80 Kermit, which Charles is preparing. It will include other fixes, various improvements, but most important it will have the long heralded modular reorganization. Anyone with items to add or fix in KERMIT-80 3.9 should hold off until this new release appears.] ------------------------------ End of Info-Kermit Digest ************************* -------