home *** CD-ROM | disk | FTP | other *** search
/ Columbia Kermit / kermit.zip / e / imail.88b < prev    next >
Text File  |  2020-01-01  |  279KB  |  6,584 lines

  1.  
  2.  
  3.  
  4.  
  5.  
  6.  
  7.  
  8.  
  9.  
  10.  
  11.  
  12.  
  13.  
  14.  
  15.  
  16.  
  17.  
  18.  
  19.  
  20.  
  21.             Columbia University Center for Computing Activities
  22.  
  23.  
  24.  
  25.  
  26.                              INFO-KERMIT DIGEST
  27.  
  28.                                   VOLUME 8
  29.  
  30.                          Christine Gianone, Editor
  31.  
  32.                             July - December 1988
  33.  
  34.  
  35. Table of Contents
  36.  
  37.  
  38. Volume 8, Number 1                                                            1
  39.   Announcing MS-DOS Kermit 2.31                                               1
  40. Volume 8, Number 2                                                            5
  41.   Announcing CTOS/BTOS Kermit 1.07                                            5
  42.   Proposal for a New Command                                                 13
  43. Volume 8, Number 3                                                           15
  44.   KERMIT-MS: Please Make Special Keys Available!                             19
  45. Volume 8, Number 4                                                           26
  46.   Announcing C-Kermit for OS/2                                               27
  47. Volume 8, Number 5                                                           36
  48.   Kermit Files Available via UUCP at Oklahoma State U                        37
  49.   Announcing Apple II Kermit Version 3.84                                    37
  50. Volume 8, Number 6                                                           48
  51.   Announcing the End of CU20B                                                48
  52.   Announcing a Maintenance Update for MS-Kermit 2.31                         49
  53.   Announcing MS-Kermit 2.31 for NEC-PC9801 with Kana-Kanji                   50
  54.   Announcing MS-DOS Kermit 2.31 for the Wang PC                              51
  55. Volume 8, Number 7                                                           58
  56. Volume 8, Number 8                                                           69
  57.   Submission of new version of Kermit-UCSD  (UCPECAN)                        69
  58. Volume 8, Number 9                                                           79
  59.   Announcing MS-DOS Kermit 2.31/A, Beta Test                                 80
  60.   Announcing a Beta Test Kermit for MS-Windows                               81
  61.   Announcing EASYK, a Menu-Driven Shell for MS-DOS Kermit                    81
  62.   Announcing Hebrew Kermit, a Test Version for the IBM PC                    81
  63.   MSDOS Kermit Script Files Available from SIMTEL20                          82
  64. Volume 8, Number 10                                                          93
  65.   Announcing a Partial Re-release of TSO Kermit                              93
  66.   Announcing a new version of Kermit for the Z100                            94
  67.   New OS/2 Kermit Files                                                      95
  68. Volume 8, Number 11                                                         105
  69.   Announcing a New Kermit for Microsoft Windows                             105
  70.   Announcing Version 3.1 of Turbo Pascal Kermit                             107
  71.   New xxu.c Utility Program                                                 107
  72.  
  73.  
  74. INFO-KERMIT DIGEST V8 #1                                                 Page 1
  75.  
  76. Info-Kermit Digest         Fri,  8 Jul 1988       Volume 8 : Number 1
  77.  
  78. SPECIAL ISSUE:
  79.  
  80.         Announcing MS-DOS Kermit 2.31
  81.  
  82. Send digest submissions to Info-Kermit@CU20B, requests for addition to or
  83. deletion from the Info-Kermit subscriber list to Info-Kermit-Request@CU20B.
  84.  
  85. Kermit files may be obtained over networks and by mail order.  On the
  86. Internetwork, use FTP to log in to host CU20B, CU20B.COLUMBIA.EDU, or
  87. CU20B.CC.COLUMBIA.EDU (a DECSYSTEM-20), as user ANONYMOUS, using any password,
  88. and GET the desired files from logical device KER:.  You can also get Kermit
  89. files over BITNET/EARN; to get started send a message with text HELP to
  90. KERMSRV, the Kermit file server, at host CUVMA.  For detailed instructions,
  91. read the file KER:AANETW.HLP (AANETW HLP on KERMSRV).  To order by mail,
  92. request a complete list of Kermit versions and an order form from Kermit
  93. Distribution, Columbia University Center for Computing Activities, 612 West
  94. 115th Street, New York, NY 10025 USA.
  95.  
  96. ----------------------------------------------------------------------
  97.  
  98. Date: Fri 8 Jul 88 13:39:29-EDT
  99. From: Christine Gianone <SY.CHRISTINE@CU20B.CC.COLUMBIA.EDU>
  100. Subject: Announcing MS-DOS Kermit 2.31
  101. Keywords: MS-DOS Kermit 2.31
  102.  
  103. This is to announce version 2.31 of MS-DOS Kermit for the IBM PC and PS/2
  104. families and compatibles, the DEC Rainbow, the HP-150 and 110, the Grid
  105. Compass II, and for "generic DOS".  This release replaces version 2.30 of
  106. January 1988.  The majority of the work was done by Professor Joe R. Doupnik
  107. of Utah State University, who also prepared the 2.29 and 2.30 releases.
  108.  
  109. The program requires DOS 2.0 or later, and 100K+ of memory.  The major new
  110. features include an expanded script language, new communication options, Kermit
  111. File Attribute support, improved user and system interfaces, additional
  112. terminal emulation features, and many internal fixes and enhancements, plus
  113. a new manual and other documentation.  The new features are described more
  114. fully at the end of this message.
  115.  
  116. The executable files are stored in a special printable format, called "BOO
  117. files", suitable for BITNET file transfer, electronic mail, etc.  These are
  118. decoded into .EXE files using a "BOO-file decoder" program.  The documentation
  119. is available online in plain ASCII text format, and in Scribe text formatter
  120. source format.  Following is a synopsis of the files.  The KERMSRV name is the
  121. same as the CU20B name, except the "KER:" should be omitted, and the period
  122. between the filename and filetype should be a space, e.g. KER:MSAAAA.HLP on
  123. CU20B is MSAAAA HLP on CUVMA.
  124.  
  125.  CU20B Name        Size        Description
  126.  
  127.  KER:MSAAAA.HLP      7K        Explanation of file naming conventions
  128.  KER:MSB*.*        185K total  BOO-file encoding/decoding programs (many)
  129.  KER:MSVIBM.BOO    113K        IBM PC Kermit, BOO-encoded executable
  130.  KER:MSVRB1.BOO     82K        DEC Rainbow Kermit BOO file
  131.  KER:MSVGEN.BOO     77K        Generic MS-DOS Kermit BOO file
  132.  
  133. Page 2                                                 INFO-KERMIT DIGEST V8 #1
  134.  
  135.  KER:MSVHP1.BOO     77K        HP-150 (untested)
  136.  KER:MSVHPX.BOO     77K        HP-110 and Portable PC (untested)
  137.  KER:MSVGRI.BOO     77K        Grid Compass II (untested)
  138.  KER:MSKERM.DOC    312K        MS-DOS Kermit manual, plain ASCII text
  139.  KER:MSKERM.MSS    302K        Scribe text formatter source for manual
  140.  KER:MSKERM.HLP     16K        A summary of MS-Kermit commands
  141.  KER:MSKERM.BWR     16K        List of known restrictions, bugs, etc.
  142.  KER:MSS*.*        760K total  System-independent MASM Source files (12 files)
  143.  KER:MSG*.*        120K each   System-dependent source (graphics, IBM only)
  144.  KER:MSU*.*      72-87K each   Sys-depn source (keyboard support, all systems)
  145.  KER:MSX*.*     40-168K each   Sys-depn source (port i/o, etc, all systems)
  146.  KER:MSY*.*        106K each   Sys-depn source (terminal emulation, IBM only)
  147.  KER:MSZ*.*        181K each   Sys-depn source (term emul, cont'd, IBM only)
  148.  KER:MSV*.MAK        2K each   Microsoft MAKE files for each version
  149.  KER:MSV*.BAT        2K each   Batch files to build each version
  150.  KER:MSV*.LNK        1K each   LINK command files for each version
  151.  
  152. The utility program MSUCHK.C (and .BOO), contributed by Phil Benchoff, allows
  153. convenient determination of MS-Kermit's keyboard codes on the IBM PC
  154. family.  And the files MSIXSE.* are the XSEND program, contributed by Mark
  155. Zinzow, for constructing MS-Kermit command files that send entire directory
  156. trees from one DOS system to another.
  157.  
  158. Be sure to read the MSKERM.BWR file before trying to use the new version, or
  159. reporting any problems with it.
  160.  
  161. Here are the minimum files needed for the new release ("xxx" stands for the
  162. specific version, IBM, RB1, HP1, HPX, or GEN):
  163.  
  164. 1. For everybody: The documentation -- MSKERM.DOC, MSKERM.HLP, MSKERM.BWR.
  165.  
  166. 2. For those who already have Kermit on their PC:  MSVxxx.BOO.
  167.  
  168. 3. If you don't have an MSBPCT "BOO-file decoder", also get at least MSBPCT.BAS
  169.    (the slow BASIC version) and MSBPCT.BOO (an .EXE version from C, which
  170.    you can decode with MSBPCT.BAS).
  171.  
  172. 4. For those who want to make modifications to the sources:
  173.    MSS*.*, MSGxxx.* (if any), MSXxxx.*, MSYxxx.* (if any), MSZxxx.* (if any),
  174.    MSVxxx.MAK (or .BAT if you don't have MAKE), and MSVxxx.LNK.
  175.  
  176. The systems for which we don't yet have the new version ready are still in
  177. the Kermit distribution as before, under the MSV, MSX, and MSY prefixes.  These
  178. will be replaced as the new ones appear.
  179.  
  180. The IBM PC, DEC Rainbow, and HP-150 versions may also be ordered on diskette
  181. from Columbia, along with typeset, printed copies of the manual.  The IBM
  182. version is available on 5.25-inch 360K DS DD diskettes, and on 3.5-inch 720K DS
  183. diskettes for the PS/2 family.  The Rainbow version is on RX50.  The HP-150
  184. version is on 3.5" diskette.
  185.  
  186. New features for release 2.31 include...
  187.  
  188.  
  189. SCRIPT LANGUAGE:
  190.  
  191.  
  192. INFO-KERMIT DIGEST V8 #1                                                 Page 3
  193.  
  194. Kermit's script language has been significantly expanded.  It has become a
  195. regular little programming language, similar to what's found in the commercial
  196. packages:
  197.  
  198.  - Substitutable parameters in macro invocations, similar to DOS Batch.
  199.  - Variables can be defined, referenced, and undefined.
  200.  - GOTO and labels for transfer of control within a TAKE file or macro
  201.  - An IF statement for testing conditions including SUCCESS or FAILURE of the
  202.    preceding  command (like SEND, GET, RECEIVE, INPUT, etc), or the time of
  203.    day, or whether a file exists, or whether a variable is defined, etc.
  204.  - A loop counting mechanism.
  205.  - A way to reexamine text that has already been INPUT.
  206.  - A way to test for the modem signals CD, CTS, and DSR.
  207.  - Limited interaction with user ("Type any key when ready...")
  208.  
  209.  
  210. A WIDER RANGE OF COMMUNICATION OPTIONS:
  211.  
  212.  - SET PORT COMn (n = 1 to 4).  Support for COM1 through COM4 on the IBM PC/AT
  213.    and PS/2 series.
  214.  - SET PORT BIOSn (n = 1 to 4).  Support for communication ports 1 to 4, using
  215.    Bios-level drivers, like those supplied with certain local area networks.
  216.  - SET PORT NET [<nodename>] as in 2.30, for NetBios-based local area networks.
  217.  - SET PORT UB-NET1 for Ungermann-Bass Net/One.
  218.  
  219. Ungermann-Bass code contributed by Henrick Levkowetz (Philips Kista AG,
  220. Stockholm) and Renne Rehmann (Switzerland)
  221.  
  222. FILE ATTRIBUTES:
  223.  
  224. Kermit File Attributes packets are now exchanged with other Kermit programs
  225. that know about them.  This lets MS-DOS Kermit know the size of incoming files
  226. in advance, lets it reject them if there's not enough disk space, and lets
  227. MS-Kermit display the percent done for incoming as well as outbound files.
  228. Attribute packets also allow arriving files be stored with their original
  229. creation dates and times.  This mechanism has been tested successfully with
  230. Kermit-370 on IBM mainframes, PDP-11 Kermit, and with MS-DOS Kermit itself.
  231.  
  232. A new MAIL command, that lets MS-Kermit send a file to a Kermit server with
  233. instructions (in an Attribute packet) to deliver it as mail to a specified user
  234. (as yet, no Kermit servers can do this).
  235.  
  236.  
  237. USER AND SYSTEM INTERFACE:
  238.  
  239.  - A new HELP command was added (a concise screenful of text, nothing fancy).
  240.  - A new transaction logging feature was added.
  241.  - CD is now a synomym for CWD.
  242.  - Macros can be invoked by name, without DO.
  243.  - Improved operation under MS-Windows (PgUp/PgDn now work).
  244.  - Error messages redesigned to be more helpful.
  245.  - Long Kermit commands in TAKE files can be continued onto subsequent lines.
  246.  - Many new SHOW commands, for different groupings of paramaters.
  247.  - Macro definitions can be shown individually.
  248.  - The Kermit command line interface now permits full 8-bit character inputs.
  249.  - "kermit < todo.lst > todo.log" works.
  250.  
  251. Page 4                                                 INFO-KERMIT DIGEST V8 #1
  252.  
  253.  - "sort < todo.lst | kermit" works.
  254.  - Kermit's return status code is now user-settable, via SET ERRORLEVEL.
  255.  - ERRORLEVEL now set correctly after SEND.
  256.  
  257. TERMINAL EMULATION:
  258.  
  259.  - Tektronix emulation improvements.
  260.  - User can identify graphics adapter explicitly.
  261.  - VGA 640x480 graphics now supported.
  262.  - VT102 scrolling within region speeded up.
  263.  - A new \Kholdscrn verb has been added (VT100 "hold screen").
  264.  - The VT102 emulator now supports escape sequences to control local echoing.
  265.  - Reply to ESC Z "what are you?" query changed from "ESC[?6;2c" to "ESC[?6c".
  266.  - VT52 printer-control escape sequences ESC W and ESC X now work.
  267.  - Ctrl-@ now sends ASCII NUL (0) by default.
  268.  
  269. OTHER CHANGES:
  270.  
  271. The SERVER command now accepts an optional time limit, to make the server
  272. shut down after a certain amount of elapsed time or at a given time of day.
  273.  
  274. The SPACE command no longer requires presence of CHKDSK.COM, and therefore
  275. no longer hangs the MS-Kermit server if it finds something wrong with the disk.
  276.  
  277. Some internal forward references were juggled so that MS-Kermit 2.31 can
  278. be assembled with Microsoft MASM 5.1 (2.30 could not be...)
  279.  
  280. Echoed XOFFs are now detected and no longer cause a deadlock.
  281.  
  282. Under network operation, files opened read-only now have the DOS DENY-NONE bit
  283. set so that competing tasks may access them simultaneously, such as when they
  284. are run by Pushing or RUN within Kermit.
  285.  
  286. The inevitable bug fixes.
  287.  
  288. FINAL WORDS:
  289.  
  290. Like any Kermit program, MS-DOS Kermit is for everyone to use and share.
  291. Please take or order the files you need.  If you're getting files over a
  292. network, please be judicious -- don't ask for KER:MS*.*, or the networks will
  293. be choked for months!  Once you get a working copy at your site, share it with
  294. others, rather than having them make further network requests.
  295.  
  296. Send complaints, bug reports, suggestions, comments, or even praise and
  297. testimonials (where they're deserved) to Info-Kermit@CU20B (.CC.COLUMBIA.EDU).
  298. And thanks once again on behalf of the world's thousands or millions of MS-DOS
  299. Kermit users to Joe Doupnik for his skill, generosity, and patience.   And
  300. thanks also to the beta testers.
  301.  
  302. ------------------------------
  303.  
  304. End of Info-Kermit Digest
  305.  
  306. INFO-KERMIT DIGEST V8 #2                                                 Page 5
  307.  
  308. Info-Kermit Digest         Thu, 14 Jul 1988       Volume 8 : Number 2
  309.  
  310. Departments:
  311.  
  312.   ANNOUNCEMENTS -
  313.         Announcing CTOS/BTOS Kermit 1.07
  314.  
  315.   PORTABLE 370 KERMIT -
  316.         Re: Kermit-370 Report
  317.  
  318.   MACINTOSH KERMIT -
  319.         MacKermit 0.9(40 -Transfer of Mac Applications, How?
  320.         MacKermit Flaw
  321.         MAC <> UNIX Mainframe File Transfer
  322.         Trouble de-BinHexing MacKermit 0.9(40)
  323.         MAC II Session Logging
  324.         MacKermit 0.9(40)
  325.         Suggested Enhancement for MacKermit
  326.         RE: Basic MacKermit Help
  327.         MacKermit Functionality
  328.         Bug in MacKermit?
  329.         Problems with MacKermit 0.9(40)
  330.  
  331.   MISCELLANY -
  332.         C64 kermit Download Problem (2 messages)
  333.         C-Kermit for Intel 310 Xenix 3.0?
  334.         Proposal for a New Command
  335.  
  336. Send digest submissions to Info-Kermit@CU20B, requests for addition to or
  337. deletion from the Info-Kermit subscriber list to Info-Kermit-Request@CU20B.
  338.  
  339. Kermit files may be obtained over networks and by mail order.  On the
  340. Internetwork, use FTP to log in to host CU20B, CU20B.COLUMBIA.EDU, or
  341. CU20B.CC.COLUMBIA.EDU (a DECSYSTEM-20), as user ANONYMOUS, using any password,
  342. and GET the desired files from logical device KER:.  You can also get Kermit
  343. files over BITNET/EARN; to get started send a message with text HELP to
  344. KERMSRV, the Kermit file server, at host CUVMA.  For detailed instructions,
  345. read the file KER:AANETW.HLP (AANETW HLP on KERMSRV).  To order by mail,
  346. request a complete list of Kermit versions and an order form from Kermit
  347. Distribution, Columbia University Center for Computing Activities, 612 West
  348. 115th Street, New York, NY 10025 USA.
  349.  
  350. ----------------------------------------------------------------------
  351.  
  352. Date: Wed, 6 Jul 88 16:27:20 edt
  353. From: Joel Dunn <joeld@uncecs.edu>
  354. Subject: Announcing CTOS/BTOS Kermit 1.07
  355. Keywords: CTOS, BTOS, Convergent NGEN, Burroughs B20
  356.  
  357. Last fall I sent you a copy of Kermit that I had ported to CTOS on Convergent
  358. NGEN systems and BTOS on the Burroughs B20.  I just dropped another version
  359. in the mail that adds a few features and fixes a few bugs.  This version I
  360. call CTOS-Kermit 1.07, and the most significant feature is probably its VT100
  361. emulation in terminal mode.  It isn't a perfect VT100, but it works for most
  362. things without blowing up.  I have had this version in distribution since
  363. January, and have have heard positive feedback, and I have not heard from
  364.  
  365. Page 6                                                 INFO-KERMIT DIGEST V8 #2
  366.  
  367. anyone that was having problems.
  368.  
  369. --Joel Dunn, UNC-Chapel Hill Administrative Data Processing
  370.  
  371. [Ed. - Thanks, Joel!  The new version has replaced the old one as KER:CT*.*
  372. in the Kermit Distribution.  The files and naming conventions are listed in
  373. KER:CTAAAA.DOC.  The program is based on an old release of C-Kermit, but Joel
  374. says that when he gets C-Kermit 4E(070) he will try to add the CTOS/BTOS
  375. support to it, so it will be yet another of the regular C-Kermits.]
  376.  
  377. ------------------------------
  378.  
  379. Date: Fri, 1988 Jul 1   22:21 EDT
  380. From: (John F. Chandler)   PEPMNT@CFAAMP.BITNET
  381. Subject: Re: Kermit-370 Report
  382. Keywords: Kermit-370
  383.  
  384. > Subject: IBM 370 Kermit 4.0 bugs (?), questions
  385. > Keywords: IBM 370 Kermit 4.0
  386. >
  387. > I'm using TSO-Kermit 4.0 on a COMPAREX 7/88 with MVS/XA with VTAM and
  388. > IBM7171 controller. After downloading some files in server mode, I give a
  389. > REMOTE KERMIT TDUMP NAMES command to see the amount of files transferred.
  390. > After that i give a REMOTE KERMIT STATUS command. Now again a REMOTE KERMIT
  391. > TDUMP NAMES command and here the size of the last file is changed to 0K.
  392. > This happens only in server mode, but also with CMS-Kermit 4.0.
  393.  
  394. Yes, the last file in the list is vulnerable -- this can also happen if
  395. you interrupt a multi-file transfer via ^X/^Z.  Someday, I'll see about
  396. fixing this.
  397.  
  398. > The REMOTE TYPE command works, but the TYPE command at the prompt level
  399. > doesn't work. I got a "MODULE 'LIST' NOT FOUND" error. The COPY command
  400. > doesn't work either.
  401.  
  402. TYPE assumes you have a LIST command available on your system -- if you
  403. don't, it should be easy enough to get one.  The same is true of COPY
  404. (IBM and other vendors offer packages including LIST, COPY, MERGE, and
  405. others).
  406.  
  407. > The DIRECTORY command doesn't work with a parameter or with wildcards as
  408. > DIR *.FOR
  409.  
  410. Yes, DIR just uses the LISTCAT facility under TSO.  If you want more
  411. sophisticated options than "LISTC" or "LISTC LEVEL(...)", you'll have
  412. to issue the command directly.
  413.  
  414. > When I'm using long packets, the packet size is at about 995 for a long time
  415. > (many Kbytes) and then suddenly shrinks to about 500 or less. There are only
  416. > 2 or 3 repeat packet for 200K data. So this feature does not satisfy me.
  417.  
  418. By my calculation, 3 repeat packets at 100K of data will reduce the size
  419. to about 700, but the size will then gradually recover.  The size
  420. shouldn't drop to 500 without at least 7 transmission errors.  All this
  421. aside, the throughput with 1000-byte packets is not a great deal more
  422. than that with 500-byte packets, even with a noise-free line, unless the
  423.  
  424. INFO-KERMIT DIGEST V8 #2                                                 Page 7
  425.  
  426. the line turn-around time is large.  See the Kermit Newsletter (summer
  427. 1988) for a discussion of throughput.
  428.  
  429. > How does the kermit HELP command work? I only get "HELP NOT AVAILABLE+",
  430. > while the TSO HELP KERMIT command works well.
  431.  
  432. It seems I left detailed descriptions of the HELP command out of both
  433. the CMS and TSO Kermit documentation!  The syntax is very simple:
  434. for TSO, type either HELP or HELP <subcmd> for information about
  435. Kermit as a whole or about a particular subcommand.  For CMS, type
  436. just HELP.
  437.  
  438. ------------------------------
  439.  
  440. Date: Fri, 10 Jun 88 23:29:11 EDT
  441. From: mholden@ajpo.sei.cmu.edu
  442. Subject: MacKermit 0.9(40 -Transfer of Mac Applications, How?
  443. Keywords: MacKermit 0.9(40)
  444.  
  445. I had been experimenting with MacKermit 0.9(40) without documentation and
  446. having no success in transferring a binary file to my Mac. Upon perusal of
  447. the newly released documentation, I confirmed a suspicion that my problems
  448. were due to MacKermit's inability to perform MacBinary conversion. However,
  449. it is not obvious to me how to compensate and obtain a binary file. Is there
  450. some arcane process involving ResEdit required to reassemble data and
  451. resource forks transmitted as two separate files? (I assume that is how the
  452. pieces must be sent.) Any help with this problem would be appreciated.
  453.  
  454. - Maretta Holden
  455.  
  456. [From Paul Placeway - Since Kermit (and hence MacKermit) is designed to send
  457. only single fork files, the best way to transfer applications is to use
  458. something like BinHex or StuffIt to put the data and resource forks and the
  459. finder info into the data fork of a single file, send that to the other Mac,
  460. and unpack it again.]
  461.  
  462. ------------------------------
  463.  
  464. From: dlt@csuna.UUCP (Dave Thompson)
  465. Subject: MacKermit Flaw
  466. Date: 9 Jun 88 16:57:04 GMT
  467. Keywords: MacKermit 0.9(40)
  468.  
  469. I just finished a cursory evaluation of Kermit 0.9(40) and found one rather
  470. fatal flaw.  There is no option to make the delete key send a backspace
  471. rather than a delete character.  While many of our mainframes (if not most)
  472. accept, or can be made to accept the delete key, we have a few that require
  473. use of the backspace character.  While I guess you *can* use control-h,
  474. that's sort-of tedious.  For those of you who wrote this program, how about
  475. an option to change this (rather as VersaTerm of MacTerminal does)?
  476.  
  477. Dave Thompson             uucp:   {ihnp4|hplabs|psivax}!csun!csuna!dlt
  478. CSUN Computer Center         phone:  (818) 885-2790
  479. 18111 Nordhoff Street, Northridge, CA 91330
  480.  
  481. [From Paul - Bind a macro of a Control-H to the delete key; the easist way to
  482.  
  483. Page 8                                                 INFO-KERMIT DIGEST V8 #2
  484.  
  485. type in the macro is either "\^h" or "\010" (don't include the double quotes).]
  486.  
  487. ------------------------------
  488.  
  489. Date: 16 Jun 88 14:24:25 GMT
  490. From: ut-sally!ut-emx!jezebel@gatech.edu (The Mad Hatter)
  491. Subject: MAC <> UNIX Mainframe File Transfer
  492. Keywords: MacKermit, C-Kermit
  493.  
  494. I have a seemingly simple question:
  495.  
  496. Whenever I do a file transfer from the unix mainframes to the Mac II, the
  497. text file gets all messaed up. Line sequences are disrupted, square boxes
  498. (what does that character mean?) are inserted at the beginning of lines). I
  499. use kermit via a Red Ryder program.  Any suggestions/solutions for this
  500. problem ? Is there any way I can transfer a file EXACTLY the way it is ?
  501. Also, I use Macwrite to open the file on the MAC II.
  502.  
  503. Jim The Mad Hatter
  504.  
  505. [From Paul - The main problem with transfering files between Unix and a Mac is
  506. the newline character.  On Unix, a newline is a Linefeed (Control-J), but on a
  507. Mac it is a Return (Control-M).  If a text file is transfered to a Mac in
  508. binary mode, everything will look fine except that wherever there should be a
  509. newline, there is a little box character instead (indicating a Linefeed
  510. character).  So the real question is do you want the file transfered exactly
  511. as-is, in which case you would use binary mode, or do you want the newlines
  512. converted, in which case you want to use text mode.]
  513.  
  514. ------------------------------
  515.  
  516. Date: Sat, 18 Jun 88 15:49:26 PDT
  517. From: burke%pollux.usc.edu@oberon.USC.EDU (Sean Patrick Burke)
  518. Subject: Trouble de-BinHexing MacKermit 0.9(40)
  519. Keywords: MacKermit 0.9(40)
  520.  
  521.   I have downloaded MacKermit 0.9(40) to my Mac, but have had trouble
  522. with the BinHex step:  BinHex version 5.0 says "no checksum error".  The
  523. MacKermit file is stated to be encoded in BinHex version 4 format.
  524.  
  525.   Do my troubles arise from incompatible versions of BinHex?  Are there other
  526. possibilities I may have overlooked?  Any advice will be appreciated.
  527.  
  528. Thanx much,
  529. Sean Burke
  530.  
  531. [From Paul - Make sure the first line of the file says
  532.  "(This file must be converted with BinHex 4.0)"
  533.  
  534. ckmker.hqx should be 100781 bytes long on a Mac or a Unix box; if it's
  535. too short, re-ftp it.]
  536.  
  537. ------------------------------
  538.  
  539. Date: Thu, 23 Jun 88 08:34:33 PDT
  540. From: GDCWOOD%CALSTATE.BITNET@CUVMA.CC.COLUMBIA.EDU
  541.  
  542. INFO-KERMIT DIGEST V8 #2                                                 Page 9
  543.  
  544. Subject: MAC II Session Logging
  545. Keywords: MacKermit 0.9(36)b4
  546.  
  547. I am having problems with writing a session log when using MacKermit .9(36)B4
  548. on a MAC II.  When choosing write session log from the settings menu,
  549. Kermit goes through the motions of writing a log; however, after discontinuing
  550. the log, the file Kermit session is nowhere to be found!
  551.  
  552. Any pointers on where i'm going astray would be greatly appreciated!
  553.  
  554. Thank you in advance for your time on this one.
  555.  
  556. Dan Calderwood
  557. Humboldt State University
  558. Academic Computing
  559. BITNET:GDCWOOD@CALSTATE
  560.  
  561. [From Paul - There seems to be some problems with the log files being created
  562. invisible on some Macs.  I haven't had it happen to me, and the code that
  563. creates the log file looks fine, so I really don't know what is going on.  The
  564. only thing I can say off hand is make sure you have version 0.9(40).]
  565.  
  566. ------------------------------
  567.  
  568. Date: Tue, 05 Jul 88 14:56:27 EDT
  569. From: <LARRY%NRCVM01.BITNET@cuvma.cc.columbia.edu>
  570. Subject: MacKermit 0.9(40)
  571. Keywords: MacKermit 0.9(40)
  572.  
  573. I am running MacKermit 0.9(40) on a Mac II. I am getting older. My eyes
  574. are getting poorer. And MacKermit uses the top left 1/4 or the screen on
  575. my Mac II! Is there a possibility of MacKermit using the full screen
  576. on a Mac II? Before my eyes give out completely?
  577.  
  578. Larry Bradley
  579. Communications Manager
  580. National Research Council of Canada
  581. Networks Branch
  582. Ottawa, Canada
  583. (613) 993-0240
  584.  
  585. [From Paul - The non-changable MacKermit terminal emulator window is a
  586. problem, but it will take quite a bit of work, because MacKermit knows deep in
  587. its heart that the terminal window is a fixed size and each character is also
  588. a fixed size.  If anyone else does it first, I'll be more than happy to include
  589. it (actually, I'll probably jump up and down 8-)
  590.  
  591. MacKermit 0.9(40) already does background file transfers under MultiFinder
  592. (although it doesn't unhighlite the top window correctly yet).
  593.  
  594. Typeing into the file dialog box will take a bit of work; I'll take a look
  595. at it.]
  596.  
  597. ------------------------------
  598.  
  599. Date:  Tuesday, 19 January 1988 19:59 edt
  600.  
  601. Page 10                                                INFO-KERMIT DIGEST V8 #2
  602.  
  603. From:  David M. Rosenberg <Rosenberg>
  604. Subject:  Suggested Enhancement for MacKermit
  605. Keywords: MacKermit
  606.  
  607. The Macintosh normally uses the OPTION key in combination with certain other
  608. keys (E, I, U, N, & `) as dead key accents.  These dead keys produce no
  609. character themselves.  However, when they are followed by certain
  610. (alphabetic) characters, the dead keys apply the specified accent to the
  611. specified alphabetic character.  This behavior is normally desirable.
  612. However, within Kermit's terminal emulator, you may want to use the OPTION
  613. key as a general modifier (as set in the "Set modifiers
  614. ..." dialog).  In that case, you would probably like all "OPTION-letter"
  615. combinations to act in a uniform mannor, not have a few of them be dead
  616. keys.
  617.  
  618. The VersaTerm terminal emulator provides a switch (to be set with ResEdit)
  619. that controls whether "OPTION dead keys" are enabled or disabled while
  620. VersaTerm is running.  If they are disabled, OPTION-E generates a keydown
  621. event with only one keystroke.
  622.  
  623. I request that you consider adding a similar switch to Macintosh Kermit
  624. (probably in the "Set modifiers ..." dialog) to control whether dead keys
  625. are enabled or disabled while the terminal emulator is running.
  626.  
  627.  /David M. Rosenberg (Rosenberg@BCO-MULTICS.ARPA)
  628.  
  629. [From Paul - As soon as OSU gets a hold of a real MPW 2.0 (we are running
  630. MPW 2.0 a4, w/o the script manager stuff), I will add exactly this, except
  631. that it will be (yet another) dialog settable option.  Actually, the code is
  632. allready there, but not tested.
  633.  
  634. I have had a number of different requests for additions to the key mapping
  635. scheme.  I personally would like to use the Option key as a Meta key (sets the
  636. 8th bit, for Emacs), but I also know of people who would like to use Option as
  637. a control key, as a way of addressing extended characters (vowels with
  638. accents, for example).  There are probably also thing that no one is
  639. interested in doing, so I am taking a survey: what do you want Kermit to do
  640. with the keyboard?  Please send responses to me (paul@tut.cis.ohio-state.edu).
  641. I will summarize the results to Info-Kermit, and work on implementing whatever
  642. isn't but needs to be.]
  643.  
  644. ------------------------------
  645.  
  646. Date: Mon, 11 Apr 88 11:55:49 EDT
  647. From: "Robert E. Zaret" <ZARET%MITVMA.BITNET@CUVMA.COLUMBIA.EDU>
  648. Subject: RE: Basic MacKermit Help
  649. Keywords: MacKermit, RedRyder
  650.  
  651. I just downloaded and then uploaded a text file from CMS-Kermit 3.1 to
  652. RedRyder 10.3 Kermit with no problems.  On the mainframe end, I used
  653. the SEND and RECEIVE commands.  The required settings depend on the
  654. type of connection:
  655.    TTY lines:
  656.         Kermit handshake set to ^Q (ASCII decimal 17)
  657.         general (terminal emulation) flow-control none
  658.    Full-screen lines (via protocol converters such as 7171 or Series/1):
  659.  
  660. INFO-KERMIT DIGEST V8 #2                                                Page 11
  661.  
  662.         Kermit handshake set to none
  663.         general (terminal emulation) flow-control XON/XOFF
  664. I generally avoid server mode (my first impression suggested it can be
  665. flakey and error-recovery can be more difficult).
  666.  
  667. ------------------------------
  668.  
  669. Date: Mon, 11 Jul 88 15:46:06 edt
  670. From: caldara@alliant.Alliant.COM (Steve Caldara)
  671. Subject: MacKermit Functionality
  672. Keywords: MacKermit 0.9(40)
  673.  
  674. Hi,
  675.    I would find the following functionality very useful if it was added to
  676.    MacKermit.
  677.  
  678.     1. Resizable window with zoom box.  (I've requested this previously)
  679.  
  680.         2. Background file transfer under MF ( Ditto above )
  681.  
  682. [Ed. - 0.9(40) has this.]
  683.  
  684.         3. The ability to 'type' the name of the file in the file dialog box
  685.            when uploading files to a remote host.  I do a lot of uploading of
  686.            postscript files to our unix hosts and then print them to a
  687.        laser writer.  I am trying to semi-automate the process with
  688.            'Quick Keys'  After I've created the postscript file using Clover-F
  689.        I use the 'Find File' DA to locate the folder it is in.  When I
  690.        select the MacKermit Send command under the File menu it pops up
  691.        with the corrent folder selected.  At this point tI would like to
  692.        be able to enter the file name from the keyboard and click open.
  693.  
  694.                         Thanx,
  695.                     Steve Caldara
  696.  
  697. P.S.  As is, MacKermit is great.  Thanx to all those who have worked on it.
  698.       These suggestions are requests to make it even betterr
  699.             ( at least for me  :-)   )
  700.  
  701. ------------------------------
  702.  
  703. Date: Wed, 13 Jul 88 15:24:38 PDT
  704. From: jak@violet.Berkeley.EDU (John A. Kunze)
  705. Subject: Bug in MacKermit?
  706. Keywords: MacKermit 0.9(40)
  707.  
  708. I've noticed a fairly annoying bug in MacKermit when connected to Unix
  709. through several versions.  The terminal emulation (with vt100 or vt102
  710. termcap) has a glitch around column 72 that I suspect has something to do
  711. with internal margin bell handling.  If you edit this message and try moving
  712. the cursor back and forth near the ends of the lines in "vi" (using the
  713. commands w, e, l, etc.), you'll notice that "vi" gets confused about the
  714. state of the screen: characters get overwritten with the wrong information
  715. and the line seems to lose a character.
  716.  
  717.     John Kunze
  718.  
  719. Page 12                                                INFO-KERMIT DIGEST V8 #2
  720.  
  721.  
  722. ------------------------------
  723.  
  724. Date: Thu, 14 Jul 88 10:15:19 EDT
  725. From: akk2@tut.cc.rochester.edu (Atul Kacker)
  726. Subject: Problems with MacKermit 0.9(40)
  727. Keywords: MacKermit 0.9(40)
  728.  
  729. I have been using MacKermit 0.9(40) for some time now.  I like the way
  730. keyboard remapping is done.  The keyboard shortcuts (in place of using the
  731. mouse) are nice, but some of them do not work !  In particular, Command-Q
  732. for quitting and Command-S for sending files don't do anything.  Command-Q
  733. just makes my modem beep...
  734.  
  735. Is it only me or is there really a bug here ?  My Mac is a Mac+.
  736.  
  737. Atul Kacker
  738. (akk2@tut.cc.rochester.edu)
  739.  
  740. ------------------------------
  741.  
  742. Date: Thu, 30 Jun 88 11:35:48 PDT
  743. From: GRAFFMAN%VAP@ames-io.ARPA
  744. Subject: C64 kermit Download Problem - Digest V7 #15
  745. Keywords: C64 Kermit
  746.  
  747. Another solution to the C64 download problem short of fixing the hex
  748. file is to run C64DXL.BAS to create the C64DXL.C000 and NOT reset
  749. the 64. Then RUN C64DXL.C000 on the kermit hex file. The end
  750. address will appear but the system will hang. Write the address down,
  751. reset the 64 (power off 'n' on) and run C64DXL.C0000 again to
  752. create the executable.
  753.  
  754. Ira Graffman    NASA/Goddard Space Flight Center
  755.  
  756. [Ed. - See message below...]
  757.  
  758. ------------------------------
  759.  
  760. Date: Thu, 30 Jun 88 22:27:12 EST
  761. From: ray@j.cc.purdue.edu (Ray Moody)
  762. Subject: C64 Kermit Downloading Problem
  763. Keywords: C64 Kermit
  764.  
  765. RE: Info-Kermit Digest V7 #15
  766. >   Many people have a problem with dowloading and converting C-64 Kermit.
  767.  
  768.     The part about the RTS vs. BRK instruction is really a matter of taste.
  769. Most monitors will catch the "BRK" instruction, and exit to BASIC if you use
  770. an RTS instruction.
  771.  
  772.     As to the checksum problem, C64DXL doesn't use the checksum in the last
  773. line anyway.  As soon as it sees the ";00" at the start of the line, it stops.
  774. (Perhaps it should use the checksum....)
  775.  
  776.     Anyway, I am looking for a better way of distributing Kermit over the
  777.  
  778. INFO-KERMIT DIGEST V8 #2                                                Page 13
  779.  
  780. networks.  As anybody who has tried the current method knows knows, it is gross
  781. and ugly.  There will be a new version of Kermit before too much longer.
  782.  
  783.                                 Ray
  784.  
  785. ------------------------------
  786.  
  787. Date: Thu, 7 Jul 88 22:04:22 GMT
  788. From: ASBPF-ED@usfk-emh.arpa
  789. Subject:  C-Kermit for Intel 310 Xenix 3.0?
  790. Keywords: C-Kermit
  791.  
  792. How do I get a copy of Kermit for an Intel 310 running Xenix v3.0?  The
  793. computer does not have a C compiler on it.  Which make file parameters
  794. ckuker.mak?
  795.  
  796.                                                Rodney B. Roeber
  797.  
  798. [Ed. - C-Kermit for Unix, Xenix, etc, only comes in source form.  Is it true
  799. that people have Xenix with no C compiler?  If so, is there anyone out there
  800. who would like to volunteer to distribute Kermit on diskette for these
  801. systems?]
  802.  
  803. ------------------------------
  804.  
  805. Date: Mon, 6 Jun 88 15:04:34 PDT
  806. From: farhad%lyra.usc.edu@oberon.USC.EDU (Farhad Arbab)
  807. Subject: Proposal for a New Command
  808. Keywords: Kermit Protocol
  809.  
  810. I would like to suggest adding a new command to kermit to make
  811. it possible for an arbitrary program to talk to a remote host.
  812. I think this is easy to do for kermits running under Unix, but
  813. it may not be so easy to do for some other systems, e.g. DOS.
  814. My comments, therefore, really refer to C-Kermit under Unix.
  815.  
  816. What I need to do is this:  I want to connect to a remote host that is
  817. running a program P, and carry out a conversation with P.  Since the
  818. behavior of P is very predictable, the conversation on my part is
  819. mostly mechanical.  I would like to replace me on my end with a local
  820. program, L, essentially a sh script.
  821.  
  822. I know about script in kermit but script is too restrictive for my
  823. purpose.  (With script, you can specify only a sequence of expect/send
  824. strings.  The optional sends within an expect string basically allow
  825. you to get the remote to a state where the next send would be
  826. appropriate. I need to send different things depending on the string I
  827. receive, and I need to do some "computation" to figure out what I'm
  828. going to send.)
  829.  
  830. What I need is a new command, say "pipe", which works just as connect,
  831. but instead of connecting the line to /dev/tty, it connects the line to
  832. the standard input and output of a given process: "pipe L" will then do
  833. what I want.  When L terminates, I want to return to the command mode
  834. of kermit.  L can be an arbitrary program (anything that can be
  835. executed via the ! escape).
  836.  
  837. Page 14                                                INFO-KERMIT DIGEST V8 #2
  838.  
  839.  
  840. How easy or difficult is it to add this feature to kermit?  Do you know
  841. of other easy ways to accomplish the same effect?  Is there another
  842. communications program that provides such a feature?
  843.  
  844. Farhad Arbab
  845. arbab@cse.usc.edu
  846.  
  847. [Ed. - There are several possible methods that could be added to C-Kermit.
  848. One of them could be a fully elaborated Kermit-style script language,
  849. complete with variables, IF's, and GOTOs, similar to MS-DOS Kermit 2.31.
  850. This is one of the many, many, many items on the "to-do" list for C-Kermit.]
  851.  
  852. ------------------------------
  853.  
  854. End of Info-Kermit Digest
  855.  
  856. INFO-KERMIT DIGEST V8 #3                                                Page 15
  857.  
  858. Info-Kermit Digest         Wed, 20 Jul 1988       Volume 8 : Number 3
  859.  
  860. MS-DOS KERMIT ISSUE:
  861.                       MS-Kermit v2.29d for HZ100
  862.                       Scan codes for Kermit 2.30
  863.                           More DG/1 Success
  864.                  Kermit for the DG/1 and Archimedes?
  865.                    Problems with Generic MS-Kermit
  866.               Tek4027 or color calls in MSKERM or QKKERM
  867.             KERMIT-MS: Please Make Special Keys Available!
  868.                   MS-Kermit 2.31-test5 Take File Bug
  869.                  MS-Kermit 2.31-test5 Screen Rollback
  870.                      MS-Kermit and VT100 Printers
  871.       Kermit Key Redefinition of Esc-chr: To Show on Mode Line.
  872.                 MS-Kermit (and others) Screen Rollback
  873.                     Selecting MS-Kermit .INI Files
  874.                     Bug in MS-Kermit V2.31-test6 ?
  875.                         MS-Kermit vs Smartcom
  876.                  MS-Kermit Tek-4010 Emulation Colors
  877.                   Bug in Kermit-MS on Tandy 3000HD?
  878.                              Kermit 2.31
  879.               Did AT&T 6300 Graphics Get Canned in 2.31?
  880.  
  881. Send digest submissions to Info-Kermit@CU20B, requests for addition to or
  882. deletion from the Info-Kermit subscriber list to Info-Kermit-Request@CU20B.
  883.  
  884. Kermit files may be obtained over networks and by mail order.  On the
  885. Internetwork, use FTP to log in to host CU20B, CU20B.COLUMBIA.EDU, or
  886. CU20B.CC.COLUMBIA.EDU (a DECSYSTEM-20), as user ANONYMOUS, using any password,
  887. and GET the desired files from logical device KER:.  You can also get Kermit
  888. files over BITNET/EARN; to get started send a message with text HELP to
  889. KERMSRV, the Kermit file server, at host CUVMA.  For detailed instructions,
  890. read the file KER:AANETW.HLP (AANETW HLP on KERMSRV).  To order by mail,
  891. request a complete list of Kermit versions and an order form from Kermit
  892. Distribution, Columbia University Center for Computing Activities, 612 West
  893. 115th Street, New York, NY 10025 USA.
  894.  
  895. ----------------------------------------------------------------------
  896.  
  897. Date: 9 Jun 1988 22:13-PDT
  898. From: SAC.DYESGPF@E.ISI.EDU
  899. Subject: MS-Kermit v2.29d for HZ100
  900. Keywords: MS-DOS Kermit 2.29d, Heath-Zenith/100
  901.  
  902. Now for the main reason for contacting you: I have been using v2.29d
  903. MS-Kermit for Z100s since early april but have been unable to toggle the
  904. printer on/off.  F12 does nothing and SF12 prints the letter L on the
  905. screen.  The documentation I have is for v2.30 for IBMs but I have not found
  906. anything for 2.29d HZ100 Kermit.  Additionally, can we expect to see v2.30
  907. with vt102 emulation for the Z100 sometime or has this project/idea been
  908. dropped?
  909.  
  910. Thank you for any and all assistance.
  911.  
  912. Al Holecek
  913. <SAC.DYESGPF@E.ISI.EDU>
  914.  
  915. Page 16                                                INFO-KERMIT DIGEST V8 #3
  916.  
  917.  
  918. [From jrd - The Z100 code for 2.29 is rather ancient. Drew Derbyshire has
  919. done extensive modifications and updating of the Z100 specific files and
  920. will shortly have a much improved Kermit. He has toyed with the idea of
  921. porting the VT102 emulator but that requires a fair amount of personal time.
  922. A test version of Z100 MS Kermit at the 2.31 level will be available
  923. sometime soon this summer.]
  924.  
  925. ------------------------------
  926.  
  927. Date: Wed, 1 Jun 88 11:04:34 edt
  928. From: Mike Watson <mwatson@NSWC-OAS.ARPA>
  929. Subject: Scan codes for Kermit 2.30
  930. Keywords: MS-DOS Kermit 2.30, Scan Codes
  931.  
  932. Does anyone have a listing of all the scan codes for PC-AT compatibles?  Or a
  933. program to produce such a listing?  Specifically the scan code I am looking for
  934. at the moment is Control-Space on a Zenith 248.  Show key doesn't even
  935. recognize this as being any different than Alt-Space or Just-Plain-Space.  Is
  936. this a legal keystroke under 2.30?  It worked under 2.29B.
  937.  
  938. Please send all responses directly to me and I will summarize for the net.
  939.  
  940.              Arpa:  mwatson@nswc-oas.arpa
  941.          Usenet:  uunet!nswc-oas.arpa!mwatson
  942.  _ _ _                         _ __      _
  943. ' ) ) )      /          /)    ' )  )    ' )   /    _/_
  944.  / / / o _. /_  __. _  //      /--'      / / / __. /  _   ______
  945. / ' (_(_(__/ /_(_(_(<_(/_     /  \ o    (_(_/ (_(_(__/_)_(_) /) )_
  946.  
  947. [From jrd - the system Bios reports space bar as a plain ascii space, no matter
  948. what other keys are held down.  Since there are plenty of other legal key
  949. combinations available, such as the standard Control-@ for NULL, I'd recommend
  950. using one of them.]
  951.  
  952. [Ed. - Cute signature!  You can use the program MSUCHK to conveniently
  953. determine scan codes for use with Kermit 2.30 and later (MSUCHK is the Kermit
  954. Distribution name, the real name is SCANCHEK).]
  955.  
  956. ------------------------------
  957.  
  958. Date: Tue, 31 May 88 22:17:32 EDT
  959. From: "Robert E. Zaret" <ZARET%MITVMA.BITNET@CUVMA.COLUMBIA.EDU>
  960. Subject: More DG/1 Success
  961. Keywords: MS-DOS Kermit 2.30, DG/1
  962.  
  963. I have figured out how to use Kermit with COM1 (internal or external modem).
  964. Specifically, I can use MS-Kermit 2.30 extended development (my copy is dated
  965. 23 April 1988) on my DG/1 Model 1 with the version of DOS 2.11 that came with
  966. it.  Issuing the following command before starting Kermit activates COM1 for
  967. use with an external modem:
  968.  
  969.           mode com1:on
  970.  
  971. Issuing the following commands before starting Kermit activates the internal
  972. modem:
  973.  
  974. INFO-KERMIT DIGEST V8 #3                                                Page 17
  975.  
  976.  
  977.           mode com1:off
  978.           mode com1:on,int
  979.  
  980. Either way, Kermit will start up using BIOS1 (the BIOS version of COM1) and
  981. will be able to control the speed of the port.  I tried these commands with
  982. earlier versions of MS-Kermit, but failed.
  983.  
  984. Problems:
  985.  
  986. I have had trouble using the MPDA (Monitor, Printer, Display Adaptor) and
  987. external monitor at 1200 bps.  I think the main problem is that performance
  988. when Kermit uses BIOS is borderline at 1200 bps, and adding the adaptor and
  989. monitor slows the DG just enough.  I would probably be willing to put up
  990. with missing characters, but cannot tolerate the problems I have had with
  991. cursor keys.  I will try some experiments, but my current guess is as
  992. follows: The cursor keys do not work when I am connected to an IBM mainframe
  993. via a Series/1 protocol converter.  I believe this particular Series/1 is
  994. set up to switch the mode of the cursor keys when the user says the terminal
  995. type is VT100.  I suspect my DG/Kermit combination never "sees" the command
  996. to switch modes.  Simple problem with agravating results.
  997.  
  998. I was unable to use MS-Kermit 2.31 - test 4 (25 April 1988).  I sent details to
  999. Joe D., but thought I should let others know.
  1000.  
  1001. [From jrd - there is not much we can do if the machine is polling the serial
  1002. port via the Bios and its speed is just a hair too slow sometimes.  Normally
  1003. the solution is to request the host to pad material, but here the problem
  1004. arises at login time.  Could padding be setup as part of the login command file
  1005. (if any) or at least have the system repeat the setup soon thereafter?]
  1006.  
  1007. [Ed. - Of course, the ideal solution for the DG/1 is for someone to provide
  1008. system-specific code to take advantage of its features, bypassing the slow
  1009. Bios polling, etc.  Since the DG/1 is apparently IBM-compatible in all respects
  1010. but the serial port, probably all that is needed is an MSXDG1.ASM module, based
  1011. on the MSXIBM.ASM version, modified for the DG/1's UART & internal modem.
  1012. Volunteers?]
  1013.  
  1014. ------------------------------
  1015.  
  1016. Date: Mon, 6 Jun 88 21:00:34 EST
  1017. From: munnari!latvax8.lat.oz.au!ccmk@uunet.UU.NET
  1018. Subject: Kermit for the DG/1 and Archimedes?
  1019. Keywords: MS-DOS Kermit 2.30, DG/1, Archimedes
  1020.  
  1021. I have a Kermit user who is trying to get MS Kermit 2.30 working with his
  1022. DG/1 model 2 with IBM compatible serial port.  He says he has tried various
  1023. combinations of set port, baud, etc with no luck.  Apparently 2.29 did work!
  1024. Is there anyone else with this problem?  If so, are there any solutions?  I
  1025. noticed a message in Info-Kermit Digest V7 #12 stating version 2.30 extended
  1026. development (is that one of the new test releases?) sort of worked.
  1027.  
  1028. Also, is there a port of Kermit for the Acorn Archimedes?  If it is only on
  1029. the very latest distribution tape then can I get a .boo file and unboo
  1030. source for the Archimedes?
  1031.  
  1032.  
  1033. Page 18                                                INFO-KERMIT DIGEST V8 #3
  1034.  
  1035. Thanks
  1036.  
  1037. Mark Kosten            ACSNET: ccmk@latvax8.lat.oz
  1038. Computer Centre          ARPA: ccmk%latvax8.lat.oz@UUNET.ARPA
  1039. La Trobe University    BITNET: ccmk@latvax8.lat.oz
  1040. Bundoora 3083          or via return address on mail header
  1041. Australia              phone:  +61 3 479 2766
  1042.  
  1043. [From jrd - On the DG/1, see the note from Robert Zaret above.  On the Acorn,
  1044. I am unaware of such a port but Alan Phillips in the UK knows if anyone does.]
  1045.  
  1046. ------------------------------
  1047.  
  1048. Date: Tue 28 Jun 88 21:25:43-EDT
  1049. From: Charles Lasner <OC.LASNER@CU20B.CC.COLUMBIA.EDU>
  1050. Subject: Problems with Generic MS-Kermit
  1051. Keywords: MS-DOS Kermit 2.30, Generic Kermit, Victor 9000
  1052. Xref: Sirius, see Victor
  1053.  
  1054. MS-Kermit Generic V2.30 8 Jan 88 does not work on a Victor 9000.
  1055.  
  1056. I tried using SET PORT FILE-HANDLE
  1057. Handle: 3
  1058.  
  1059. and only got a screen full of "3" characters! (and it crashed)
  1060.  
  1061. using handles 1-4 yield unusable results, but at least I could escape back
  1062. from them
  1063.  
  1064. Using 4 caused "device error" to appear infinitely.
  1065.  
  1066. using 1 and 2 causes console echoing (obviously talking to CON:).
  1067.  
  1068. BTW, this never worked before, but at least this generic version warns you
  1069. when you put in fantasy values; former versions crashed with higher values of
  1070. file-handle value.
  1071.  
  1072. Can anyone guess at what's not so "generic" about Victor 9000.  We do have the
  1073. Victor-specific modules, so in theory, the generic version should be fixed.
  1074.  
  1075. cjl (OC.LASNER@CU20B)
  1076.  
  1077. [From jrd - Bryan Peterson is the resident Victor 9000 guru.  Was ANSI.SYS
  1078. installed to read that backspace escape sequence?  On Generic Kermit and file
  1079. handle 3: goodness knows, that's a "normal" DOS handle but the value is not
  1080. enforced by any standard.  Since Generic Kermit polls the serial port device
  1081. via DOS it will operate at only low speeds.  Thus, work on Generic Kermit to
  1082. cater for the Victor 9000 becomes an academic (sic) exercise.]
  1083.  
  1084. [Ed. - Also, if SET PORT FILE-HANDLE doesn't work, you can also try SET PORT
  1085. DEVICE <device-name>.]
  1086.  
  1087. ------------------------------
  1088.  
  1089. Date: Tue, 24 May 88 12:42:12 pdt
  1090. From: Alex Woo <woo@pioneer.arc.nasa.gov>
  1091.  
  1092. INFO-KERMIT DIGEST V8 #3                                                Page 19
  1093.  
  1094. Subject: Tek4027 or color calls in MSKERM or QKKERM
  1095. Keywords: MS-Kermit, Tektronix Emulation
  1096.  
  1097. Has anyone considered putting color calls into either of these programs?
  1098. I'm not sure which tekxxxx after 4014 has color, but it would be nice
  1099. to display color on an EGA screen.
  1100.  
  1101. Alex Woo
  1102. woo@pioneer.arc.nasa.gov
  1103.  
  1104. [From jrd - MS Kermit 2.31 uses the ANSI coloring commands ESC [ 3x ; 4x m
  1105. (x = 1 - 7) to control coloring of individual dots in Tek graphics mode.
  1106. SET TERMINAL COLOR also works when the terminal type is TEK4010.]
  1107.  
  1108. ------------------------------
  1109.  
  1110. Date: 31-May-88 at 16:52 MEZ
  1111. From: Eckart Meyer <I7100501%DBSTU1.BITNET@CUNYVM.CUNY.EDU>
  1112. Subject: KERMIT-MS: Please Make Special Keys Available!
  1113. Keywords: MS-DOS Kermit, VT100 PF Keys
  1114.  
  1115. I use kermit connect mode to have a VT100 terminal on a VAX.  It's a great
  1116. thing but one is missing:  Please, please find a way to make the upper four
  1117. keys of the keypad available in connect mode since they are very frequently
  1118. used by DEC software (Keys PF1...PF4) especially by the editor EDT.  I know
  1119. it's not a trival thing but it should work.  I have a program (without
  1120. sources) which uses these keys just like any other key - the program is an EDT
  1121. for MS-DOS.
  1122.  
  1123. Eckart Meyer                                     Address:   Schleinitzstr. 23
  1124. Inst. f. Nachrichtentechnik                                 3300 Braunschweig
  1125. Technical University of Braunschweig                             West Germany
  1126.                                                  Phone:      +49 531 391 2454
  1127. BITNET:   I7100501@DBSTU1
  1128. Internet: I7100501%DBSTU1.BITNET@CUNYVM.CUNY.EDU
  1129.  
  1130. [From jrd - the four top keys vary considerably from keyboard to keyboard, but
  1131. normally involve Scroll Lock, Num Lock, etc.  The system Bios does not regard
  1132. these keys as being suitable for reporting individually to Bios callers (such
  1133. as MS-Kermit).  The only ways of seeing them individually are to drop down
  1134. below the Bios and fight with any other programs grabbing raw keyboard info.
  1135. MS-Kermit declines to enter those jousting lists. (I have PC-EDT here too.)
  1136. MS Kermit/IBM is shipped with DEC's PF1-4 keys residing on IBM's F1-4 keys.]
  1137.  
  1138. ------------------------------
  1139.  
  1140. Date: Tue, 7 Jun 88 17:53 CET
  1141. From: Helmut Waelder <ZRWA001%DTUZDV1.BITNET@CUVMA.CC.COLUMBIA.EDU>
  1142. Subject: MS-Kermit 2.31-test5 Take File Bug
  1143. Keywords: MS-DOS Kermit 2.31, Take Files
  1144.  
  1145. Have you ever tried the new MS-Kermit 2.31-test5 release with MS-DOS 2.11 ?
  1146. I have discovered that kermit doesn't find any TAKE file, not even the
  1147. MSKERMIT.INI file. Here is an example session:
  1148.  
  1149. C>kermit
  1150.  
  1151. Page 20                                                INFO-KERMIT DIGEST V8 #3
  1152.  
  1153.  ..
  1154. Kermit-MS>take mskermit.ini
  1155. ?Can't find take-file: ake mskermit.ini
  1156. Kermit-MS>
  1157.  
  1158. This happend on a Olivetti M24 with MS-DOS 2.11 and on a Sperry XT with
  1159. MS-DOS 2.11. There are no such problems with MS-DOS 3.x.
  1160.  
  1161. Helmut
  1162. Acknowledge-To: Helmut Waelder <ZRWA001@DTUZDV1>
  1163.  
  1164. [From jrd - Thanks, the problem has been widely reported, and it's fixed in
  1165. the released version of 2.31.  Just in case there is an interest, the problem
  1166. related to tagging files opened read-only with access Deny-None so they could
  1167. be shared, e.g. on LAN file servers.  Kermit now skips the access condition if
  1168. the DOS version is earlier than 3.0.]
  1169.  
  1170. ------------------------------
  1171.  
  1172. Date: Wed, 1 Jun 1988 17:30:50 EDT
  1173. From: Kenneth.H.Lee@cunixc.columbia.edu
  1174. Subject: MS-Kermit 2.31-test5 Screen Rollback
  1175. Keywords: MS-DOS Kermit 2.31, Screen Rollback, CGA
  1176.  
  1177. The latest test Kermit is acting differently in the way that it handles stored
  1178. screens.  Instead of scrolling smoothly through the lines, it blanks the
  1179. screen and puts up the new screen whereever page up or page down is hit.
  1180.  
  1181. /k
  1182.  
  1183. [From jrd - "test5" had revisions to screen initialization which skipped
  1184. sampling the kind of display adapter until the info was needed later.
  1185. Consequently, it used the CGA snow removal screen blanking.  The current
  1186. material tests the kind of display adapter right away and omits snow removal
  1187. for all except CGAs.  MS Kermit 2.30 might produce a disappearing cursor
  1188. with some of the new VGA boards for the same reason; works ok in 2.31.
  1189. Later conversations with Kenneth indicates his CGA board has pecularities
  1190. when asked to scroll a group of lines. Version 2.30 scrolled one line at a
  1191. time, but 2.31 does it in large groups, for speed.]
  1192.  
  1193. ------------------------------
  1194.  
  1195. Date: Wed, 15 Jun 88 14:49:28 CDT
  1196. From: moore@ncsc.ARPA (Moore)
  1197. Subject: MS-Kermit and VT100 Printers
  1198. Keywords: Ms-Kermit 2.31, Transparent Printing
  1199.  
  1200. Is this a known bug in 2.31-test?, or a bug at all?  With older versions of
  1201. Kermit (2.29) and all our Wyse terminals here, we can (with a UNIX alias)
  1202. issue the command p to print a file: p is aliased to echo ^[[5i cat the file
  1203. then echo ^[[4i.  Ever since 2.30 came out (and maybe a little earlier)
  1204. we've had to replace those commands with echo ^[[?5i and ^[[?4i; an
  1205. unfortunate side effect is that output also goes to the screen (which the
  1206. previous string avoids).  Any comments?
  1207.  
  1208. Jim
  1209.  
  1210. INFO-KERMIT DIGEST V8 #3                                                Page 21
  1211.  
  1212.  
  1213. [From jrd - Both forms should work with MS Kermit 2.31.]
  1214.  
  1215. ------------------------------
  1216.  
  1217. Date: Thu, 23 Jun 1988 11:39:22 EDT
  1218. From: Mohib N Durrani <mnd@cunixc.cc.columbia.edu>
  1219. Subject: Kermit Key Redefinition of Esc-chr: To Show on Mode Line.
  1220. Keywords: MS-DOS Kermit, Escape Character
  1221.  
  1222. I redefined the MS-Kermit Escape Character ^] for the ` (ie. \29) and it works
  1223. fine, but the new ` key does not display on the Mode line at the bottom of the
  1224. kermit screen, the old ^] continues to be displayed.
  1225.  
  1226.     How can I change the        Esc-chr: ^]
  1227.     to                          Esc-chr: `
  1228.  
  1229. The previous version of kermit used to automatically change the display of the
  1230. Esc-chr.
  1231.  
  1232. Thanks.
  1233.  
  1234. [From jrd - The Kermit escape character must be a control code and version
  1235. 2.31 rejects non-control codes with an explicit message.  The easy solution
  1236. is to assign verb \Kexit to the accent key (ALT X has this definition too)
  1237. even though ^] will still show on the status line (as it should in this
  1238. case).]
  1239.  
  1240. ------------------------------
  1241.  
  1242. Date: Wed, 20 Apr 88 15:00:31-1000
  1243. From: david@uhccux.uhcc.hawaii.edu (David Lassner)
  1244. Subject: MS-Kermit (and others) Screen Rollback
  1245. Keywords: MS-DOS Kermit 2.31, Screen Rollback
  1246.  
  1247. One of the terminal emulation features included in VersaTerm (an
  1248. excellent package for the Mac) which is incredibly useful is the
  1249. capability to "scroll the screen on full-screen erase" or similar.
  1250. The idea is that before clearing a screen its contents are saved in
  1251. the scrollback buffer so that you can see what you've done during
  1252. your session, even on a full-screen oriented host where scrolling
  1253. almost never occurs.  Would be great to see this in future versions
  1254. of MS-Kermit, MacKermit, and others.
  1255.  
  1256. David Lassner, University of Hawaii Computing Center, 808/948-7351
  1257. INTERNET: david@uhccux.uhcc.hawaii.edu   BITNET: david@uhccux
  1258. UUCP:     {ihnp4,uunet,ucbvax,dcdwest}!ucsd!nosc!uhccux!david
  1259.  
  1260. [From jrd - a good idea David. Next time we'll try to do that. Thanks.]
  1261.  
  1262. ------------------------------
  1263.  
  1264. Date: 27 Jun 88 21:10:55 GMT
  1265. From: herbst@uncecs.edu (Robert T. Herbst)
  1266. Subject: Selecting MS-Kermit .INI Files
  1267. Keywords: MS-DOS Kermit, Initialization Files
  1268.  
  1269. Page 22                                                INFO-KERMIT DIGEST V8 #3
  1270.  
  1271.  
  1272. #include <stdio.h>
  1273. /* C program to allow the selection of a number of .ini files for
  1274. * use with mskermit. The .ini file are the users choice.  A file keys
  1275. * containing the definitions of set key in the .ini file is displayed
  1276. * using the MSDOS type command. This program "select.c" has been
  1277. * compiled using POWER C and has run successfully.
  1278. * Robert T. Herbst University of North Carolina at Wilmington.
  1279. */
  1280. main()
  1281. {
  1282.     int c;
  1283.     for(;;)
  1284.     {
  1285.     printf("Enter 4 for 2400b 2 for 1200b h to exit\n");
  1286.     system("type keys");
  1287.     while( ( c = getchar()) != '\n')
  1288.     switch(c)
  1289.     {
  1290.         case '4': system( "kerm230 -f mskermit.ini");
  1291.             break;
  1292.         case '2': system( "kerm230 -f m1200.ini");
  1293.             break;
  1294.         case 'k': system("kerm230");
  1295.             break;
  1296.         case 'h': exit(0);
  1297.             break;
  1298.         default : break;
  1299.     }
  1300.     }
  1301.  
  1302. }
  1303.  
  1304. [From jrd - Thanks Robert.  Neat.  One might also note that a Batch file can
  1305. be used with a similar mechanism via %1.  MS-Kermit 2.31 has so many new
  1306. control features in scripts and macros that it is possible to merge several
  1307. ini files into one, with separate macros defined for each condition and then
  1308. use one or another as extended Kermit commmands.  The new User's Manual has
  1309. full details.]
  1310.  
  1311. ------------------------------
  1312.  
  1313. Date: Wed, 29 Jun 88 17:50 CET
  1314. From: Helmut Waelder <ZRWA001%DTUZDV1.BITNET@CUVMA.CC.COLUMBIA.EDU>
  1315. Subject: Bug in MS-Kermit V2.31-test6 ?
  1316. Keywords: MS-DOS Kermit 2.31
  1317.  
  1318. When I'm downloading files from an IBM host (CMS-kermit 4.0) with MS-kermit
  1319. v2.31-test6 the files don't have a date or timestamp in the directory.
  1320. This happens with PC-DOS 3.3 and MS-DOS 3.2, not with MS-DOS 3.1
  1321. Greetings
  1322.          Helmut
  1323.  
  1324. [Ed. - This is fixed in the final 2.31 release.]
  1325.  
  1326. ------------------------------
  1327.  
  1328. INFO-KERMIT DIGEST V8 #3                                                Page 23
  1329.  
  1330.  
  1331. Date: Tue, 5 Jul 88 13:48 MDT
  1332. From: Joe Doupnik <JRD%USU.BITNET@CUVMA.CC.COLUMBIA.EDU>
  1333. Subject: MS-Kermit vs Smartcom
  1334. Keywords: Smartcom
  1335.  
  1336.         Just a comment on a situation seen locally.  If a person runs the
  1337. Hayes Smartcom program SCOM and then runs Kermit the computer may hang.
  1338. As best that I can make out SCOM captures the serial port interrupts but
  1339. when it exits it leaves those interrupts pointing at itself (or at whatever
  1340. occupies the memory next).  Thus, Kermit or another program can come along
  1341. and get stung by those loose ends.  So, if some people remark on the situation
  1342. we have a probable answer.  There's nothing Kermit can do about it.
  1343.         Regards,
  1344.         Joe D.
  1345.  
  1346. [Ed. - We have had some calls about this too.  The offender is apparently
  1347. Smartcom II, not III.  One caller claimed to be able to clear up the problem
  1348. by "reconfiguring the modem" in Smartcom.  Failing that, obviously, one can
  1349. Ctrl-Alt-Del to put things back...  Also see the comments below on the
  1350. Tandy 3000.]
  1351.  
  1352. ------------------------------
  1353.  
  1354. Date: Thu, 7 Jul 88 13:19:47 CDT
  1355. From: boerner@emx.utexas.edu (Brendan B. Boerner)
  1356. Subject: MS-Kermit Tek-4010 Emulation Colors
  1357. Keywords: MS-DOS Kermit 2.31, Terminal Emulation, Color Graphics
  1358.  
  1359. Hello, Does anyone know how to use the different modes with the Tek-4010
  1360. emulation in Kermit 2.30?  I saw that depending on memory on the EGA
  1361. adpater, one can choose between more resolution/less color vs. less
  1362. resolution/more color.  Currently, when I use it, it goes into B/W mode.
  1363. How do I change this?
  1364.  
  1365. Thank you.
  1366.  
  1367. Brendan
  1368.  
  1369. [Ed. - Normally, you only get two colors when in Tek mode.  These are
  1370. the foreground and background colors you chose in your SET TERMINAL COLOR
  1371. command.  As of 2.31, if you have host-resident software that can include the
  1372. ANSI color-changing commands in Tek output, then you can have lots of colors,
  1373. but the exact number depends upon which display adapter and monitor you have.]
  1374.  
  1375. [From jrd - I suspect you are using CGA mode, which has only black and white
  1376. graphics. If Kermit senses that you are using an EGA board then it defaults to
  1377. 16 color 640 by 350 if the board says the monitor type is suitable. If you
  1378. have an ordinary CGA monitor (not an EGA monitor or better) then b/w is the
  1379. best you can do.  Also, check the switches on the display board.]
  1380.  
  1381. ------------------------------
  1382.  
  1383. Date: Sat, 9 Jul 88 18:01:10 pdt
  1384. From: steinmetz!unioncs!boehmr@uunet.UU.NET (Richard Boehm)
  1385. Subject: Bug in Kermit-MS on Tandy 3000HD?
  1386.  
  1387. Page 24                                                INFO-KERMIT DIGEST V8 #3
  1388.  
  1389. Keywords: MS-DOS Kermit 2.31, Tandy 3000HD
  1390.  
  1391. I would like to report what I believe to be a bug in Kermit-MS 2.31-test5 and
  1392. 2.31-test6.  When starting the program, exiting the program, or executing a
  1393. "take" file, DOS presents the message:
  1394.  
  1395.      Unexpected HW interrupt 0C at 0070:0455.
  1396.  
  1397. I have used all versions of Kermit since version 2.29b, and this behaviour
  1398. manifested itself starting with version 2.31-test5 (I am currently using
  1399. 2.31-test4).
  1400.  
  1401. I have just downloaded Kermit-MS 2.31 (the official release).  This version
  1402. also has the bug, but the bug doesn't appear when exiting the program.
  1403.  
  1404. My machine is a Tandy 3000HD AT clone.  It uses the Phoenix BIOS, is
  1405. an 80286 machine, and runs MS-DOS 3.2.
  1406.  
  1407.  ..steinmetz!unioncs!boehmr                      Richard E. Boehm, Jr.
  1408.    boehmr@unioncs.UUCP
  1409.    e_boehmr@union.BITNET
  1410.  
  1411. [Ed. - This does not happen on IBM systems, or any of various other IBM
  1412. compatibles we have tested, so it is apparently specific to the Tandy
  1413. model.  I assume that the only problem is the appearance of the error message,
  1414. and the operation of the program is unaffected?  Anyone with Tandy 3000
  1415. expertise is encouraged to track this one down!]
  1416.  
  1417. [From jrd - I think we found the root cause of this item, and essentially the
  1418. same problem when Kermit is run after running Smartcom II or PC-Talk III.
  1419. MSVIBM.BOO, dated 17 July 1988, is issued to cure it.
  1420.  
  1421. Here is the technical description: When Kermit resets the serial port
  1422. (relinquishes ownership) it tells the UART to disconnect its chip controlling
  1423. the interrupt (IRQ) line on the PC bus.  That disconnection can result in a
  1424. stray interrupt which appears to Kermit as not related to the serial port.
  1425. In such a case Kermit passes the interrupt to the previous owner which is
  1426. usually the ROM Bios.  The Phoenix version 1.00 ROM Bios in the Tandy 3000
  1427. apparently cannot handle a stray interrupt and issues the message above, but
  1428. most machines absorb it without comment.  In the case of Smartcom II and
  1429. PC-Talk III, both programs redirect the serial port interrupt to themselves
  1430. but do not return it when the program exits!  Thus, they remain the previous
  1431. owner; but their memory is used by another program, or worse.  This usually
  1432. crashes the system later on when something (noise?) arrives at the serial
  1433. port.  The 17 July version of IBM PC MS Kermit absorbs the stray interrupt
  1434. when resetting the serial port.  While this keeps Kermit operational, the
  1435. Smartcom II and PC-Talk III situations remain dangerous to the system after
  1436. they exit; rebooting is suggested.]
  1437.  
  1438. ------------------------------
  1439.  
  1440. Date: Sun, 10 Jul 88 17:24:23 EDT
  1441. From: elsie!ado@ncifcrf.gov (Arthur David Olson)
  1442. Subject: Kermit 2.31
  1443. Keywords: MS-DOS Kermit 2.31
  1444.  
  1445.  
  1446. INFO-KERMIT DIGEST V8 #3                                                Page 25
  1447.  
  1448. We've picked up Kermit 2.31; thanks as usual to all involved for a job well
  1449. done.
  1450.  
  1451. One thing that seems to be missing from the various "change lists" (though
  1452. it's described in the mskerm.doc opus) is the addition between version 2.30
  1453. and version 2.31 of "-" as a line continuation file in "take" and
  1454. "initialization" files.  Might be worth adding a note, since I've seen a fair
  1455. number of initialization files with
  1456.  
  1457.     set key \330 -
  1458.  
  1459. commands in them (for people who don't need different terminal emulations,
  1460. and who want to get a minus when they tap the key way over there in the corner
  1461. on recent IBM-style keyboards.)  To paraphrase mskerm.doc, change the above to
  1462.  
  1463.     set key \330 \45
  1464.  
  1465.     ado@ncifcrf.gov            ADO is a trademark of Ampex.
  1466.  
  1467. [Ed. - Thanks for the hint!  It's been added to MSKERM.BWR and MSR231.UPD.]
  1468.  
  1469. [From jrd - Usually one can add a trailing space to the hyphen since the
  1470. space is trimmed off before the string is interpreted. \45 is easier to see
  1471. when editing.]
  1472.  
  1473. ------------------------------
  1474.  
  1475. Date: Tue, 12 Jul 88 09:34:12 edt
  1476. From: rob@asrcmv.albany.edu (Robert D. Seals)
  1477. Subject: Did AT&T 6300 Graphics Get Canned in 2.31?
  1478. Keywords: MS-DOS Kermit 2.31, AT&T 6300, Tektronix Emulation
  1479.  
  1480. One of the great things about 2.29[bc] and 2.30 was the autodetection
  1481. and support for AT&T 6300 graphics - which doesn't seem to be
  1482. present in 2.31 (OK, I admit, I haven't reread the documentation...)
  1483.  
  1484. rob
  1485.  
  1486. [Ed. - This is news to us.  We don't have an AT&T 6300 to try it on, but
  1487. the DEC VAXmate has the same kind of adapter, and it is "autodetected" with
  1488. no problem.  But 2.31 contains a new feature, SET TERMINAL GRAPHICS <name>,
  1489. so that you can specify the actual adapter type in case Kermit guesses
  1490. wrong.  On an AT&T 6300, you can include SET TERMINAL GRAPHICS ATT in your
  1491. MSKERMIT.INI file.]
  1492.  
  1493. ------------------------------
  1494.  
  1495. End of Info-Kermit Digest
  1496.  
  1497. Page 26                                                INFO-KERMIT DIGEST V8 #4
  1498.  
  1499. Info-Kermit Digest         Wed, 3 Aug 1988       Volume 8 : Number 4
  1500.  
  1501. Departments:
  1502.  
  1503.   ANNOUNCEMENTS -
  1504.         CU20B IP Host Address to Change
  1505.         Okstate Dialup Number to Change
  1506.         Kermits Needed
  1507.         Announcing C-Kermit for OS/2
  1508.  
  1509.   MS-DOS KERMIT -
  1510.         Automatic Switching Between Text and Graphics in MS-Kermit
  1511.         MS-Kermit 2.31-test6
  1512.         MS-DOS Kermit 2.30
  1513.         MS-Kermit 2.31-test6
  1514.         IBMPC Kermit 2.31 UB problem
  1515.         MS-Kermit 2.31 -- SET PORT UB-Net1
  1516.         MS-Kermit 2.31 Bug (Feature?)
  1517.         MS-Kermit v2.31 Scripts
  1518.  
  1519. Send digest submissions to Info-Kermit@CU20B, requests for addition to or
  1520. deletion from the Info-Kermit subscriber list to Info-Kermit-Request@CU20B.
  1521.  
  1522. Kermit files may be obtained over networks and by mail order.  On the
  1523. Internetwork, use FTP to log in to host CU20B, CU20B.COLUMBIA.EDU, or
  1524. CU20B.CC.COLUMBIA.EDU (a DECSYSTEM-20), as user ANONYMOUS, using any password,
  1525. and GET the desired files from logical device KER:.  You can also get Kermit
  1526. files over BITNET/EARN; to get started send a message with text HELP to
  1527. KERMSRV, the Kermit file server, at host CUVMA.  For detailed instructions,
  1528. read the file KER:AANETW.HLP (AANETW HLP on KERMSRV).  To order by mail,
  1529. request a complete list of Kermit versions and an order form from Kermit
  1530. Distribution, Columbia University Center for Computing Activities, 612 West
  1531. 115th Street, New York, NY 10025 USA.
  1532.  
  1533. ----------------------------------------------------------------------
  1534.  
  1535.  
  1536. Date: Mon, 1 Aug 88 12:00:00 EDT
  1537. From: Christine M Gianone <SY.CHRISTINE@CU20B.CC.COLUMBIA.EDU>
  1538. Subject: CU20B IP Host Address to Change
  1539. Keywords: CU20B, FTP
  1540.  
  1541. As of August 15th, 1988, Columbia University Computer Center IP host
  1542. addresses will change by replacing the 32 with a 40 in the third octet of
  1543. the IP address.  In particular, the host number of CU20B will change from
  1544. 128.59.32.128 to 128.59.40.128.  If you have trouble FTP'ing Kermit files
  1545. from CU20B on or after August 15th, you can specify the new host number until
  1546. your site's host tables are updated to reflect the change.
  1547.  
  1548. ------------------------------
  1549.  
  1550. Date: Mon, 01 Aug 88 15:46:23 -0500
  1551. From: Mark Vasoll <vasoll%a.cs.okstate.edu@RELAY.CS.NET>
  1552. Subject: Okstate Dialup Number to Change
  1553. Keywords: Okstate
  1554.  
  1555.  
  1556. INFO-KERMIT DIGEST V8 #4                                                Page 27
  1557.  
  1558. On August 6, 1988 the phone number for the UUCP and Kermit servers at
  1559. Oklahoma State University will change:
  1560.  
  1561.     Old number:   (405) 624-6953
  1562.     New number:   (405) 744-6953
  1563.  
  1564. A "that number has been changed" recording has been promised, but I
  1565. don't hold a lot of hope for it.
  1566.  
  1567. Mark Vasoll
  1568. Computing and Information Sciences   Email:  vasoll@a.cs.okstate.edu
  1569. Oklahoma State University
  1570. Stillwater, Oklahoma
  1571.  
  1572. ------------------------------
  1573.  
  1574. Date: Fri 29 Jul 88 17:27:32-EDT
  1575. From: Frank da Cruz <SY.FDC@CU20B.CC.COLUMBIA.EDU>
  1576. Subject: Kermits Needed
  1577. Keywords: Future Kermits
  1578.  
  1579. Now that Kermit is about seven years old, it has spread to cover nearly every
  1580. major computer and operating system, and many minor ones -- well over 300 in
  1581. all.  But there are still some notable systems for which no Kermit programs
  1582. exist.  Among them:
  1583.  
  1584.  - The IBM System/34, System/36, System/38, AS/400.  Some of these machines
  1585.    lack asynchronous communication interfaces, so Kermit file transfer will
  1586.    be possible only if the commonly-used protocol converters can be put into
  1587.    transparent mode, as the IBM 370-series mainframe Kermits do for their
  1588.    3270 protocol converters.
  1589.  
  1590.  - Support in IBM mainframe Portable Kermit-370 for DOS/VSE and CICS.
  1591.  
  1592.  - Wang VS and other minicomputer systems.
  1593.  
  1594. If you know of anyone working on these projects, or are interested in doing
  1595. it yourself (for fame but not fortune), please let us know!
  1596.  
  1597. ------------------------------
  1598.  
  1599. Date: Fri, 29 Jul 88 14:54:36 -0100
  1600. From: PD Software <syspds%central1.lancaster.ac.uk@NSS.Cs.Ucl.AC.UK>
  1601. Subject: Announcing C-Kermit for OS/2
  1602. Keywords: C-Kermit, OS/2 Kermit
  1603.  
  1604.                        Announcing C-Kermit for OS/2
  1605.  
  1606. A version of C-Kermit for OS/2 is now available for beta testing.  Any bugs,
  1607. suggestions or complaints should be sent to C.Adie@uk.ac.edinburgh.  This
  1608. first version is 1.0a - the 'a' indicates it is a pre-release version.
  1609.  
  1610. I've used the letter 'O' as the system-specific identifier for OS/2, so
  1611. the release files are named as follows:
  1612.       CKOKER.EXE      The executable program
  1613.       CKOKER.DOC      Documentation
  1614.  
  1615. Page 28                                                INFO-KERMIT DIGEST V8 #4
  1616.  
  1617. The sources for the program and the Scribe source for the documentation
  1618. will be made available when beta testing is over (ie when bug reports
  1619. slow to a trickle?).
  1620.  
  1621. C-Kermit for OS/2 runs on OS/2 Standard edition 1.0 (or on 1.1 in a
  1622. Presentation Manager window, although this is untested).  It makes use of
  1623. the COM.SYS (or perhaps COM01.SYS) device driver, which must be loaded in
  1624. the CONFIG.SYS file.  There are believed to be no major bugs, but the
  1625. following minor problems exist:
  1626.  
  1627. *  Stop bits: There is currently no  mechanism for altering the number of  stop
  1628.    bits, which is fixed at 1.
  1629. *  Server breakout:  There is  no way  of stopping  server operation  from  the
  1630.    keyboard, short of Control-Break.
  1631. *  Dial: The DIAL command has not  been tested.  It is impossible to  terminate
  1632.    the dialling process short of pressing Control-Break.
  1633. *  Multiple copies: Multiple  copies of Kermit  accessing multiple comms  ports
  1634.    simultaneously has not been tested.
  1635. *  Terminal emulation: No  particular terminal is  emulated, and therefore  the
  1636.    keyboard has only  the natural key-to-ascii-code  mapping.  This means  that
  1637.    (for instance)  the Del  key does  not produce  code 127,  but a  code of  0
  1638.    followed by 224.  The only exception  to this "natural" mapping is that  the
  1639.    backspace key is  explicitly mapped into  127.  It is  hoped to implement  a
  1640.    VT100 emulator and a keyboard remapper  to circumvent these problems.   This
  1641.    may also improve the  speed of terminal writing,  which is slow compared  to
  1642.    MS-Kermit.
  1643. *  Multiple send from command line:  At present, a wildcard file  specification
  1644.    cannot be given as the argument to the '-s' command line option.  Thus, only
  1645.    a single file can be sent using a command line.
  1646. *  Flow control from keyboard: In connect mode, if a Control-S is typed at  the
  1647.    keyboard to pause a long file listing, it may or may not work, depending  on
  1648.    the state of the device driver's flow control flags.
  1649.  
  1650. There are also some further minor bugs or areas for improvement which are
  1651. detailed at the end of the documentation.
  1652.  
  1653. Chris Adie
  1654. 26/07/88
  1655.  
  1656. [Ed. - Many thanks!  The files are in KER:CKOKER.*.  We normally do not
  1657. release programs without source code, but we make an exception in this case
  1658. because of the high level of interest in OS/2.  Test at your own risk!]
  1659.  
  1660. ------------------------------
  1661.  
  1662. Date: Tue,28 Jun 88 16:24:40 BST
  1663. From: P.WADE@uk.ac.hull.cc.vme
  1664. Subject: Automatic Switching Between Text and Graphics in MS-Kermit
  1665. Keywords: MS-DOS Kermit 2.31
  1666.  
  1667. Someone here is using a package which uses DEC VT100 codes during its text
  1668. output and Tektronix 4010 codes for its graphics output.  The package expects
  1669. the terminal to automatically switch between text and graphics modes at the
  1670. appropriate points.  If MS-Kermit is used with the package, automatic switching
  1671. from text to graphics takes place in some cases (for example when the package
  1672. sends ESCAPE Control-L) but not in others (for example when the package sends a
  1673.  
  1674. INFO-KERMIT DIGEST V8 #4                                                Page 29
  1675.  
  1676. Tektronix drawing command).  In comparison, UNITERM on an Atari ST always
  1677. switches when necessary.
  1678.  
  1679. Should the codes which cause MS-Kermit to automatically switch from text to
  1680. graphics mode be expended to include the Tektronix FS, GS, RS, and US commands?
  1681. This automatic switching can, of course, be disabled if necessary via the
  1682. Kermit DISABLE TEK command.
  1683.  
  1684. Phil Wade,
  1685. Hull University Computer Centre.
  1686.  
  1687. [From jrd - While automatic switching for all Tek commands (there are more
  1688. than the FS, GS, RS, US set, clearly) can be convenient for isolated packages
  1689. it both adds a lot of code to MS Kermit and exposes users to many unexpected
  1690. switches when files are displayed with embedded control codes. Overall, I
  1691. think we are better off with the just the current two automatic entry points.]
  1692.  
  1693. ------------------------------
  1694.  
  1695. Date: Fri, 01 Jul 88 11:36:28 BST
  1696. From: CPA1@UK.AC.CAM.PHX
  1697. Subject: MS-Kermit 2.31-test6
  1698. Keywords: MS-DOS Kermit 2.31
  1699.  
  1700. I've come across a number of peculiarities while using 2.31-6 from batch
  1701. files, or MAKE files, with a PS2/60.  When the command
  1702.  
  1703.     kermit send dirname\filename.C dirname\filename.C
  1704.  
  1705. is entered (having set up a kermit on an EPSON AX in server mode) the
  1706. filename used at the other end is
  1707.  
  1708.     dirname\filename.CX  (Note the added X in the extension)
  1709.  
  1710. I assume that this is a bug -it's certainly annoying!  The addition of the X
  1711. does not occur with extension names of 2 or more characters.  If the send
  1712. command is not the last on the line then the X is not added either, i.e.
  1713.  
  1714.     kermit send dirname\filename.C dirname\filename.C, quit
  1715.  
  1716. saves
  1717.  
  1718.     dirname\filename.C
  1719.  
  1720. at the remote end.  The "sending as" within kermit works fine.
  1721.  
  1722. Another quirk is when Control-C is used while a file transfer is in
  1723. progress.  This causes the server kermit to hang up waiting for more
  1724. packets.  The way I've found around this is to ask it for a "rem dir" a few
  1725. times and Control-C out it, producing the "host not receiving" (not
  1726. verbatim!) error message.  Eventually the server gets the message and gives
  1727. a directory listing.
  1728.  
  1729. Yours,
  1730. Chris.
  1731.  
  1732.  
  1733. Page 30                                                INFO-KERMIT DIGEST V8 #4
  1734.  
  1735. [From jrd - Chris, you need to reread the MS Kermit User's Manual. First,
  1736. the SEND command can take two arguments: local filname and optionally the
  1737. name to use in the outgoing packet. It does not take a list of files to be
  1738. sent; use a list of SENDs. Second, Control-C is a non-protocol exit and does
  1739. not notify the other side. Hence the host waits for more packets in the file
  1740. exchange. Use Control-Z instead, as indicated on the file transfer screen
  1741. status line.]
  1742.  
  1743. ------------------------------
  1744.  
  1745. Date: Thu, 7 Jul 88 20:02:31 bst
  1746. From: Leila Burrell-Davis <leilabd@uk.ac.sussex.cvaxa>
  1747. Subject: MS-DOS Kermit 2.30
  1748. Keywords: MS-DOS Kermit 2.30
  1749.  
  1750. MS-Kermit is normally pretty good about renaming files that don't conform to
  1751. MS-DOS filename conventions, but it is distinctly upset by a file with a
  1752. colon in its name, e.g.  episode:1
  1753.  
  1754. I found this rather annoying when I left it overnight downloading a lot of
  1755. files from a un*x machine, only to discover the next morning that it had
  1756. given up as soon as it got to the filename with the colon in.
  1757.  
  1758. You may like to pass on details of this `feature' to the program's author
  1759. (:-).
  1760.  
  1761. Leila Burrell-Davis, Computing Service, University of Sussex, Brighton, UK
  1762. Tel: +44 273 678390   Fax: +44 273 678335  JANET: leilabd@uk.ac.sussex.cvaxa
  1763. ARPA: leilabd%cvaxa.sussex.ac.uk@cunyvm.cuny.edu
  1764. BITNET: leilabd@cvaxa.sussex.ac.uk      UUCP: leilabd@cvaxa.uucp
  1765.  
  1766. [From jrd - DOS uses colon as a device name terminator/separator. Try this at
  1767. DOS level-
  1768.  
  1769.         C> copy oldfile new:name.ext
  1770.  
  1771. DOS will tell you that there is an illegal destination (no device "new:").
  1772. DOS also uses multiletter device names which may optionally end in a colon.
  1773. Kermit cannot determine whether or not you mean use a specific device and
  1774. thus does not change the colon to another legal letter.]
  1775.  
  1776. ------------------------------
  1777.  
  1778. Date: Thu, 07 Jul 88 15:36:00 BST
  1779. From: Mr. A. O. V. Blanc <YMUMAL@uk.ac.umist.central-services.prime-a>
  1780. Subject: MS-Kermit 2.31-test6
  1781. Keywords: MS-DOS Kermit 2.31
  1782.  
  1783. We have obtained a copy of MS-Kermit 2.31-test6  for  testing.   I  have
  1784. been running it on several PC models (IBM, Opus, and Comcen) and testing
  1785. it against our Kermit-CMS version  4.0,  the  IBM  generic  Kermit  with
  1786. updates up to SC88092.  During the tests -- which went very well, on the
  1787. whole -- the following difficulties appeared:
  1788.  
  1789. (1)  The reported size of the packets received never gets larger than 4,
  1790.      which is clearly incorrect.
  1791.  
  1792. INFO-KERMIT DIGEST V8 #4                                                Page 31
  1793.  
  1794.  
  1795. (2)  The file size used to calculate the percentage transferred is quite
  1796.      far off, which some files terminating when only 25%  had  allegedly
  1797.      been  transferred.   Different  ways  of measuring file size may be
  1798.      responsible, but users would surely feel happier if the  percentage
  1799.      were changed to 100% when the transfer terminates successfully.
  1800.  
  1801. (3)  The  two  Kermits  cooperated  nicely when connected through a 7171
  1802.      comms processor, but not when using a CAMTEC PAD.  In this case the
  1803.      connection  works normally until the Kermits begin trading packets.
  1804.      Then the IBM apparently insists on a fifteen or twenty second pause
  1805.      before  accepting  each  line of input.  This results in a transfer
  1806.      rate of 3 to 4 packets per minute, which is intolerably  slow,  and
  1807.      the  input  problem remains even after Kermit ceases to execute.  I
  1808.      know this sounds like a CMS problem, especially since it  seems  to
  1809.      occur  under similar circumstances with MS-Kermit version 2.30.  It
  1810.      does not happen, however, when I use version 2.29c!
  1811.  
  1812. (4)  I am not surprised that MS-Kermit receives only the  date  and  not
  1813.      the  time  of a file's creation, since the warning of this appeared
  1814.      in the MSTIBM.BWR file.  When files are transferred from the  micro
  1815.      to CMS, however, both date and time seem to be ignored.
  1816.  
  1817. Thanks to Joe Doupnik for his extensive work improving MS-Kermit and its
  1818. script facility.
  1819.  
  1820. [From jrd - some of the 2.31test series had a file length arithmetic problem
  1821. which lost the high order part of the file size figure sent by the host.
  1822. Fixed some time ago.
  1823.  
  1824. CMS Kermit apparently sends the date of a file but not the time. In that case
  1825. MS Kermit records the time as 00:00:00, which is legal, but DOS then shows
  1826. this as blanks on the screen. The release version of MS Kermit 2.31 adds two
  1827. seconds to let DOS display some numbers for the time.
  1828.  
  1829. The long pause does appear to be either a CMS or a front end processor
  1830. problem; see if using regular length packets helps these fellows. You might
  1831. say to MS Kermit  SET SEND PAUSE <millisec>  because some IBM systems do not
  1832. become ready for reception until long after turning around the line. If this
  1833. were the case the mainframe would miss the start of a packet and would have
  1834. to wait for a timeout and a repeat; the timeout is normally 13 seconds to
  1835. receive an acknowledgement.]
  1836.  
  1837. ------------------------------
  1838.  
  1839. Date: Fri, 22 Jul 88 18:00:06 PDT
  1840. From: doug%Tybalt.Caltech.Edu%CITIAGO.BITNET@CUVMB.CC.COLUMBIA.EDU
  1841. Subject: IBMPC Kermit 2.31 UB problem
  1842. Keywords: MS-DOS Kermit 2.31, Ungermann-Bass
  1843.  
  1844.     I recently downloaded IBMPC Kermit v2.31.  An excellent package, as
  1845. usual.  The UB NetCI interface looked particularly interesting, since we
  1846. currently require a resident BIOS rerouter to use the CI from Kermit.
  1847.  
  1848.     Unfortunately, as soon as characters were received from the CI, the PC
  1849. hung.  After reducing other software to its bare essentials, I began to
  1850.  
  1851. Page 32                                                INFO-KERMIT DIGEST V8 #4
  1852.  
  1853. suspect Kermit.  After downloading MSXIBM.ASM and DEBUGing through
  1854. KERMIT.EXE, a probable cause appeared.
  1855.  
  1856.     The IRET in RPOST would send the CS off to never-never land, causing the
  1857. machine to hang.  This was because the "call rpost" in UBRECV assumed a near
  1858. (intra-segment) call, and only pushed the IP+2 on the stack, and not the CS.
  1859. As a solution, I recommend making UBRECV and all other interrupt routines
  1860. type PROC FAR, since that more accurately describes what they are. (IRET =
  1861. far RET + POPF).  Then UBRECV would assume a far call.
  1862.  
  1863.     I certainly appreciate all the work that you are putting into Kermit.
  1864. It fills an essential gap here at Caltech.
  1865.  
  1866.     There are a couple more minor problems with the UB NetCI interface in
  1867. IBMPC Kermit v2.31.  I downloaded the source this evening and MSXIBM.ASM
  1868. required the following changes before the UB CI would work:
  1869.  
  1870.     1. Change all interrupt procedures to PROC FAR (mentioned earlier)
  1871.  
  1872.     2. Add lines between outch3: and outch3a: as follows -
  1873.  
  1874. outch3: cmp     clone,0                 ; real uart?
  1875.         je      outch3a                 ; e = yes
  1876.         cmp     clone,'N'               ; network?
  1877.         je      outch8                  ; e = yes, using netbios
  1878. add-->  cmp     clone,'U'               ; UB?
  1879. add-->  je      outch8                  ; e = yes, using netbios setup
  1880.         jmp     outch6                  ; default for others ('B' clones)
  1881. outch3a:push    cx                      ; Save registers
  1882.  
  1883.     3. In ubsend, xmtbuf does not contain our info- xmtbufx does; so change
  1884.  
  1885.         jcxz    ubsend1                 ; dont send zero chars           [ohl]
  1886.         mov     bx, offset xmtbuf       ; buffer address in es:bx        [ohl]
  1887.         mov     ax, datas
  1888.         mov     es, ax
  1889. to
  1890.         jcxz    ubsend1                 ; dont send zero chars           [ohl]
  1891. changed-->mov     bx, offset xmtbufx    ; buffer address in es:bx        [ohl]
  1892.         mov     ax, datas
  1893.         mov     es, ax
  1894.  
  1895.     These three minor changes appear to fix the interface.  Thank you again
  1896. for providing the program and source.
  1897.  
  1898. -Doug Schafer
  1899. doug@tybalt.caltech.edu
  1900.  
  1901. [Ed. - Thanks for the fixes!  They have been passed along to Joe Doupnik,
  1902. and there should be a maintenance release of 2.31 soon, which fixes this and
  1903. a couple of other problems.]
  1904.  
  1905. ------------------------------
  1906.  
  1907. Date: Fri, 15 Jul 88 15:42:24 EDT
  1908. From: cucard!dasys1!eravin@columbia.edu (Ed Ravin)
  1909.  
  1910. INFO-KERMIT DIGEST V8 #4                                                Page 33
  1911.  
  1912. Subject: MS-Kermit 2.31 -- SET PORT UB-Net1
  1913. Keywords: MS-DOS Kermit 2.31, Ungermann-Bass
  1914.  
  1915. I just tried using MS-Kermit 2.31 on our Ungerman-Bass network.  The
  1916. MS-Kermit docs say that when I issue the "CONNECT" command after "SET PORT
  1917. UB-Net1", I should see the NET-CI prompt and be able to use the usual NET-CI
  1918. commands.  Alas, I saw nothing.  Not only that, but when I tried to exit
  1919. Kermit, my PC went into a "waiting state", where the PC hangs up, but a
  1920. chirp is issued every few seconds to let you know that it is still trying to
  1921. access something or someone on the network that can't be found.
  1922.  
  1923. I can still use U/B's NET-CI driver (NETCICOM.DRV) without any problem: not
  1924. only that, but if I use NETCICOM to redirect a com port (such as COM3), and
  1925. then tell MS-Kermit to use port BIOS3, everything works fine and I can
  1926. access the NET-CI interface without difficulty.  Any idea why I can't use
  1927. the "SET PORT UB-Net1" command?
  1928.  
  1929.             Ed Ravin
  1930.  
  1931.             eravin@dasys1.UUCP
  1932.         or
  1933.             eravin@trintex.UUCP
  1934.  
  1935. [Ed. - See previous message.  The problem will be corrected shortly.]
  1936.  
  1937. ------------------------------
  1938.  
  1939. Date: Tue, 19 Jul 88 00:31 PDT
  1940. From: CARL FUSSELL <CARL%SCU.BITNET@CUVMB.CC.COLUMBIA.EDU>
  1941. Subject: MS-Kermit 2.31 Bug (Feature?)
  1942. Keywords: MS-DOS Kermit 2.31
  1943.  
  1944. I just downloaded the latest version of MS Kermit (2.31).  There seems to be
  1945. some unexplainable behavior when running a script.  We run a typical script
  1946. to login to a VAX/VMS system.  The script worked fine under version
  1947. 2.31-test5 but caused a strange problem when used under the final 2.31
  1948. distribution.  The problem that appears is that TAB characters aren't
  1949. displayed correctly.  Instead of tabbing to the next tab stop, the display
  1950. is tabbed to column 80 with the remaining text wrapping to the next line.
  1951. This does not happen using the same script with V2.31-test5.
  1952.  
  1953. The script line that causes the problem is the initial CLEAR statement.
  1954. Leave it in and the problem described above appears.... remove it (or
  1955. comment it out) and it works.  Fortunately for us, everything works ok with
  1956. the line removed.
  1957.  
  1958. Anyway, just tho't I'd pass this along...  (the script is attached if
  1959. interested.)
  1960.  
  1961. Carl Fussell
  1962. Santa Clara Univ
  1963. CARL@SCU.BITNET
  1964.  
  1965. [Ed. - We have had one other similar complaint -- i.e. that the CLEAR
  1966. command, which is supposed to clear only the serial port input buffer,
  1967. also clears the tab stops.  But no matter how hard we try, we can't
  1968.  
  1969. Page 34                                                INFO-KERMIT DIGEST V8 #4
  1970.  
  1971. reproduce it.  We'll try again to pin it down.  Meanwhile, the workaround
  1972. is to put a "SET TERM TAB" command after the CLEAR command.]
  1973.  
  1974. ------------------------------
  1975.  
  1976. Date: Mon, 25 Jul 88 13:01:26 -0500
  1977. From: Mark Bergman <bergman@ROCKEFELLER.ARPA>
  1978. Subject: MS-Kermit v2.31 Scripts
  1979. Keywords: MS-DOS Kermit 2.31, Scripts
  1980.  
  1981. Many thanks for all the improvements in MS-Kermit v2.31!  Its already making
  1982. lives easier here by providing a much-requested auto-dialing capability.  I
  1983. do have a question about using the INPUT command as part of a take file
  1984. (actually mskermit.ini) to allow the user to choose what number to dial.  I
  1985. want to ask the user whether to connect to the Rockefeller University system
  1986. or not (expanding to other choices in the future).
  1987.  
  1988. I was trying statements in the form:
  1989.  
  1990.  echo Connect to the Rockefeller system? (yes or no)\13
  1991.  input 45 yes @CON
  1992.  if success goto CONTINUE
  1993.  
  1994. which fails (according to an "if success" test) when the first key is
  1995. pressed, regardless of the input.  No <RETURN> is needed, and the input is
  1996. not echoed.
  1997.  
  1998. I also tried statements in the form:
  1999.  
  2000.  echo Connect to the Rockefeller system? (yes or no)\13
  2001.  input 45 @CON yes
  2002.  if success goto CONTINUE
  2003.  
  2004. This succeeds (according to an "if success" test), in every case.  The input
  2005. is not checked, but it always succeeds.  The input text also must be
  2006. followed by two <RETURNS> before it is recognized, and the characters are
  2007. not echoed while this is run.
  2008.  
  2009. I wanted to know if you had an example of a script that read and tested
  2010. input from the keyboard.  I also wanted to know if there was any way to
  2011. supress the "?Timeout" error message.  The script provides it's own error
  2012. message.
  2013.  
  2014.                                         Mark Bergman
  2015.                     Rockefeller University
  2016.                     Computing Services
  2017.                     Box 247
  2018.                     1230 York Avenue
  2019.                     New York, New York 10021
  2020.  
  2021.                                         212-570-7510
  2022.  
  2023.   BEST PATH==>     bergman@rocky5.rockefeller.edu
  2024.             bergman@rockvax.bitnet
  2025.             bergman@rocky2.uucp
  2026.  
  2027.  
  2028. INFO-KERMIT DIGEST V8 #4                                                Page 35
  2029.  
  2030. [Ed. - Version 2.31 does not include a mechanism to do what you want.
  2031. INPUT only looks at the communication port, not the keyboard.  A future
  2032. release will include this feature.]
  2033.  
  2034. ------------------------------
  2035.  
  2036. End of Info-Kermit Digest
  2037.  
  2038. Page 36                                                INFO-KERMIT DIGEST V8 #5
  2039.  
  2040. Info-Kermit Digest         Wed, 10 Aug 1988       Volume 8 : Number 5
  2041.  
  2042. Departments:
  2043.  
  2044.   ANNOUNCEMENTS -
  2045.     Columbia University IP Host Numbers to Change
  2046.         Kermit Files Available via UUCP at Oklahoma State U
  2047.         Announcing Apple II Kermit Version 3.84
  2048.  
  2049.   MS-DOS KERMIT -
  2050.     Typo in MSKERM DOC
  2051.         Support for AT&T 6300 Graphics Card
  2052.         Kermit-MS 2.31 and AT&T Graphics
  2053.         MS-Kermit and a Local Printer
  2054.         MS-Kermit 2.31 (?More parameters are needed) Message
  2055.         MS-Kermit 2.31 Internal Modem Problems
  2056.         RE: MS-Kermit 2.31 Bug (Feature?)
  2057.         MS-Kermit 2.31 IBM 17 July 1988 Eats nulls
  2058.         VT200 Support in MS-Kermit?
  2059.         MS-Kermit with Async Servers?
  2060.         MS-DOS Kermit 2.31 and Local Printer
  2061.         MS-Kermit 2.3x Bug?  2.31 And Z-100's ZPC
  2062.         Z-248 COM3 and Kermit
  2063.         MSUCHK/SCANCHECK and the Rainbow
  2064.  
  2065.   MACINTOSH KERMIT -
  2066.         MacKermit Checksums
  2067.         MacKermit and Binary File Transfer
  2068.         Transferring application via MacKermit 0.9(40)
  2069.         MacKermit Escape Sequence Keys
  2070.  
  2071. Send digest submissions to Info-Kermit@CU20B, requests for addition to or
  2072. deletion from the Info-Kermit subscriber list to Info-Kermit-Request@CU20B.
  2073.  
  2074. Kermit files may be obtained over networks and by mail order.  On the
  2075. Internetwork, use FTP to log in to host CU20B, CU20B.COLUMBIA.EDU, or
  2076. CU20B.CC.COLUMBIA.EDU (a DECSYSTEM-20), as user ANONYMOUS, using any password,
  2077. and GET the desired files from logical device KER:.  You can also get Kermit
  2078. files over BITNET/EARN; to get started send a message with text HELP to
  2079. KERMSRV, the Kermit file server, at host CUVMA.  For detailed instructions,
  2080. read the file KER:AANETW.HLP (AANETW HLP on KERMSRV).  To order by mail,
  2081. request a complete list of Kermit versions and an order form from Kermit
  2082. Distribution, Columbia University Center for Computing Activities, 612 West
  2083. 115th Street, New York, NY 10025 USA.
  2084.  
  2085. ----------------------------------------------------------------------
  2086.  
  2087. Date: Mon, 8 Aug 1988 8:08:08 EDT
  2088. From: Christine M. Gianone <cmg@cunixc.cc.columbia.edu>
  2089. Subject: Columbia University IP Host Numbers to Change
  2090. Keywords: IP Host Numbers, CU20B, Internet, FTP
  2091.  
  2092. Effective August 15th, 1988, Columbia University Computer Center Internet
  2093. hosts will be moved from subdomain 32 to subdomain 40.  In particular, the
  2094. IP host number for CU20B.CC.COLUMBIA.EDU (the host from which Kermit files
  2095. can be FTP'd over the Internet) will change from 128.59.32.128 to
  2096.  
  2097. INFO-KERMIT DIGEST V8 #5                                                Page 37
  2098.  
  2099. 128.59.40.128.  This change was posted to the NIC, but Internet sites that
  2100. have not updated their host tables, or which do not route through a domain
  2101. server, will not be able to reach CU20B by mail, FTP, or other Internet
  2102. protocol except by specifying the host number.
  2103.  
  2104. ------------------------------
  2105.  
  2106. Date: Mon, 8 Aug 1988 13:55:09 EDT
  2107. From: Christine M. Gianone <cmg@cunixc.cc.columbia.edu>
  2108. Subject: Kermit Files Available via UUCP at Oklahoma State U
  2109. Keywords: UUCP, Okstate
  2110.  
  2111. Just a reminder for those who might not know about this already: a complete,
  2112. though not necessarily totally up to date, collection of Kermit files is
  2113. available via dialup UUCP or Kermit protocol at Oklahoma State University,
  2114. courtesy of Mark Vasoll and the folks in the Department of Computing and
  2115. Information Sciences.
  2116.  
  2117. Full instructions (too long to list here) are in the file KER:AANOKS.HLP on
  2118. CU20B, or AANOKS HLP on KERMSRV at CUVMA.  If you're not on the Internet or
  2119. BITNET directly, you can send mail to KERMSRV asking it to send you this
  2120. file:
  2121.  
  2122.     To: KERMSRV@CUVMA.BITNET
  2123.     (through whatever UUCP/BITNET gateway you know about)
  2124.  
  2125.     Subject: (doesn't matter)
  2126.  
  2127.     Message text:
  2128.     mail aanoks.hlp
  2129.  
  2130. Thanks once again to Oklahoma State for providing this service.
  2131.  
  2132. ------------------------------
  2133.  
  2134. Date: Mon, 01 Aug 88 07:28:02 PDT
  2135. From: Ted Medin <MEDIN-T@SHARK.NOSC.MIL>
  2136. Subject: Announcing Apple II Kermit Version 3.84
  2137. Keywords: Apple II, ProDOS
  2138.  
  2139. Kermit-65 is truly a network project with many people sugesting improvements
  2140. and reporting bugs & problems. We wont name them all but here are a couple of
  2141. major contributors, Murray Sewall & Bob Holley. Again thanks to all.  Here
  2142. are the changes:
  2143.  
  2144. 1. Prodos text files now 8 bits wide.
  2145. 2. More vt100 fixes (will it ever stop!)
  2146. 3. Start of packet now variable
  2147. 4. Multiple protocols (Kermit and Xmodem)
  2148. 5. TAKE command
  2149. 6. DOS 3.3 more robust (almost bullet proof?)
  2150. 7. Lots of little fixes and options
  2151.  
  2152. [Ed. - Thanks, Ted!  The new files are in KER:APP*.*.]
  2153.  
  2154. ------------------------------
  2155.  
  2156. Page 38                                                INFO-KERMIT DIGEST V8 #5
  2157.  
  2158.  
  2159. Date: Sat, 30 Jul 88 14:06:40 EDT
  2160. From: Harold Pritchett <HAROLD@UGA.UGA.EDU>
  2161. Subject: Typo in MSKERM DOC
  2162. Keywords:  MS-DOS Kermit 2.31 Manual
  2163.  
  2164. Found the following in the current version of the MS-DOS KERMIT USERS GUIDE
  2165. from KERMSRV at CUVMA.
  2166.  
  2167. In Figure 1-3, the first line of the sample script reads:
  2168.  
  2169.      if defined % %to start            ; Make sure number specified.
  2170.  
  2171. It should read:
  2172.  
  2173.      if defined %\1%%to start            ; Make sure number specified.
  2174.  
  2175. Just thought you would like to correct it.  (I really don't read with that
  2176. sort of accuracy.  I cut the script out and tried to run it and it didn't
  2177. work...)
  2178.  
  2179. Harold
  2180.  
  2181. [Ed. - Hmmm...  Our copy of the manual reads "if defined %\1 goto start".
  2182. Don't know where "if defined % %to start" came from, but in any case, it
  2183. SHOULD read "if defined \%1 goto start" (\%1 instead of %\1).  There's also
  2184. a problem with the previous figure, 1-2.  Namely, the ermsg macro won't
  2185. print the entire error message, but only the first word, e.g.  "ermsg No
  2186. login prompt!" will only print "No".  Version 2.31 of Kermit doesn't have a
  2187. straightforward way to pass an arbitrary string of blank-separated words to
  2188. a macro, so the messages have to be made into single "words" like this:
  2189. "ermsg No_login_prompt!".  Sorry for the errors.]
  2190.  
  2191. ------------------------------
  2192.  
  2193. Date: 19 Jul 88 22:57:00 CDT
  2194. From: "DAVID GAY" <gay@chewi.che.wisc.edu>
  2195. Subject: Support for AT&T 6300 Graphics Card
  2196. Keywords: MS-DOS Kermit 2.31
  2197.  
  2198. Version 2.30 of MS-Kermit supports the AT&T 6300 graphics card for the
  2199. Tektronix emulation just fine, but version 2.31 does not. The setup shows
  2200. that the terminal graphics is set to AT&T 640x400, but the plots are done in
  2201. CGA 640x200 mode. I am using the MSVIBM.EXE which was FTPed directly from
  2202. Columbia on July 8, 1988.
  2203.  
  2204. and more......
  2205.  
  2206. After discovering that MSKERMIT 2.31 July 1, 1988 did not correctly support
  2207. the 640x400 graphics mode of my AT&T 6300 as v2.30 did, I FTPed waht looked
  2208. like an updated version of MSVIBM.EXE from COLUMBIA.  Unfortunately, the bug
  2209. is still present. Even though the SHOW TERMINAL lists ATT as the graphics
  2210. mode, everything is done in CGA 640x200 (ugh!)  mode. Please inform me when
  2211. this bug gets fixed so that i may retrieve a new copy.
  2212.  
  2213. Thanks,
  2214.  
  2215. INFO-KERMIT DIGEST V8 #5                                                Page 39
  2216.  
  2217.  
  2218. David Gay
  2219. gay@chewi.che.wisc.edu
  2220.  
  2221. [Ed. - We don't have a PC with an ATT graphics board here to test it on,
  2222. but we do have a VAXmate, which has an ATT compatible board.  It normally
  2223. runs in CGA mode, but if you SET TERMINAL GRAPHICS ATT, then it uses the
  2224. higher resolution graphics just fine.]
  2225.  
  2226. [From jrd - I wonder if we are encountering a Bios version number effect.
  2227. For AT&T machines Kermit looks for the word 'OLIVETTI' in the Bios starting
  2228. at locations FC00:0h or FC00:2014H.  It would be worthwhile using Debug to
  2229. probe your Bios.]
  2230.  
  2231. ------------------------------
  2232.  
  2233. Date: Tue, 26 Jul 88 16:53:05 EST
  2234. From: Anne Chetham-Strode <ACS%UVMVM.BITNET@CUVMB.CC.COLUMBIA.EDU>
  2235. Subject: Kermit-MS 2.31 and AT&T Graphics
  2236. Keywords: MS-DOS Kermit 2.31
  2237.  
  2238. I am testing Kermit-MS 2.31 on an AT&T PC6300.  The PC6300 supports CGA mode
  2239. as well as AT&T's "high resolution" 640 x 400 mode.  Kermit-MS 2.30 detected
  2240. and supported the high resolution mode automatically.  Version 2.31 does not
  2241. seem to support it.  "Set Term Graphic ATT" and "Set Term Graphic EGA" both
  2242. produce CGA mode.
  2243.  
  2244. Also "Set Term Graphic VGA" on a PS/2 Model 30 produces CGA mode.
  2245.  
  2246. If you intend to work on this, I am happy to help with testing.
  2247.  
  2248. Thank you for your attention.
  2249.  
  2250. [From jrd - see comments above on AT&T 6300's.  Selecting SET TERM GRAPH VGA
  2251. in turn selects video mode 17 (11H) for monochrome monitors or mode 18 (12H)
  2252. for color monitors, 640x480 by 2 or 16 colors, respectively, provided the
  2253. display adapter passes the EGA tests.  My brief PS/2 Model 30 specs show it
  2254. to support mode 17 (mono) but not 18 (color) so Kermit must have decided that
  2255. a color display was being used or that the adapter did not pass the EGA tests.
  2256. If it passed the EGA tests then 640x350 mode would have been employed.  There
  2257. is no Model 30 nearby so I can not easily check it.  Help on Model 30's would
  2258. be appreciated.]
  2259.  
  2260. ------------------------------
  2261.  
  2262. Date: Wed, 20 Jul 88 15:06:25 GMT
  2263. From: Dermot O'Beirne <DOBEIRNE%IRLEARN.BITNET@CUVMB.CC.COLUMBIA.EDU>
  2264. Subject: MS-Kermit and a Local Printer
  2265. Keywords: MS-DOS Kermit 2.31, Printer, 3COM EtherSeries
  2266.  
  2267. Does MS-Kermit 2.3x "talk" to the printer port continually when Cntrl-Prt-Scn
  2268. is in use ...... even when there is no input to or output from the Host?
  2269.  
  2270. We link to a spooled printer on a LAN by linking PRN: to the network printer.
  2271. The spool file automatically closes and writes to the printer if no data is
  2272. sent to this file for a period of 30 seconds (3Com EtherSeries !). This works
  2273.  
  2274. Page 40                                                INFO-KERMIT DIGEST V8 #5
  2275.  
  2276. fine in MS-Kermit Version 2.29 when, with the Ctrl-Prt-Scn activated, and no
  2277. keyboard activity for 30 seconds the printer echoes.  With MS-Kermit 2.3x this
  2278. is not the case.  Even with no activity (unless Kermit generates some ?), the
  2279. print spool file stays open indefinitely or until explicitely closed by
  2280. unlinking the printer or dumping a new file at it.  I have read how 2.30 uses
  2281. the interrupts etc. but cannot understand why there is this difference.  Is
  2282. this a bug or feature?
  2283.  
  2284. Any ideas? Is there a SET option to overcome this?
  2285.  
  2286. Thanks.
  2287. Dermot O'Beirne                <DOBEIRNE@IRLEARN>
  2288. Computer Centre
  2289. University College Dublin
  2290. Ireland
  2291.  
  2292. [From jrd - While there seems to be a SET command for almost everything, there
  2293. is not one for your case, nor Kermit does send keep-alives to the printer. It
  2294. uses just normal DOS writes to the standard printer device (file handle 4)
  2295. when characters arrive, and that 'file' is opened and maintained by DOS rather
  2296. than by Kermit.  Buffering strategies by DOS will affect when the network
  2297. spooler flushes the queue.  Experience here with a similar LAN arrangement
  2298. (Novell NetWare) gives unpredictable results with many applications programs,
  2299. so you have company but no firm answers.]
  2300.  
  2301. ------------------------------
  2302.  
  2303. Date: Thu, 21 Jul 88 08:27 EDT
  2304. From: Pete Kanaitis <X979PK0P@VB.CC.CMU.EDU>
  2305. Subject: MS-Kermit 2.31 (?More parameters are needed) Message
  2306. Keywords: MS-DOS Kermit 2.31
  2307.  
  2308. I just downloaded MS-Kermit V2.31 1 July 1988.  I noticed that certain
  2309. commands with a question mark on the command line (without a carriage
  2310. return) cause the "?More parameters are needed" error message.  Typing "?"
  2311. or "set ?" (no carriage return on either one) as the *FIRST* command after
  2312. starting KERMIT from DOS will display the normal help, then the "?More
  2313. parameters..." message.
  2314.  
  2315. Pete Kanaitis
  2316. Allegheny-Singer Research Institute
  2317. X979PK0P@VB.CC.CMU.EDU
  2318.  
  2319. [From jrd - Thanks Pete.  This will be fixed in the next rlease.]
  2320.  
  2321. ------------------------------
  2322.  
  2323. Date: Mon, 25 Jul 88 14:14:44 +0300
  2324. From: Ze'ev (Steed) Shtadler <steed%TECHUNIX.BITNET@cuvmb.cc.columbia.edu>
  2325. Subject: MS-Kermit 2.31 Internal Modem Problems
  2326. Keywords: MS-DOS Kermit 2.31, SmartLink.
  2327.  
  2328. I got the new MS-Kermit 2.31 version, translated MSVIBM.BOO and tried it.
  2329. Well, it couldn't talk to the internal modem.  I couldn't make it operate in
  2330. any way I tried.  MS-Kermit 2.30 had no problems with the same hardware
  2331. configuration.
  2332.  
  2333. INFO-KERMIT DIGEST V8 #5                                                Page 41
  2334.  
  2335.  
  2336. I use an XT Turbo clone (COPAM 401), MS-DOS 3.30, SmartLink 1200B as COM1:,
  2337. no other RS232 cards, a real-time clock and 2 parallel ports.
  2338.  
  2339. Did anyone experience a similar behavior with this new version?  Using this
  2340. modem, Using other modems?
  2341.  
  2342. Any suggestions for solving/determining the problem?
  2343.  
  2344.                         Thanks in advance,
  2345.                                 Ze'ev Shtadler.
  2346.  
  2347. Ze'ev Shtadler.
  2348. Computer Science Faculty,
  2349. Technion, Israel Institute of Technology,
  2350. Haifa Israel
  2351. Phone (work):   (+972) 4-294304
  2352.  
  2353. BITNET:  steed@techunix
  2354. Domain:  steed@techunix.technion.ac.il
  2355. UUCP:    ...!psuvax1!techunix.bitnet!steed
  2356. ARPANET: steed%techunix.bitnet@cunyvm.cuny.edu
  2357. CSNET:   steed%techunix.bitnet@csnet-relay.csnet
  2358.  
  2359. [From jrd - I presume the IRQ is set to 4 and the i/o address to 03F8H, for
  2360. COM1. If the modem were not found Kermit would say either the port is
  2361. unavailable or operates through the Bios.  Otherwise, if the modem can be
  2362. dialed but the screen shows nothing then the IRQ jumper is not set to 4 for
  2363. COM1.  After that I'd be guessing.]
  2364.  
  2365. [Ed. - We have heard that certain internal modems can be configured by the
  2366. accompanying software (e.g. Smartcom for Hayes), and that this same software
  2367. can leave the modem in a state that is unusable by any other program,
  2368. including Kermit.  Sometimes fooling with the modem's configuration program
  2369. can clear up the problem.]
  2370.  
  2371. ------------------------------
  2372.  
  2373. Date: Thu, 4 Aug 88 14:37 MDT
  2374. From: <JRD%USU.BITNET@CUVMB.CC.COLUMBIA.EDU>
  2375. Subject: RE: MS-Kermit 2.31 Bug (Feature?)
  2376. Keywords: MS-DOS Kermit 2.31, Tabs
  2377.  
  2378. In response to Carl Fussell's message in Info-Kermit Digest V8 #4:
  2379.  
  2380. For the life of me I can see no reason why CLEAR and Tab stops
  2381. could interact this way. The VT102 will clear all tab stops when it receives
  2382. the command ESC [ 3 g from the host.  As a quick check you might add the
  2383. STATUS command to the script file in a few places; tabs will be shown at the
  2384. end of the display.  Even then, the terminal emulator is not activated until
  2385. the Connect command.  A puzzle, or a damaged BOO file.
  2386.  
  2387. ------------------------------
  2388.  
  2389. Date: 1 Aug 88 14:01:12 EDT (Mon)
  2390. From: chris@bitsy.cooper.EDU (Christopher P. Lent)
  2391.  
  2392. Page 42                                                INFO-KERMIT DIGEST V8 #5
  2393.  
  2394. Subject: MS-Kermit 2.31 IBM 17 July 1988 Eats nulls
  2395. Keywords: MS-DOS Kermit 2.31
  2396.  
  2397. Hi,
  2398.     Sorry to complain, but MS-Kermit 2.31 17 July 1988 removes nulls from
  2399. the terminal session.  Normally this is not a problem, BUT we have an
  2400. application which captures binary data from a Perkin-Elmer FT-IR Model 1600
  2401. via the LOG SESSION command.  Since MS-Kermit 2.31 eats nulls, we have to
  2402. remain at MS-Kermit 2.30 for now.
  2403.  
  2404. REPEAT BY:
  2405.  
  2406. If you have a UNIX system around an easy way to test this is to go:
  2407.  
  2408. $ echo "A NULL \0 <---" >null.lis
  2409. $ od -c null.lis
  2410. 0000000   A       N   U   L   L      \0       <   -   -   -  \n
  2411. 0000016
  2412. $ #                                  ^------- Note the null in the file
  2413. $ # at this point turn on debugging with SET DEBUG ON
  2414. $ # then reconnect and go
  2415. $ cat t.lis
  2416.  
  2417. On Version 2.30 the null appears as ^@ and is also put in the SESSION LOG
  2418. file, while on 2.31 the null is gone from both the terminal session and the
  2419. session log file.
  2420.  
  2421. In conclusion, Kermit is wonderful but this problem makes our user rather
  2422. unhappy.  Admittedly it's not Kermit's intended use but it would prove
  2423. useful from time to time.  Hope a fix will make it into the next MS-Kermit
  2424. release.
  2425.  
  2426. Thanks,
  2427. Chris Lent
  2428. chris@cooper.cooper.edu
  2429. chris%cooper.cooper.edu@cmcl2.NYU.EDU
  2430. philabs!phri!cooper!chris
  2431.  
  2432. [From jrd - Chris, to increase terminal emulation throughput both NULLs and
  2433. DEL characters are filtered out right in the serial port interrupt routine.
  2434. DEL is passed if Tektronix emulation is active.  Both were supposed to be
  2435. passed when DEBUG was active but a small glitch did not let NULL through
  2436. even then.  We dare not let these two characters get into the log file since
  2437. they are normally fillers and would make a great mess when the log file
  2438. is later edited or manipulated by ordinary tools.  I suppose a case could be
  2439. made that both should be passed if transparent printing were active, but
  2440. again filler characters from the host would also go to the printer.  Still,
  2441. they should not be in the session log.]
  2442.  
  2443. ------------------------------
  2444.  
  2445. Date: Wed, 6 Apr 88 15:59:30 EST
  2446. From: LES@vaxi.UWO.CDN
  2447. Subject: VT200 Support in MS-Kermit?
  2448. Keywords: VT200 Support
  2449.  
  2450.  
  2451. INFO-KERMIT DIGEST V8 #5                                                Page 43
  2452.  
  2453. What we would really like to see Kermit do is emulate at least one feature
  2454. of the VT200 - the ability to download soft character sets.  That is, we
  2455. would like to use Kermit with an application which manages and down loads
  2456. software character sets to the terminal in order to allow a wide variety of
  2457. characters to be displayed.
  2458.  
  2459. Do you know:
  2460.  
  2461.  - if such a feature is planned?
  2462.  - if anyone out there have implemented such a feature
  2463.  - if, given we have the source and resources, its a big job?
  2464.  
  2465. Thanks in advance,
  2466. Les Flodrowski
  2467. Social Science Computing Lab, U. of Western Ontario.      LES@UWOVAX.BITNET
  2468.  
  2469. [From jrd - VT220 support has been discussed but no one has volunteered to
  2470. take on the task.  Once VT220 support starts it needs to be rather complete
  2471. because people naturally then expect all the features.  A partial work around
  2472. is to load the SET TRANSLATION INPUT table from a Take file, to change in
  2473. coming characters, and to use SET KEY for outgoing ones.  In any case, the
  2474. displayable characters are restricted to those built into the Bios and display
  2475. adapter; i.e., the ordinary ASCII plus IBM extended ASCII set.  Since sources
  2476. are available you might wish to know that the job is certainly not small.]
  2477.  
  2478. ------------------------------
  2479.  
  2480. Date: Mon, 1 Aug 88 11:54:33 edt
  2481. From: Jim Gogan <ugogan@uncecs.edu>
  2482. Subject: MS-Kermit with Async Servers?
  2483. Keywords: MS-DOS Kermit 2.31
  2484.  
  2485. I would like to hear from anyone using MS-Kermit 2.31 over a network to an
  2486. async communications server (such as IBM's LAN Async Communications Server
  2487. Program or Novell's ACS).  We are looking for something (anything!!) that
  2488. will let us attach up to 8 hardwired async lines to a communications server
  2489. on a PC network (primarily token ring and Ethernet hardware) without having
  2490. to use ASCOM, Smartcom or Crosstalk!!  Is this possible with Kermit 2.31?
  2491. What products are any good (not just available) for async server capability?
  2492. I don't care about dial-in capabilities; we just want to get access to other
  2493. campus hosts for the users on the PC nets.  Please e-mail any and all
  2494. success/horror stories & I'll summarize if appropriate.  Thanks in advance.
  2495.  
  2496.    Jim Gogan
  2497.    Univ. of North Carolina
  2498.    Microcomputing Support Center
  2499.    Phone: (919) 962-0101
  2500.    E-mail: ugogan@ecsvax (UUCP or BITNET) or ugogan@ecsvax.uncecs.edu
  2501.  
  2502. [From jrd - IBM's Async Comms Server needs a special protocol across the
  2503. network, and I suspect so does Novell's ACS. I have the specs on the IBM
  2504. item; in fact, I have a MS student doing an analysis of it. However, it
  2505. does not mesh well with the current Kermit procedures. There might be some
  2506. async servers, such as the updated IBM ASC from Yale, which expect the client
  2507. to use the local Bios path to the serial port; Kermit supports that route
  2508. via SET PORT BIOSn (n=1-4).]
  2509.  
  2510. Page 44                                                INFO-KERMIT DIGEST V8 #5
  2511.  
  2512.  
  2513. ------------------------------
  2514.  
  2515. Date: Thu, 21 Jul 88 07:59:00 CDT
  2516. From: moore@ncsc.ARPA (Moore)
  2517. Subject: MS-DOS Kermit 2.31 and Local Printer
  2518. Keywords: MS-DOS Kermit 2.31
  2519.  
  2520. Sorry, I should have looked closer before I complained about the local
  2521. printer problem a while back.  Seems that, prior to 2.30, parity none
  2522. allowed transparent printing with no problem; with 2.31, at least on our
  2523. 4.3bsd host, parity must be set to space in order to let both transparent
  2524. printing and file transfers take place (setting parity to even disables
  2525. transfers completely!).
  2526.  
  2527. Thanks for the assistance.
  2528.  
  2529. Jim
  2530.  
  2531. [Ed. - The behavior of parity is rather complicated.  UNIX sends even parity
  2532. during a terminal session.  MS-Kermit, by default, strips parity of incoming
  2533. characters during CONNECT.  UNIX Kermit, once it is invoked, transfers files
  2534. with no parity (uses all 8 bits for data).  Thus normally, you don't need to
  2535. SET PARITY to anything when connecting a PC to a UNIX system -- both
  2536. terminal connection and file transfer work.  Transparent printing, however,
  2537. as of 2.30, is intended to work with full 8-bit binary bytes to allow for
  2538. printing of special characters.  If you don't tell MS-Kermit that incoming
  2539. characters have parity, it will route them to the printer that way, which of
  2540. course produces a messed-up listing.  So for transparent printing with a VAX
  2541. on the other end, you should SET PARITY EVEN (or SPACE or anything else),
  2542. but then to transfer files, also remember to SET PARITY EVEN (or SPACE
  2543. or...) on the VAX Kermit too.]
  2544.  
  2545. ------------------------------
  2546.  
  2547. Date: Tue, 2 Aug 88 12:24:21 edt
  2548. From: jlupo%icc@afit-ab.arpa (James A. Lupo)
  2549. Subject: MS-Kermit 2.3x Bug?  2.31 And Z-100's ZPC
  2550. Keywords: MS-DOS Kermit 2.31, Z-100, ZPC
  2551.  
  2552. I have noted a slight bug in MS-Kermit, Versions 2.30 and 2.31, with the
  2553. "set term color" command.  If bright foreground colors are specified, as
  2554. in:
  2555.     set term color 1 37 44
  2556.  
  2557. The terminal display does indeed show bright white characters on a blue
  2558. background.  However, receipt of a terminal reset sequence, specifically,
  2559. <ESC>[0m, causes the display to drop to low intensity white on black.
  2560. The screen remains this way until one escapes to the Kermit command mode
  2561. and subsequently re-connects to terminal mode.  At that point, new text
  2562. is displayed once again in high intensity until the next reset sequence is
  2563. received.
  2564.  
  2565. Side Note for Z-100 Users
  2566.  
  2567. I recently upgraded my Z-100 low profile computer with the Heath User's
  2568.  
  2569. INFO-KERMIT DIGEST V8 #5                                                Page 45
  2570.  
  2571. Group ZPC IBM-PC emulation software and a Scottie Systems Scottie Board with
  2572. COM1 adapter.  I had great difficulty in getting Version 2.30 to run, and
  2573. have not isolated all problems yet.  However, Version 2.31 runs hot out of
  2574. its BOO file translation with no modification.  The ZPC package emulates a
  2575. CGA display, and both the VT102 and TEK4010 modes so far operate flawlessly.
  2576. Even though running Kermit in emulation this way does not make use of the
  2577. 640x480 resolution available on a Z-100, it still makes a very servicable
  2578. comm package (at least until the Z-100 Kermit reaches the 2.3x level??).
  2579.  
  2580. LtCol James A. Lupo
  2581. Assistant Professor of Physics
  2582. Air Force Institute of Technology (AFIT/ENP)
  2583. Wright-Patterson AFB, OH 45433-6583
  2584.  
  2585. E-Mail:  jlupo@afit-ab.arpa
  2586.  
  2587. [From jrd - ESC [ 0 m means reset the display attributes of bold, blink, and
  2588. inverse rather than reset the terminal (ESC c), so a dim display is the
  2589. natural result.  Color change is not a host command and probably relates to
  2590. your non-Connect mode screen attributes (which Kermit uses until told
  2591. otherwise).  ZPC emulation: that's good news!]
  2592.  
  2593. ------------------------------
  2594.  
  2595. Date: Thu, 4 Aug 88 10:09:35 CDT
  2596. From: moore@ncsc.ARPA (Moore)
  2597. Subject: Z-248 COM3 and Kermit
  2598. Keywords: Zenith Z-248, COM3, MS-DOS Kermit
  2599.  
  2600. I know the subject of COM3 on the Z-248 has been beaten to death, but I got
  2601. an interesting little program from Zenith called COM3.COM, which is supposed
  2602. to map COM[1-2] I/O to COM3.  Well, it works at 1200baud with Kermit, but at
  2603. any higher speeds characters get lost regardless of the handshaking.
  2604.  
  2605. Has anyone been able to get COM3.COM to work with Kermit?  I'll post results
  2606. if they're compelling.
  2607.  
  2608. Thanks.
  2609.  
  2610. Jim
  2611. Moore@NCSC.arpa
  2612.  
  2613. ------------------------------
  2614.  
  2615. Date: Mon, 25 Jul 88 18:33:15 EST
  2616. From: John C Klensin <KLENSIN@INFOODS.MIT.EDU>
  2617. Subject: MSUCHK/SCANCHECK and the Rainbow
  2618. Keywords: MS-DOS Kermit, Rainbow Kermit
  2619.  
  2620. I probably should have anticipated this, but the MSKERMIT scan program,
  2621. SCANCHECK (distributed as MSUCHK) goes crazy when one attempts to run it
  2622. on a Rainbow under MSDOS 2.1.  SHOW KEY from within Kermit, while
  2623. tedious, works fine.
  2624.  
  2625. This might at least be worth a note in a .BWR file.
  2626.  
  2627.  
  2628. Page 46                                                INFO-KERMIT DIGEST V8 #5
  2629.  
  2630. [Ed. - Noted.]
  2631.  
  2632. ------------------------------
  2633.  
  2634. Date: Fri, 22 Jul 88 11:10:58 EDT
  2635. From: mates%applga.uucp@umix.cc.umich.edu (Valerie Mates)
  2636. Subject: MacKermit Checksums
  2637. Keywords: MacKermit 0.9(40)
  2638.  
  2639. Hi!  I'm using MacKermit 9(40) on a Mac II, connected to a Sun running
  2640. C-Kermit 4E.  During file transfers when it should be using type 1
  2641. checksums, MacKermit's display window says it's using type 0 checksums.  It
  2642. also says that the window size is 00.  Several large, binary files have been
  2643. corrupted during transfers with "type 0" checksums.  Any idea what's
  2644. happening?
  2645.  
  2646. Thanks!
  2647. Valerie Mates
  2648.  
  2649. [Ed. - The block check display is probably a mistake.  Kermit doesn't
  2650. transfer files without error checking.  We'd have to know a lot more detail
  2651. about the corrupted binary files -- what system was on the other end, what
  2652. version of Kermit, what kind of connection, etc.]
  2653.  
  2654. ------------------------------
  2655.  
  2656. Date: Fri, 15 Jul 88 00:42:26 EDT
  2657. From: mholden@ajpo.sei.cmu.edu
  2658. Subject: MacKermit and Binary File Transfer
  2659. Keywords: MacKermit 0.9(40), MacBinary
  2660.  
  2661. The problem I am trying to work with MacKermit is to transfer a Stuffit
  2662. file (which is MacBinary) from the SIMTEL20 archives to my Macintosh via
  2663. another, UNIX, ARPANET host.  The master files are being retained in binary
  2664. on SIMTEL20 to conserve disk space.  Since I have no disk space there and
  2665. am also unaware of a Mac file ASCII conversion program on the DEC20) I
  2666. cannot convert it there before transfer. I do have disk space on the UNIX
  2667. host but I also am not aware of a conversion program resident there. I
  2668. have done XMODEM transfer from there since one of the system people kindly
  2669. put it up for me so I am not desperate.  However I was curious as to whether
  2670. there exists a technique for double file transfer with subsequent glueing
  2671. back together. I would like to suggest that support of MacBinary transfer
  2672. would make the program enormously more useful for Mac users.
  2673.  
  2674. - Maretta Holden
  2675.  
  2676. [Ed. - See the next message...]
  2677.  
  2678. ------------------------------
  2679.  
  2680. Date: Fri, 15 Jul 88 10:21:36 PDT
  2681. From: dplatt@coherent.com (Dave Platt)
  2682. Subject: Transferring application via MacKermit 0.9(40)
  2683. Keywords: MacKermit 0.9(40), MacBinary
  2684.  
  2685. If you wish to transfer a Mac application (or other file containing a
  2686.  
  2687. INFO-KERMIT DIGEST V8 #5                                                Page 47
  2688.  
  2689. resource fork) with MacKermit 0.9(40), I'd suggest running the file through
  2690. BinHex 5.0 (not 4.0).  BinHex 5.0 will translate the file into MacBinary
  2691. format; the resulting file is pure data (no resource fork) and can be
  2692. uploaded using MacKermit's binary-file transmission option.  The resulting
  2693. file can be downloaded in the same way, and then reconverted via BinHex 5.0.
  2694.  
  2695. The biggest advantage to using BinHex 5.0, as opposed to PackIt or StuffIt,
  2696. is that the uploaded file on the mainframe/BBS is stored in MacBinary
  2697. format.  It's therefore compatible with any terminal emulator/downloader
  2698. that supports automatic MacBinary conversion; for example, one could
  2699. download such a file using MacTerminal 2.0, Red Ryder, MicroPhone, FreeTerm,
  2700. or VersaTerm... and all of these programs would automatically convert the
  2701. file back into a Mac application as it was received.  Similarly, any of
  2702. these programs could be used to upload an application or document to a
  2703. mainframe or BBS in MacBinary format; the resulting file can be downloaded
  2704. using binary MacKermit and then run through BinHex 5.0 to restore it to its
  2705. original form.
  2706.  
  2707. Dave Platt                                             VOICE: (415) 493-8805
  2708.   USNAIL: Coherent Thought Inc.  3350 West Bayshore #205  Palo Alto CA 94303
  2709.   UUCP: ...!{ames,sun,uunet}!coherent!dplatt     DOMAIN: dplatt@coherent.com
  2710.   INTERNET:   coherent!dplatt@ames.arpa,    ...@sun.com,    ...@uunet.uu.net
  2711.  
  2712. [Ed. - And obviously, automatic MacBinary translation is high on the list
  2713. for Mac Kermit too.]
  2714.  
  2715. ------------------------------
  2716.  
  2717. Date: Tue 26 Jul 88 17:19:30-PDT
  2718. From: Mike O'Rourke <MIKEO%HADAR.usc.edu@oberon.usc.edu>
  2719. Subject: MacKermit Escape Sequence Keys
  2720. Keywords: MacKermit 0.9(40)
  2721.  
  2722. Greetings.  I am a MicroComputer Consultant for the University of Southern
  2723. California.  We recently received MacKermit 0.9(40) and MSKermit 2.30 for
  2724. our Labs and have a question for you.
  2725.  
  2726. Many of our users connect to IBM Mainframes (MVS and CMS) using XVT100
  2727. emulation (VT102 really), and have trouble trying to find out the escape
  2728. sequences - PF Keys, Break keys, Insert Keys, Clear Screen, etc.  If you
  2729. have such mappings, most notably for the Mac, we would dearly appreciate any
  2730. info you could offer.
  2731.  
  2732. Thanks in Advance,
  2733.  
  2734. Mike O'Rourke
  2735.  
  2736. [Ed. - Mac Kermit 0.9(40) does not come with a default set of VT100 key
  2737. mappings.  Users are encouraged to put together Mac Kermit settings files
  2738. for this and send them in, so we can add them to the distribution.]
  2739.  
  2740. ------------------------------
  2741.  
  2742. End of Info-Kermit Digest
  2743.  
  2744. Page 48                                                INFO-KERMIT DIGEST V8 #6
  2745.  
  2746. Info-Kermit Digest         Wed, 17 Aug 1988       Volume 8 : Number 6
  2747.  
  2748. Today's Topics:
  2749.  
  2750.                         Announcing the End of CU20B
  2751.              Announcing a Maintenance Update for MS-Kermit 2.31
  2752.           Announcing MS-Kermit 2.31 for NEC-PC9801 with Kana-Kanji
  2753.                Announcing MS-DOS Kermit 2.31 for the Wang PC
  2754.                         Binary File Types on the VAX
  2755.      Kermit 3.82 and the Applied Eng. Serial Pro Work Together Finally
  2756.                    CMS Kermit and 7171 Protocol Converter
  2757.                        CMS-Kermit with Commtex Cx-80
  2758.               KERMIT through Commtex Cx-80 Protocol Converter
  2759.                 Kermit Through Protocol Converters Under CMS
  2760.  
  2761. ----------------------------------------------------------------------
  2762.  
  2763. Date: Tue, 15 Aug 88 19:16 MDT
  2764. From: Christine M. Gianone <SY.CHRISTINE@CU20B.CC.COLUMBIA.EDU>
  2765. Subject: Announcing the End of CU20B
  2766. Keywords: CU20B, CUNIXC, Kermit Distribution, Network, FTP
  2767.  
  2768. This will be the final issue of the Info-Kermit Digest that is sent from our
  2769. last DEC-20 (once a fleet of four).  DEC-20s were important to the original
  2770. development of Kermit -- much of the protocol was designed "around" them, one
  2771. of the first Kermit programs was written for the DEC-20, and the DEC-20 style
  2772. of "user interface" is found in many popular Kermit programs -- MS-Kermit,
  2773. CP/M Kermit, C-Kermit, etc.
  2774.  
  2775. Kermit's new "home" will be on CUNIXC.CC.COLUMBIA.EDU, a VAX 8700 running
  2776. UNIX (Ultrix).  The IP host number is 128.59.40.130.  You should be able to
  2777. get files from it with anonymous FTP, just as you did from CU20B, except that
  2778. now you're talking to a UNIX, rather than TOPS-20, FTP server.  The Kermit
  2779. files are in directories kermit/a, kermit/b, kermit/c, kermit/d, and kermit/e.
  2780. You can also send mail to Info-Kermit and Info-Kermit-Request at the new
  2781. address.
  2782.  
  2783. The process is roughly like this: FTP to CUNIXC.CC.COLUMBIA.EDU, login
  2784. anonymous, any password, and then CWD (or CD) to kermit, which contains a
  2785. file read.me, which you should read.  Then CWD (or CD) to a, b, c, d, or e
  2786. and GET or MULTIPLE GET (or MGET) the files you want.  The AA*.* files are
  2787. still DEC-20 oriented, but will be updated soon.  Report problems to
  2788. info-kermit-request@CUNIXC.CC.COLUMBIA.EDU.  Also, remember that Columbia's
  2789. Internet host numbers changed on August 15th, as announced in the last
  2790. Info-Kermit Digest, so if you have trouble connecting, try using CUNIXC's
  2791. host number 128.59.40.130, instead of the host name.
  2792.  
  2793. Since CUNIXC is a heavily-used machine, please try not to FTP files from it
  2794. during peak hours.  There are presently no formal restrictions, so please
  2795. don't overburden the system.  If the response is very slow, cancel your
  2796. session and try again later.
  2797.  
  2798. During a brief transition period, CU20B will continue to house the Kermit
  2799. files, so if you have problems with CUNIXC, you can still return to "old
  2800. faithful".
  2801.  
  2802.  
  2803. INFO-KERMIT DIGEST V8 #6                                                Page 49
  2804.  
  2805. To order by mail, request a complete list of Kermit versions and an order
  2806. form from Kermit Distribution, Columbia University Center for Computing
  2807. Activities, 612 West 115th Street, New York, NY 10025 USA.
  2808.  
  2809. ------------------------------
  2810.  
  2811. Date: Wed, 10 Aug 88 19:16 MDT
  2812. From: Christine M. Gianone <SY.CHRISTINE@CU20B.CC.COLUMBIA.EDU>
  2813. Subject: Announcing a Maintenance Update for MS-Kermit 2.31
  2814. Keywords: MS-DOS Kermit 2.31, IBM PC, Ungermann-Bass
  2815.  
  2816. This is to announce a maintenance release of MS-DOS Kermit 2.31 for the IBM
  2817. PC family only.  This version is being issued in order to correct several
  2818. problems in the original 2.31 release.  If you are affected by these
  2819. problems, then you should get the update, otherwise you needn't bother.  The
  2820. version number remains 2.31, but the date is changed to August 15, 1988 (as
  2821. displayed in the startup herald, or by the "version" command).  The items
  2822. that were fixed are:
  2823.  
  2824.  - Ungermann-Bass Net One support had two fatal bugs, now fixed.
  2825.  
  2826.  - The Bios in some machines objected to the stray interrupt generated when
  2827.    Kermit disconnected from the serial port.  Also solves problem of
  2828.    an earlier task, such as Smartcom II or PC-Talk III, forgetting to replace
  2829.    the serial port interrupt it borrowed and causing a crash during Kermit
  2830.    operations.
  2831.  
  2832.  - A serial port with COM2 addressing (port 02E8H, IRQ 3) placed in the
  2833.    COM1 Bios data area, such as happens with some PCjr's, could not be
  2834.    found.  Corrected to switch to COM2 addressing if the Bios data area
  2835.    for COM1 is 02E8H.  COM1 is normally port 03F8H and IRQ 4.
  2836.  
  2837.  - The main Kermit command line parser sometimes gave a spurious message
  2838.    "?more parameters are needed" when query mark help was used.
  2839.  
  2840.  - A bug in finding labels in scripts, as in the commands
  2841.  
  2842.         GOTO end
  2843.         :end1
  2844.         commands
  2845.         :end
  2846.  
  2847.    which let label ":end1" be jumped to by mistake if ":end1" appeared first.
  2848.  
  2849.  - The test for available disk space when receiving files with attributes
  2850.    packets inflated the disk space rather than the file size.  Revised to
  2851.    proper form: inflate file size by 17/16, decline file if not enough
  2852.    disk space.
  2853.  
  2854.  - The script CLEAR command could sometimes interfere with terminal emulation
  2855.    (lose tab settings, etc).
  2856.  
  2857. [Ed. - Thanks to Joe Doupnik for the rapid work in making these corrections!
  2858. The new files are installed in Kermit distribution in the usual places, as
  2859. well as in the new place (kermit/a on CUNIXC) and a more detailed description
  2860. from Joe of this release has been added to the MSKERM.BWR file.]
  2861.  
  2862. Page 50                                                INFO-KERMIT DIGEST V8 #6
  2863.  
  2864.  
  2865. ------------------------------
  2866.  
  2867. Date: SAT, 06 AUG 88 17:19:25 JST
  2868. From: "H.Fujii" <KEIBUN@JPNKEKVM.BITNET>
  2869. Subject: Announcing MS-Kermit 2.31 for NEC-PC9801 with Kana-Kanji
  2870. Keywords: NEC-PC9801, Japan, Kana, Kanji, MS-DOS Kermit 2.31, Fujitsu FMR-60
  2871.  
  2872. We (Hirofumi Fujii & Akihiro Shirahasi) have developed MS-Kermit for NEC
  2873. PC-9801, the most popular MS-DOS machine in Japan (more than IBM-PC!!).  The
  2874. big difference from other machines is Kana & Kanji treatment.  Kana & Kanji
  2875. are the Japanese characters developed from ancient Chinese characters. The
  2876. number of authorized characters is about 6000 (Level-1: 2965; Level-2:
  2877. 3388).  These are specified by 2-byte codes.  Our program can handle these
  2878. characters in addition to usual MS-Kermit operation (including VT100,
  2879. TEK4014 emulation).
  2880.  
  2881. For Kana & Kanji treatment, it is very important that the MS-Kermit has the
  2882. 8-bit display mode.  So, we have set the default mode to 8-bit display.
  2883.  
  2884. We have already tested this program with my friends and collaborators.  We
  2885. have tried to send Kana-Kanji mail and Kana-Kanji phone on VAX/VMS using
  2886. this Kermit.  Of course these utilities caused some problems because these
  2887. programs are not designed for 2-byte cahracters.  But many of the problems
  2888. are recoverable (e.g., by pressing ^W) and we found this program is very
  2889. convenient to communicate using Kana & Kanji.
  2890.  
  2891. So, we want to distribute this program as a 'free to use' program.  We have
  2892. named the modules as follows:
  2893.  
  2894.    MSVP98.BOO  -- BOO file for MSVP98.EXE (MS-Kermit for NEC PC-9801)
  2895.    MSVP98.DOC  -- Manual for NEC PC-9801 Kermit
  2896.  
  2897.    MSGP98.ASM  -- Macro Assembler source file for TEK4014 emulator
  2898.    MSUP98.ASM  -- Macro Assembler source file for Keyboard
  2899.    MSXP98.ASM  -- Macro Assembler source file for Machine-dependent interface
  2900.    MSYP98.ASM  -- Macro Assembler source file fir VT102 emulator
  2901.  
  2902. These are the files for MS-Kermit 2.31 for NEC-PC9801.  Please add these
  2903. files to your KERMSRV.
  2904.  
  2905.                                              Hirofumi Fujii
  2906.                                       KEK On-line group, Dept. of Physics,
  2907.                              National Laboratory for High-Energy Physics (KEK)
  2908.  
  2909.                                         BITNET address: KEIBUN@JPNKEKVM
  2910.                                 HEPNET(DECNET) address: KEKVAX::KEIBUN
  2911.  
  2912. P.S.
  2913. 1) I tried to SET KEY xxx <Kana-Kanji characters>. It worked well!  This is
  2914. owing to 8-bit character handling at MS-Kermit command level (a new feature
  2915. of version 2.31).
  2916. 2) I have already started to install this program to another Japanese MS-DOS
  2917. machine, Fujitsu (mainframer) FMR-60 (their mainframe-machine is IBM
  2918. compatible but their peronal computer is NOT! -- Why?  I think it is due to
  2919. Japanese character sets).  At the present time, it works without Tek4014
  2920.  
  2921. INFO-KERMIT DIGEST V8 #6                                                Page 51
  2922.  
  2923. emulation.
  2924.  
  2925. [Ed. - Many thanks for the valuable work you have done, and for contributing
  2926. it to the Kermit library!  The files have been installed in Kermit
  2927. Distribution under the names you listed above.]
  2928.  
  2929. ------------------------------
  2930.  
  2931. From: Volker Edelhoff <edelhoff@UNIDO.BITNET>
  2932. Date: Fri, 22 Jul 88 14:39:42 +0200
  2933. Subject: Announcing MS-DOS Kermit 2.31 for the Wang PC
  2934. Keywords: MS-DOS Kermit 2.31, Wang PC
  2935.  
  2936. I adapted the machine-specific file MSXWNG.ASM for the Wang PC to assemble
  2937. and link with the rest of MS-DOS Kermit version 2.31 files correctly.  I did
  2938. no improvement to the functionality of it - I just added some externals and
  2939. dummy procedures to satisfy the linker.
  2940.  
  2941. BUT: the machine independent files of Kermit have been improved immensely
  2942. (script file, more comprehensive messages, better SHOW facilities, ...), so
  2943. that this adaption will be quite useful for the WANG PC/APC world, anyway!
  2944.  
  2945. The new files are as follows:
  2946.  
  2947.   Name         Size   Description
  2948.   MSVWNG.BAT   1223   Batch file for assembling and linking
  2949.   MSVWNG.BOO  71012   BOO-version of MSVWNG.EXE
  2950.   MSVWNG.EXE  63616   The WANG PC/APC Kermit Version 2.31
  2951.   MSVWNG.LNK    204   Link-script file used by MSVWNG.BAT
  2952.   MSVWNG.MAK   1890   MAKE-file for MS-Make
  2953.   MSXWNG.ASM  25256   The ASM-source code
  2954.  
  2955.                                 Volker Edelhoff
  2956.                                 Hennener Str. 35
  2957.                                 D-5860 Iserlohn
  2958.                                 West Germany
  2959.  
  2960. [Ed. - Many thanks, Volker!  The Wang PC version of MS-Kermit has not been
  2961. updated since 2.29, in May 1986, so this announcement should be most welcome
  2962. to Wang PC users!  The files have been added to Kermit Distribution under
  2963. the names listed above.  Now if only someone would write Kermit programs for
  2964. the Wang VS systems...]
  2965.  
  2966. ------------------------------
  2967.  
  2968. Date: Tue, 26 Jul 88 09:34:32 CST
  2969. From: Bob Parks <C27901RP%WUVMD.BITNET@CUVMB.CC.COLUMBIA.EDU>
  2970. Subject: Binary File Types on the VAX
  2971. Keywords: VAX/VMS Kermit, SAS
  2972.  
  2973. My problem is most likely due to ignorance of VAX file formats but here is
  2974. the problem.
  2975.  
  2976. SAS has a file format called TRANSPORT format and one uses it to transport
  2977. SAS internal data sets between machines (VAX, PC, CMS, etc).  The transport
  2978. file is written in ASCII and apparently uses all 256 different 'characters'.
  2979.  
  2980. Page 52                                                INFO-KERMIT DIGEST V8 #6
  2981.  
  2982.  
  2983. The problem is that the transport data set is 80 characters per record, file
  2984. type fixed (e.g., no end-of-record indication - specifically no CR/LF
  2985. characters).  If such a transport file is transfered between the VAX
  2986. (KERMIT-32) and the PC, then at the end of each and every record there is a
  2987. CR/LF on the PC side.  I have tried SET FILE TYPE BINARY and SET FILE TYPE
  2988. FIXED and no SET FILE TYPE command and receieved the same files identically.
  2989.  
  2990. Possibly the VAX stores the transport data set with CR/LF even though it is
  2991. fixed length file (and maybe there is some fixup in SAS / VAX when creating
  2992. the transport data set).  BUT what I wanted, and what works between CMS and
  2993. the PC is CMS KERMIT's SET FILE BINARY that does not, on the PC side, get
  2994. CR/LF on each record.
  2995.  
  2996. Any suggestions appreciated.
  2997.  
  2998. Bob
  2999.  
  3000. ------------------------------
  3001.  
  3002. Date: Thu, 28 Jul 88 09:46:04 CDT
  3003. From: "Wayne J. Hauber" <GA.WJH%ISUMVS.BITNET@CUVMB.CC.COLUMBIA.EDU>
  3004. Subject: Kermit 3.82 and the Applied Eng. Serial Pro Work Together Finally
  3005. Keywords: Kermit-65, Super Serial Card, Apple II
  3006.  
  3007. A couple of months ago, I asked about Kermit 3.8* and the Applied Engineering
  3008. Serial Pro.  The Serial Pro is supposed to be Super Serial Card compatible but
  3009. I was having no luck getting the card to work with Kermit's Super Serial Card
  3010. driver.
  3011.  
  3012. I enabled interrupts like the Kermit manual (and everyone else) suggested by
  3013. closing the dip switches.  Kermit refused to work; when you enter CONNECT at
  3014. the Kermit-65 prompt the system freezes with the interrupt switches off and
  3015. appears to drop into the monitor with the switches on.
  3016.  
  3017. I tried lots of things and even browsed the Kermit source.  However, last
  3018. night I got Kermit to work with my Serial Pro.  The fix was unexpected and
  3019. seems a little weird to me.  I opened the switches that enable IRQ interrupts
  3020. for the 6551 and 6818 (switches 1 and 3) and closed the switches which enable
  3021. NMI interrupts (switches 2 and 4).  This should have disabled IRQ and enabled
  3022. NMI interrupts.
  3023.  
  3024. What is weird is this:  NMI interrupts are not supported on the Apple II
  3025. series of computer and IRQ is.  The switches I set should be the wrong
  3026. switches.  Non maskable interrupts (NMI) are supposed to crash an Apple II
  3027. into the monitor.  This sounds a lot like the symptoms I had.  Most people I
  3028. talked with thought it wouldn't hurt to leave all the interrupt switches on;
  3029. that was wrong in Kermit's case.  It doesn't hurt some of the other
  3030. communication programs on the market though.
  3031.  
  3032. My guess is that the AE Serial Pro User's Guide contains a mislabeled diagram
  3033. for its switch settings.  The switches labelled as NMI are really IRQ and
  3034. vice-versa.
  3035.  
  3036. (Ted, Mark or Murph):
  3037. With what should be the wrong switch settings, Kermit 3.82 works quite well
  3038.  
  3039. INFO-KERMIT DIGEST V8 #6                                                Page 53
  3040.  
  3041. with my Serial Pro.  My previous comments about the Serial Pro have been
  3042. placed in a .BWR (beware) file at Columbia and have been read elsewhere.
  3043. Could someone with the appropriate clout update the .BWR file telling Serial
  3044. Pro users to open interrupt switches 1 & 3 and close 2 & 4?
  3045.  
  3046. Wayne Hauber
  3047. Iowa State University
  3048. gawjh@isumvs
  3049.  
  3050. [Ed. - Your new message has been placed in APPLE.BWR.]
  3051.  
  3052. ------------------------------
  3053.  
  3054. Date: Wed, 3 Aug 88 11:11 EDT
  3055. From: <TLEWIS%UTKVX3.BITNET@cuvmb.cc.columbia.edu>
  3056. Subject: CMS Kermit and 7171 Protocol Converter
  3057. Keywords: CMS Kermit, Portable 370 Kermit, Protocol Converters
  3058.  
  3059. I am having major difficulties transferring files between a pc and CMS
  3060. through a 7171 protocol converter.  After a number of packets has been
  3061. transferred, the error message "Unable to receive acknowlegement from host"
  3062. appears.  The number of packets varies before this error appears and
  3063. sometimes it works perfectly with no error.  Is anyone else having problems
  3064. like this?  I have tried different bauds to no avail.  It seems to be an
  3065. intermitten problem.  Not only is it happening under CMS, but also on our
  3066. MUSIC/SP system which also goes through the 7171.  It leads me to believe
  3067. that it is something with the 7171 causing it.  Any information or advice
  3068. would be greatly appreciated!
  3069.  
  3070. Terry Lewis
  3071. University of Tennessee at Martin
  3072. TLEWIS@UTKVX1
  3073.  
  3074. [Ed. - If the problems are mainly when you're sending from the PC to the
  3075. host, they are likely traceable to small or insufficient input buffers on the
  3076. 7171.  Make sure you're running an up-to-date 7171 system, and up-to-date
  3077. Kermits.  We can transfer 1000-byte packets from PCs into our CMS system
  3078. through a 7171 without even a hiccup.  And of course, also make sure you have
  3079. your communication parameters set correctly -- usually, SET PARITY EVEN is
  3080. the only change you have to make to the PC from its defaults.]
  3081.  
  3082. ------------------------------
  3083.  
  3084. Date: Fri, 22 Jul 1988   11:52:26   CET
  3085. From: A0045%DK0RRZK0.BITNET@CUVMB.CC.COLUMBIA.EDU
  3086. Subject: CMS-Kermit with Commtex Cx-80
  3087. Keywords: CMS Kermit, Kermit-370, CX-80
  3088.  
  3089. We have a small IBM mainframe, which we got as donation from IBM for the
  3090. scientific research in general and in particular for the 600-year-anniversary
  3091. of Cologne University.  At the moment, we have a protocol converter named
  3092. Commtex Cx-80 as a test device, and we tried in vain to get Kermit running
  3093. through that box.  I must confess, I'm not very intimate with IBM mainframes,
  3094. with their special vocabulary and their four-digit names, but I'll try to
  3095. explain the configuration.
  3096.  
  3097.  
  3098. Page 54                                                INFO-KERMIT DIGEST V8 #6
  3099.  
  3100. The IBM mainframe is a 4341 running VM/SP Rel.5 (CMS), the communications
  3101. controller is a 3705, where several remote cluster controllers are attached
  3102. via BSC (not SNA, as I was told). The Cx-80 can be used to attach
  3103. asynchroneous terminals and is in turn connected to a local network with
  3104. TCP/IP based on Ethernet.  We have installed CMS-Kermit 4.0 on the IBM, the
  3105. program runs succesfully through the converter, but the file transfer doesn't
  3106. work. I understand that it needs a transparent I/O mode in this
  3107. configuration, therefore I quote some lines from the Cx-80 documentation
  3108. about their transparent facilities:
  3109.  
  3110. >> In addition to the Cx-80's pass-thru mode, which is compatible
  3111. >> with many mainframe graphics and file transfer packages (e.g.,
  3112. >> products form SAS Institute and ISSCO), pass-thru operation
  3113. >> which is compatible with the IBM 7171 Hardcopy/Transparent
  3114. >> Write and Write/Read is now supported.
  3115.  
  3116. When I compare this with the CMS-Kermit documentation, I assume that the
  3117. first mode might correspond to SET CONTROLLER GRAPHICS, and the second mode
  3118. to SET CONTROLLER SERIES1.  CMS-Kermit comes up with SERIES1, which ends with
  3119. a bit more success than GRAPHICS.  What happens in SERIES1 mode is, that with
  3120. both XTYPE and Packet transfer most characters arrive correctly, but a few
  3121. are garbled (e.g. 7,- and & all appear as @). When I do a packet dump on the
  3122. CMS-machine, I see in the EBCDIC-dump all packets correctly, the RAW-dump
  3123. shows the incoming packets ok with their ASCII-code and the high-bit set, the
  3124. outgoing packets however, are dumped completely wrong.  So this dump is of no
  3125. great help either.  In GRAPHICS mode, I get an immediate disconnect from the
  3126. Cx-80, and there is a hint to this behaviour in the Cx-80 documentation:
  3127.  
  3128. >> In BSC systems, the Cx-80 will enter Pass-Though mode only if
  3129. >> the Command/WCC/X'70' sequence is written immediately following
  3130. >> device selection.
  3131. >> Any data which follows the X'70'is expected to be in the range
  3132. >> X'7F'through X'FE, inclusive. Upon receiving such data, the
  3133. >> Cx-80 will strip the most significant bit (resulting in data
  3134. >> in the range X'00'through X'7F', the full ASCII-128 set), and
  3135. >> deliver them untranslated to the addressed device.  ......
  3136. >> ... If any data following the X'70' is in the range X'40'
  3137. >> through X'7E'inclusive, the Cx-80 will interpret the data as
  3138. >> a host disconnect request and will respond by terminating the
  3139. >> connection between the terminal and the Cx-80.
  3140.  
  3141. Therefore my question: Does the CMS-Kermit conform to this description or is
  3142. it a special convention for the Cx-80 only?  Or do all the problems perhaps
  3143. occur, because CMS-Kermit is only designed for SNA-connections instead of BSC
  3144. and those two just are different in these places?
  3145.  
  3146. Jochen Roderburg
  3147. Regional Computing Center
  3148. University of Cologne
  3149. Robert-Koch-Str. 10                    Tel. :  49-221/470-4564
  3150. D-5000 Koeln 41                        Email:  A0045 @ DK0RRZK0.BITNET
  3151. West Germany
  3152.  
  3153. [Ed. - See following messages...]
  3154.  
  3155. ------------------------------
  3156.  
  3157. INFO-KERMIT DIGEST V8 #6                                                Page 55
  3158.  
  3159.  
  3160. Date: Sun, 17 Jul 1988 14:04 CDT
  3161. From: Landy Manderson <USTS034@UABTUCC>
  3162. Subject: KERMIT through Commtex Cx-80 Protocol Converter
  3163. Keywords: Protocol Converters, CX-80, TSO, 7171
  3164.  
  3165. I have a question about version 4.00 of MVS/TSO KERMIT, and I am hoping
  3166. this is the right place to send it.
  3167.  
  3168. We are using Commtex Cx-80 series 3270-to-ASCII protocol converters here at
  3169. UAB.  Our latest version of the Cx-80 software supports 7171 emulation, but
  3170. apparently only the Hardcopy/Transparent (115D7F110000) and Transparent
  3171. Write/Read (115D7F110001) commands.
  3172.  
  3173. Now, when I tried to do file transfer from PC to TSO through the Cx-80 I kept
  3174. getting a comm error 531, and it was apparently rejecting the 7171 orders
  3175. from KERMIT.  Upon inspection of the code, it appeared to be using the
  3176. command (115D7F110005), and it was sending ASCII data without the high-order
  3177. bit turned on.
  3178.  
  3179. I do not have a recent 7171 book (I do have one from 1984), so I do not have
  3180. any idea what command (....0005) is, but I changed our KERMIT to use the
  3181. Transparent Write/Read (....0001), and to set the high-order bit on in the
  3182. ASCII stream.  This seems to have worked, because now we can file transfer.
  3183.  
  3184. Is this a reasonable fix to our problem?  Is there any precedent for this,
  3185. and if so, has someone come up with a better way to do this?  Any help on
  3186. this would be appreciated.  Of course, this is not a super critical question
  3187. since we have gotten it to work as is.
  3188.  
  3189. Landy Manderson, Lead Software Specialist
  3190. User Services, The University Computer Ctr
  3191. University of Alabama at Birmingham (USA)
  3192.  
  3193. [From John Chandler - Landy, by coincidence, I have just heard from someone
  3194. else on BITNET who tried out Kermit (on CMS) through a Cx-80.  In that case,
  3195. it seemed that operation in SERIES1 mode was fine except for a few characters
  3196. that unaccountably got translated outbound to "@" signs.  I believe you'll
  3197. find that changing the write/read order was unnecessary, since the convention
  3198. in the 7171 and its ancestor, the Yale ASCII system, was to treat any buffer
  3199. address other than 0000 as a write/read.  I don't know why Kermit uses 0005,
  3200. instead of the more obvious 0001, but "that's how it has always been."  As to
  3201. the setting of the high-order bits in the ASCII data, we found that to
  3202. prevent the odd garbling of characters.  Below is the update I devised to
  3203. accomplish it.  It is unfortunate that this is necessary on the Cx-80, since
  3204. it is *not* on any other protocol converter that I know of -- in particular,
  3205. the 7171 can be configured to give a full 8-bit channel both ways for
  3206. convenient binary file transfer without 8th-bit quoting.
  3207.  
  3208. One point that interests me is one you did not mention, namely, the graphics
  3209. pass-through capability of the Cx-80.  What I heard of its documentation
  3210. suggested that you could also transfer files if you set CONTROLLER to
  3211. GRAPHICS in Kermit-TSO.  Would you mind trying it to find out?  Thanks.
  3212.  
  3213.                                   John
  3214. ======================================================================
  3215.  
  3216. Page 56                                                INFO-KERMIT DIGEST V8 #6
  3217.  
  3218. / I 08511000          $ 8511500 500                  08/02/88 14:59:32
  3219.          LA    9,0(5,2)      Save start of ASCII stuff         @SC88215
  3220. / I 08521000          $ 8521500 500                  08/02/88 14:59:32
  3221.          LR    3,2           Copy end of transmission          @SC88215
  3222. / I 08523000          $ 8523200 200                  08/02/88 14:59:32
  3223.          LR    2,9           Start of ASCII stuff              @SC88215
  3224.          SR    3,2           Length                            @SC88215
  3225.          LA    15,ON80       Set high bits                     @SC88215
  3226.          BAL   14,TRANSLAT                                     @SC88215
  3227. / I 08526500          $ 8526700 100                  08/02/88 14:59:32
  3228.          LR    6,8           Set up for log routine            @SC88168
  3229. / I 08575000          $ 8575050 50                   08/02/88 14:59:32
  3230. * For setting high bits...                                     @SC88215
  3231. ON80     DC    X'808182838485868788898A8B8C8D8E8F'             @SC88215
  3232.          DC    X'909192939495969798999A9B9C9D9E9F'             @SC88215
  3233.          DC    X'A0A1A2A3A4A5A6A7A8A9AAABACADAEAF'             @SC88215
  3234.          DC    X'B0B1B2B3B4B5B6B7B8B9BABBBCBDBEBF'             @SC88215
  3235.          DC    X'C0C1C2C3C4C5C6C7C8C9CACBCCCDCECF'             @SC88215
  3236.          DC    X'D0D1D2D3D4D5D6D7D8D9DADBDCDDDEDF'             @SC88215
  3237.          DC    X'E0E1E2E3E4E5E6E7E8E9EAEBECEDEEEF'             @SC88215
  3238.          DC    X'F0F1F2F3F4F5F6F7F8F9FAFBFCFDFEFF'             @SC88215
  3239.          DC    X'808182838485868788898A8B8C8D8E8F'             @SC88215
  3240.          DC    X'909192939495969798999A9B9C9D9E9F'             @SC88215
  3241.          DC    X'A0A1A2A3A4A5A6A7A8A9AAABACADAEAF'             @SC88215
  3242.          DC    X'B0B1B2B3B4B5B6B7B8B9BABBBCBDBEBF'             @SC88215
  3243.          DC    X'C0C1C2C3C4C5C6C7C8C9CACBCCCDCECF'             @SC88215
  3244.          DC    X'D0D1D2D3D4D5D6D7D8D9DADBDCDDDEDF'             @SC88215
  3245.          DC    X'E0E1E2E3E4E5E6E7E8E9EAEBECEDEEEF'             @SC88215
  3246.          DC    X'F0F1F2F3F4F5F6F7F8F9FAFBFCFDFEFF'             @SC88215
  3247. ]
  3248.  
  3249. ------------------------------
  3250.  
  3251. Date: Fri, 29 Jul 88 22:40:53 EDT
  3252. From: pepap%cfa3.span%CFA.BITNET@cuvmb.cc.columbia.edu
  3253. Subject: Kermit Through Protocol Converters Under CMS
  3254. Keywords: Kermit-370, Protocol Converters
  3255.  
  3256. Kermit-370, as released for CMS last winter, contained code designed to allow
  3257. file transfer through a variety of terminal emulators other than the "usual"
  3258. Yale ASCII type (Series/1, 7171, 4994, 9370 ASCII subsystem).  The code was,
  3259. in fact, modelled after similar code that was demonstrated under TSO, but I
  3260. haven't heard any reports of success under CMS, and the only negative report
  3261. was for a device that hadn't been tried under TSO.  I would like to hear from
  3262. anyone who has any kind of 3270 emulator other than the Yale type (and other
  3263. than the 3708, which is known to work with Kermit already).  The particular
  3264. types of emulators which might be expected to work include the MICOM 7400,
  3265. the Datastream/Leedata 8010/8030/874 models, the PCI 1076 and 276, and
  3266. possibly the Renex TMS-1.  Any others which claim to support a "graphics
  3267. passthrough" mode are also likely candidates.  I would be interested in any
  3268. results of trying Kermit with such emulators, whether positive or negative
  3269. (be sure to SET CONTROLLER GRAPH to the mainframe Kermit before starting the
  3270. transfer).  Actually, I now have an update that automatically detects
  3271. non-SERIES1-type terminals and also some other updates that may help in
  3272. diagnosing the problems if the existing code should prove not to work "as
  3273. is."
  3274.  
  3275. INFO-KERMIT DIGEST V8 #6                                                Page 57
  3276.  
  3277.  
  3278. ------------------------------
  3279.  
  3280. End of Info-Kermit Digest
  3281.  
  3282. Page 58                                                INFO-KERMIT DIGEST V8 #7
  3283.  
  3284. Info-Kermit Digest         Thu, 1 Sep 1988       Volume 8 : Number 7
  3285.  
  3286. Special C-Kermit Issue:
  3287.                         C-Kermit on the Amiga
  3288.                      C-Kermit 4E(070) on Masscomp
  3289.                    Lock Files Being Ignored by UUCP
  3290.                          C-Kermit and .kermrc
  3291.                      Using Kermit from Shell Mode
  3292.              C-Kermit on an X-MP Running UNICOS Version 3
  3293.                     Performance of Unix Kermit 4E
  3294.                  A Piped Connect Command for C-Kermit
  3295.                       SCO Xenix Makefile Error?
  3296.                     Xenix C-Kermit Makefile Fixes
  3297.                 Parameters for Using Kermit over X.25?
  3298.                      Problem with C-Kermit on VMS
  3299.                         Fixes for VMS C-Kermit
  3300.              C-Kermit 4E(070) Under VAX/VMS Using Parity
  3301.  
  3302. Send digest submissions to Info-Kermit@CUNIXC, requests for addition to or
  3303. deletion from the Info-Kermit subscriber list to Info-Kermit-Request@CUNIXC.
  3304.  
  3305. Kermit files may be obtained over networks and by mail order.  On the
  3306. Internetwork, use FTP to log in to host CUNIXC, CUNIXC.CC.COLUMBIA.EDU, a
  3307. VAX 8700 running UNIX (Ultrix).  The IP host number is 128.59.40.130.  You
  3308. should be able to to FTP to CUNIXC, login as user ANONYMOUS (any password)
  3309. and GET the desired files.  The Kermit files are in directories kermit/a,
  3310. kermit/b, kermit/c, kermit/d, and kermit/e.  You can also get Kermit files
  3311. over BITNET/EARN; to get started send a message with text HELP to KERMSRV,
  3312. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  3313. file k1/aanetw.hlp (AANETW HLP on KERMSRV).  To order by mail, request a
  3314. complete list of Kermit versions and an order form from Kermit Distribution,
  3315. Columbia University Center for Computing Activities, 612 West 115th Street,
  3316. New York, NY 10025 USA.
  3317.  
  3318. ----------------------------------------------------------------------
  3319.  
  3320. Date: Sat, 13 Aug 1988 17:31:22 EDT
  3321. From: Frank da Cruz <fdc@cunixc.cc.columbia.edu>
  3322. Subject: C-Kermit
  3323. Keywords: C-Kermit
  3324.  
  3325. In honor of the first anniversary of the release of C-Kermit 4E, this is a
  3326. special edition of the Info-Kermit digest devoted to C-Kermit.  Some of the
  3327. messages are quite old (but still valid) -- apologies for taking so long to
  3328. get around to them!
  3329.  
  3330. C-Kermit needs a lot of work.  Besides the obvious improvements that
  3331. everyone wants -- sliding windows, a rational script language, etc -- a
  3332. massive internal reorganization is necessary.  As some of you may recall,
  3333. the program originally ran on a VAX with 4.2BSD UNIX.  Then some conditional
  3334. compilation code was added to make it work with ATT System III and V.  Then
  3335. more conditional code was added for Venix, Xenix, etc etc etc, and then many
  3336. non-UNIX systems like Apollo Aegis, Data General AOS/VS, VAX/VMS, the
  3337. Commodore Amiga, the Apple Macintosh, etc etc etc.  As a result, the program
  3338. is a big mess, and the slightest change for one system seems to break it for
  3339. all the other systems.
  3340.  
  3341. INFO-KERMIT DIGEST V8 #7                                                Page 59
  3342.  
  3343.  
  3344. The original of breaking the program up into modules that reflected
  3345. different kinds of dependencies was a good one, but the lines of demarcation
  3346. need to be rethunk.  Only the most basic, primitive, and atomic functions
  3347. should go into the system-dependent files.  How this can best be
  3348. accomplished is still open to question.  For example, the UNIX-specific
  3349. terminal i/o functions file, ckutio.c, is a joke that can only become worse
  3350. with the appearance of new UNIX "standards": X/OPEN and Posix are just two
  3351. more UNIX environments to add to the list -- they don't make all the others
  3352. go away!  Then we have all the "hybrid" systems like Xenix, Ultrix, and
  3353. SUNOS whose characters change with every release (is it Sys V? is it BSD?
  3354. no, it's both!).
  3355.  
  3356. My favorite feature of Unix programming is the way everything is done by
  3357. files -- including header files in your program source, making files
  3358. with magic names appear and disappear to signal other programs what's going
  3359. on, etc etc.  Back in the old days, where there was just one "UNIX", all of
  3360. this was fine.  But now, with so many UNIXes to choose from, and so many
  3361. releases of each one, we find that the major way in which things change is
  3362. where these files are stored and what they are called.  Sometimes it seems
  3363. to me that most of the hair in Unix Kermit involves figuring out where
  3364. "file.h" is on each release of each version of UNIX, and the rest of it is
  3365. devoted to coping with UUCP lock files (where is it? what's it called?
  3366. what's its mode?)...
  3367.  
  3368. I can't offer any good estimate about when the next release of C-Kermit will
  3369. be ready -- "real job" and all that...  But I'm faithfully collecting all
  3370. the bug reports, fixes, and suggestions in hopes that a chunk of time will
  3371. appear some day.  The file is about 2 feet thick so far.  A sampling appears
  3372. below.
  3373.  
  3374. ------------------------------
  3375.  
  3376. Date: Tue, 15 Mar 88 17:33:10 PST
  3377. From: RCKG01M%CALSTATE.BITNET@CUVMA.COLUMBIA.EDU  (Stephen Walton)
  3378. Subject: C-Kermit on the Amiga
  3379. Keywords: Amiga, Commodore Amiga, C-Kermit/Amiga
  3380.  
  3381. I have a working version of C Kermit 4E(070) on my Amiga using Manx Aztec
  3382. C Version 3.6a (latest one).  The diffs from the old version are minimal,
  3383. but I did put in the parity stuff.  It seems to work fine, though I've
  3384. not tested it 100%--I can't get a completely transparent 8-bit link to
  3385. a computer here, and transfers with even and mark parity time out, but
  3386. ones with odd or space or no parity go just fine.  I have not tried to
  3387. compile it with any version of Lattice C.
  3388.  
  3389. I also ran Amiga Lint over it;  this Lint is very picky, but did find
  3390. a few things worth mentioning in the main sources:
  3391.  
  3392. (1) In ckcmai.c, there are two printf() calls where the number of
  3393.     items to print doesn't match the number of formatting codes.
  3394. (2) Several items, such as ttname[], are declared as CHAR in some places
  3395.     and as char in others.  Since the former is typedef'd to "unsigned
  3396.     char" for most systems, there are potential problems here.
  3397. (3) The extern item speed will have to be a long if C Kermit is to
  3398.     be ported to systems where int=16 bits.  It is an int right now.
  3399.  
  3400. Page 60                                                INFO-KERMIT DIGEST V8 #7
  3401.  
  3402.     This failing requires compiling Amiga Kermit with int set to 32
  3403.     bits, which increases the executable size by 1/3 (from 63 to 85K).
  3404.  
  3405. If you're interested in the files, let me know and I'll email context
  3406. diffs and a BOO file.
  3407.  
  3408. Stephen Walton, formerly of Ametek but now at Cal State Northridge
  3409. SWALTON@SOLAR.STANFORD.EDU (preferred) or RCKG01M@CALSTATE.BITNET
  3410.  
  3411. [Ed. - Thanks, Steve!  For now, the updated Amiga Kermit is in a "shell
  3412. archive" as CKIKER.UPD (it's been there since April)!  Your comments 1-3 are
  3413. noted and will be filtered in to the next release.]
  3414.  
  3415. ------------------------------
  3416.  
  3417. Date: Wed, 2 Mar 88 02:58:11 CST
  3418. From: Stan Barber <sob@soma.neuro.bcm.tmc.edu>
  3419. Subject: C-KERMIT 4E(070) on Masscomp
  3420. Keywords: C-Kermit 4E, Masscomp, RTU, BSD 2.10
  3421.  
  3422. The latest release of C-KERMIT will run on the MASSCOMP MC5XXX computers
  3423. (lockfile problems excluded!) with the application of the following diff to
  3424. the makefile (ckuker.mak).
  3425.  
  3426. I have also checked the most recent version of kermit out on BSD2.10 and here
  3427. are the recommended changes. (version is C-Kermit, 4E(070) 29 Jan 88)
  3428.  
  3429. [Ed. - Diffs omitted, but added to CKUKER.BWR.  Thanks, Stan!]
  3430.  
  3431. ------------------------------
  3432.  
  3433. Date: 8 Mar 88 14:56:06 EST (Tue)
  3434. From: mancol!jh@cmcl2.NYU.EDU (John Hanley)
  3435. Subject: Lock Files Being Ignored by UUCP
  3436. Keywords: Lock Files, UUCP Lock Files, C-Kermit
  3437.  
  3438. We had a problem with the lock file that kermit creates not being publicly
  3439. readable, so that uucico's started by uucp couldn't read the lock file and
  3440. therefore didn't think the port was locked.  It turns out that
  3441. creat(lockfile,0444) on our 3b2/400 (SVR3) does NOT create a -r--r--r--
  3442. file, but rather -r--------.  One workaround we used for a while was to
  3443. invoke kermit using a script that sets umask to 0, runs kermit, and then
  3444. restores umask.  Eventually I patched ttlock() right after the creat() --
  3445. context diffs enclosed below.  I have no idea how widespread this problem is
  3446. (or if it even exists beyond our own system).
  3447.  
  3448.    --John Hanley
  3449.      System Programmer, Manhattan College
  3450.      soon to be a former E.E./CS undergrad (got a job for me? :)
  3451.      ..!cmcl2.nyu.edu!manhat!jh
  3452.  
  3453. [Ed. - This problem is not specific to Sys V.  The mode (protection)
  3454. specified in the creat() call is combined by UNIX with the user's "umask".
  3455. Therefore, any user who does not like to create files that are publicly
  3456. readable will also create UUCP lockfiles that are not publicly readable.
  3457. Solution: after creating the lock file, change the mode:
  3458.  
  3459. INFO-KERMIT DIGEST V8 #7                                                Page 61
  3460.  
  3461.  
  3462.     lck_fil = creat(flfnam, 0444); <-- should really be 0644, right?
  3463.     if (lck_fil < 0) return(-1);
  3464.     fchmod(lck_fil, 0644);         <-- new
  3465.  
  3466. Actually, this will only work if the lockfile didn't exist already.  If it
  3467. did, then the mode won't change.  This is a feature of UNIX.  Anybody have a
  3468. better idea?]
  3469.  
  3470. ------------------------------
  3471.  
  3472. Date: Thu, 17 Mar 88 18:00:08 +0200
  3473. From: Ze'ev (Steed) Shtadler <steed%TECHUNIX.BITNET@CUVMA.COLUMBIA.EDU>
  3474. Subject: C-Kermit and .kermrc
  3475. Keywords: C-Kermit 4E(070)
  3476.  
  3477. I use C-kermit 4E(070) on 4.3BSD Unix.
  3478.  
  3479. I tried to use a .kermrc init file in one of my subdirectories to override
  3480. some setting done by $HOME/.kermrc but it didn't work.  It seems that kermit
  3481. doesn't look for .kermrc in the working directory if it finds $HOME/.kermrc.
  3482. This is also said in the manual, but I didn't see it even after looking
  3483. several times.
  3484.  
  3485. I would like to ask and suggest that an attempt to "take" .kermrc in the
  3486. working directory will be done even if $HOME/.kermrc was found.  This will
  3487. simplify things for me (and hopefully for others) and will be in line with
  3488. other programs like vi.  The obvious fix attempt: put "take .kermrc" in
  3489. $HOME/.kermrc will cause an infinite loop when kermit is called from the
  3490. home directory.
  3491.  
  3492.                                         Ze'ev.
  3493. Ze'ev Shtadler.
  3494. Computer Science Faculty,
  3495. Technion, Israel Institute of Thechnology,
  3496. Haifa Israel
  3497. Phone (work):   (+972) 4-293849
  3498.  
  3499. BITNET:  steed@techunix
  3500. Domain:  steed@techunix.technion.ac.il
  3501. UUCP:    ...!psuvax1!techunix.bitnet!steed
  3502. ARPANET: steed%techunix.bitnet@cunyvm.cuny.edu
  3503. CSNET:   steed%techunix.bitnet@csnet-relay.csnet
  3504.  
  3505. [Ed. - This is an interesting idea.  But it could produce some unwanted side
  3506. effects.  Anybody who cd'd to somebody else's directory would automatically
  3507. and unavoidably be affected by that person's .kermrc.  Better to leave
  3508. things as they are, and use explicity 'take' or 'set' commands to override
  3509. your $HOME/.kermrc environment.]
  3510.  
  3511. ------------------------------
  3512.  
  3513. Date: 21 Mar 88 21:12:04 GMT
  3514. From: jeff@colgate.CSNET (Jeff Norden)
  3515. Subject: Using Kermit from Shell Mode
  3516. Keywords: GNU EMACS, Shell, C-Kermit
  3517.  
  3518. Page 62                                                INFO-KERMIT DIGEST V8 #7
  3519.  
  3520.  
  3521. I'm not sure, but I think somebody asked about running kermit under shell
  3522. mode a while ago.  Anyway, here is some code that I find useful.  The result
  3523. is that I can log onto machines with primitive operating systems (VMS and
  3524. ATT system V :-), and still have the features of shell-mode available for
  3525. command history, etc.  It's also handy to be able to run a file transfer in
  3526. an emacs window.  The transfer is in the "background", but you can also
  3527. monitor or stop it easily.
  3528.  
  3529. [Ed. - Thanks for the idea, Jeff.  The entire text of this long message,
  3530. including the GNU EMACS code, have been put with the C-Kermit files, as
  3531. KER:CKUGNU.MSG.]
  3532.  
  3533. ------------------------------
  3534.  
  3535. Date: Tue, 7 Jun 88 12:01:20 mdt
  3536. From: Jim Ivey <isew01!miv@uiucdcs.UUCP>
  3537. Subject: C-Kermit on an X-MP Running UNICOS Version 3
  3538. Keywords: C-Kermit, Cray X-MP, UNICOS
  3539.  
  3540. I have been attempting to bring up C-Kermit on our local X-MP running UNICOS
  3541. Version 3.  Unfortunately, it is neither System V nor BSD4.2, but rather a
  3542. combination to purportedly "bring you the best of both worlds."  It is
  3543. closest to System V, and when I compile it using "make sys5", it compiles
  3544. fine, producing "wermit."  However when I try transferring files there are
  3545. numerous Timeout errors, and the transfer rate is usually about 10-30 baud.
  3546. Yes, 10 baud.  Now obviously this isn't because of the slowness of the
  3547. machine.  So far I haven't been able to find the answer, though I did make a
  3548. change in the routine msleep (m).  In it, there was a constant called
  3549. CLOCK_TICK.  Normally for UXIII, it is set for 17 milliseconds per clock
  3550. tick.  However, on the Cray X-MP there are 8.5 nanoseconds per clock tick.
  3551. Therefore this conversion comes out to "#define CLOCK_TICK .0000085".  Now,
  3552. when I extracted the msleep code from the program and tested it out, this
  3553. modification worked and the old code didn't.  Unfortunately, this didn't
  3554. solve the problem and I'm still on square one.
  3555.  
  3556. Do you know if anyone has already tried to do this?  If you could give me
  3557. some help on this I would appreciate it greatly.  Also, any help would
  3558. probably have to have some explanation, because I am very new to Kermit,
  3559. though I know UNIX and UNICOS well.
  3560.  
  3561. As a side note, is there any way to change the packet size to something
  3562. above 94?  We have all 9600 lines, and with the I/O speed of the Cray, it
  3563. seems like (after we get the program working in the first place) we could
  3564. get more performance out of the larger packets.
  3565.  
  3566. Anyways, thanks ahead of time for any help you can give me.  I plan on
  3567. coming in late tonight to uucp the latest version of C-Kermit in the hopes
  3568. that that will help a little.
  3569.  
  3570. Jim Ivey                         UUCP: ...!ucbvax!ucdavis!egg-id!miv
  3571. Idaho National Engineering Lab.  INTERNET: miv@egg-id.INEL.GOV
  3572. POB 1625 M.S. 2603               Phone: (208) 526-9066
  3573. Idaho Falls, Id. 83415           FAX:
  3574.  
  3575. [Ed. - I hope the latest version helped.  It solves the second problem you
  3576.  
  3577. INFO-KERMIT DIGEST V8 #7                                                Page 63
  3578.  
  3579. mention because it can handle packets up to 1K long (or as long as you want
  3580. if you change the symbols MAXSP and MAXRP, defined in ckuker.h.)]
  3581.  
  3582. ------------------------------
  3583.  
  3584. Date: Mon, 11 Jul 88 12:57:07 CDT
  3585. From: rutgers!cs.utexas.edu!execu!keith@columbia.edu (Keith Pyle)
  3586. Subject: Performance of Unix Kermit 4E
  3587. Keywords: C-Kermit 4E(070) Performance
  3588.  
  3589. I have recently attempted to upgrade Kermit from version 4D to 4E on a Prime
  3590. EXL 316 (80386) running System V Release 3 Version 2.1.  While I encountered
  3591. no difficulities in building 4E, I noticed an obvious difference in
  3592. performance as compared to 4D.  Building Kermit without DEBUG and TLOG made
  3593. no difference.  For a file of 67429 bytes, the following times (sec.) and
  3594. effective baud rates were noted:
  3595.  
  3596.                                 time        rate
  3597. Kermit 4D with DEBUG and TLOG    119        5660
  3598. Kermit 4E with DEBUG and TLOG    268        2516
  3599. Kermit 4E without DEBUG and TLOG 268        2516
  3600.  
  3601. The EXL had no other users during these tests and was communicating with
  3602. an IBM PC XT running MS Kermit 2.31 over a direct 9600 baud line using
  3603. 8 bits, no parity.  I subsequently rebuilt both 4D and 4E with profiling
  3604. enabled.  Following is an extract of the prof(1) output for each using
  3605. the same test file as above:
  3606.  
  3607. Kermit 4D:
  3608.  %Time Seconds Cumsecs  #Calls   msec/call  Name
  3609.   46.4    7.28    7.28    1040      7.000   read
  3610.    7.7    1.22    8.50   72827      0.0167  encode
  3611.    6.7    1.05    9.55     834      1.26    write
  3612.    6.1    0.95   10.50   69179      0.0137  getchx
  3613.    5.0    0.79   11.28   67430      0.0117  zchin
  3614.    4.3    0.68   11.96     799      0.85    getpkt
  3615.    3.2    0.51   12.47   67432      0.0076  chkfn
  3616.    3.2    0.51   12.98                      _mcount
  3617.    2.9    0.46   13.44     801      0.57    spack
  3618.    2.2    0.34   13.78   77202      0.0044  dopar
  3619.    1.6    0.26   14.04                      getuid
  3620.    1.6    0.25   14.29    1602      0.156   chk2
  3621.    1.3    0.21   14.50     827      0.25    ioctl
  3622.    1.1    0.18   14.67                      monitor
  3623.    1.0    0.16   14.83                      sighold
  3624.  
  3625. Kermit 4E:
  3626.  %Time Seconds Cumsecs  #Calls   msec/call  Name
  3627.   46.6    6.62    6.62     837      7.91    write
  3628.   42.8    6.08   12.70     884      6.88    read
  3629.    3.8    0.54   13.24    1602      0.337   chk2
  3630.    0.9    0.13   13.37     801      0.16    spack
  3631.    0.6    0.09   13.46    1613      0.056   strlen
  3632.    0.6    0.09   13.55   77206      0.0012  dopar
  3633.    0.6    0.08   13.63                      _mcount
  3634.    0.6    0.08   13.71                      sighold
  3635.  
  3636. Page 64                                                INFO-KERMIT DIGEST V8 #7
  3637.  
  3638.  
  3639. Note that write calls consume more than 5 times as much of the profiled
  3640. time in version 4E.  Has any similar behaviour been noted on other versions
  3641. of Kermit? (I have tested 4D vs. 4E on a Sun and found no appreciable
  3642. difference.)  Any suggestions on what changes leading to 4E might be
  3643. responsible?
  3644.  
  3645. Keith Pyle          ...!tut.cis.ohio-state.edu!cs.utexas.edu!execu!keith
  3646.  
  3647. [Ed. - This is a very interesting report.  Both these C-Kermit versions were
  3648. developed on a VAX (with Ultrix and BSD), and also tested on some other
  3649. systems (ATT 3B20, PDP-11, etc).  One of the major goals for 4E was improved
  3650. performance.  Benchmarks showed a slight improvement when sending files
  3651. (about 10%), a large improvement when receiving (something like 40%), and a
  3652. dramatic improvement when receiving with parity (greater than 400%) -- but
  3653. this is on BSD-type systems, where the fundamental method used for packet
  3654. input was changed (the method used in ATT-type systems was not significantly
  3655. changed, but some improvements were attempted).  Little work was done on
  3656. packet output (write) from 4D to 4E, so it's hard to know how to explain the
  3657. observed behavior.  Has anyone observed a similar performance degradation on
  3658. other System-V based systems?  If so, then the System-V specific code must
  3659. be at fault; otherwise, it's something to do with the particular machine,
  3660. OS, or compiler.]
  3661.  
  3662. ------------------------------
  3663.  
  3664. Date: Mon, 13 Jun 88 15:04:34 PDT
  3665. From: Marco Franzen <franzen@dbsinf6.bitnet>
  3666. Subject: A Piped Connect Command for C-Kermit
  3667. Keywords: C-Kermit 4E(070)
  3668.  
  3669.    Date: Mon, 6 Jun 88 15:04:34 PDT
  3670.    From: farhad%lyra.usc.edu@oberon.USC.EDU (Farhad Arbab)
  3671.    Subject: Proposal for a New Command
  3672.  
  3673.    What I need is a new command, say "pipe", which works just as connect,
  3674.    but instead of connecting the line to /dev/tty, it connects the line to
  3675.    the standard input and output of a given process: "pipe L" will then do
  3676.    what I want.  When L terminates, I want to return to the command mode
  3677.    of kermit.  L can be an arbitrary program (anything that can be
  3678.    executed via the ! escape).
  3679.  
  3680. I have implemented exactly this command a couple of months ago, I named it
  3681. "|".  It used some code of the connect command and some of the "!" escape.
  3682. We use it to transfer files with the Zmodem protocol via the sz/rz program.
  3683. That protocol is more efficient on low-noise lines if either side's Kermit
  3684. cannot handle extended-length packets.
  3685.  
  3686.    How easy or difficult is it to add this feature to kermit?  Do you know
  3687.    of other easy ways to accomplish the same effect?  Is there another
  3688.    communications program that provides such a feature?
  3689.  
  3690. To simply add this command to the Unix version of C-Kermit was not a big
  3691. task.  Most of all is error-handling ("You have to set line/speed first"
  3692. messages,...).  For other OSes I can't say.
  3693.  
  3694.  
  3695. INFO-KERMIT DIGEST V8 #7                                                Page 65
  3696.  
  3697. The only problem on my Version of Unix is with the line-locking ioctl: After
  3698. calling the new "|" command, the hard lock vanishes.  But this malfunction
  3699. doesn't hurt me, since the uucp-style (/usr/spool/uucp/LCK..line) locking
  3700. still works.
  3701.  
  3702. The reason for this misbehavior is that the dup2() (and also the dup() and
  3703. fcntl(F_DUPFD)) system call releases the lock.  Neither [v]fork() nor
  3704. execlp() act this way. However, I don't know, whether this is an error of my
  3705. Operating System.
  3706.  
  3707. Marco Franzen            | Phone:  +49 531 391 3245
  3708. Inst. f. Betriebssysteme | Fax:    +49 531 391 4577
  3709.     und Rechnerverbund   | BITNET: franzen@dbsinf6.bitnet  (preferred)
  3710. TU Braunschweig          | UUCP:   franzen@infbs.uucp (uunet!infbs!franzen)
  3711. D-3300 Braunschweig      | from CSnet:  franzen%dbsinf6.bitnet@Germany.CSnet
  3712. Fed. Rep. of Germany     | from ARPA:   franzen%dbsinf6.bitnet@psuvax1.psu.edu
  3713.  
  3714. [Ed. - This idea will be looked at for the next release.  It could be used
  3715. to build a very generalized script facility to be used with C-Kermit, but
  3716. from the outside.]
  3717.  
  3718. ------------------------------
  3719.  
  3720. Date: Tue Mar 15 20:51:01 1988
  3721. From: yunexus!ists!mike@uunet.UU.NET (Mike Clarkson)
  3722. Subject: SCO Xenix Makefile Error?
  3723. Keywords: SCO Xenix
  3724.  
  3725. For Ckermit 4E(070) 29 Jan 88, in the Makefile
  3726.  
  3727.     # CKUKER.MAK, Version 2.11, 29 January 1988
  3728.     #
  3729.     # -- Makefile to build C-Kermit for Unix and Unix-like systems --
  3730.     ...
  3731.     #SCO Xenix/286 2.2.1, e.g. for IBM PC/AT, PS/2 Model 50, etc.
  3732.     sco286:
  3733.     make wermit "CFLAGS= -DXENIX -DUXIII -DDEBUG -DTLOG -F 3000 -i -M2le" \
  3734.         "LNKFLAGS = -F 3000 -i -M2le"
  3735.  
  3736. looks OK, but
  3737.  
  3738.     #SCO Xenix 2.2.1 for IBM PC, XT, PS2/30, or other 8088 or 8086 machine
  3739.     sco86:
  3740.     make wermit "CFLAGS= -DXENIX -DUXIII -DDEBUG -DTLOG -F 3000 -i -Mle" \
  3741.         "LNKFLAGS = -F 3000 -i -Mle"
  3742.  
  3743. looks suspicious:
  3744.  
  3745.     1) SCO Xenix doesn't have 2.2.1 for the 8086.  Their latest (and last?)
  3746. release for the 8086 is 2.1.3.
  3747.  
  3748.     2) SCO Xenix doesn't have large model (-Ml) for the 8086.
  3749.  
  3750.  
  3751. Has this makefile actually been tested?  Last time I tried, it wouldn't work
  3752. medium model.
  3753.  
  3754. Page 66                                                INFO-KERMIT DIGEST V8 #7
  3755.  
  3756.  
  3757. Mike Clarkson                        mike@ists.UUCP
  3758. Institute for Space and Terrestrial Science
  3759. York University, North York, Ontario,
  3760. CANADA M3J 1P3                        (416) 736-5611
  3761.  
  3762. [Ed. - Hasn't been tested at Columbia.  We don't have any Xenix systems
  3763. handy to test things on.  It was sent in by somebody.  See below for a
  3764. different version that was sent in by somebody else.]
  3765.  
  3766. ------------------------------
  3767.  
  3768. Date: 15 Jul 88 17:43:17 EDT (Fri)
  3769. From: uunet!jetson!john@rutgers.edu (John Owens)
  3770. Subject: Xenix C-Kermit Makefile Fixes
  3771. Keywords: SCO Xenix
  3772.  
  3773. There are some changes to the makefile I made for Xenix.  I added support
  3774. for 386 Xenix, and changed the model for 86 xenix from large to medium.
  3775. (Large model doesn't run on the 8086 Xenixes.)
  3776.  
  3777. # CKUKER.MAK, Version 2.11, 29 January 1988
  3778.  . . .
  3779. # for SCO Xenix/286 2.2.1 with development system 2.2 on 80286, "make sco286"
  3780. # for SCO Xenix/86  2.2.1 with development system 2.2 on 8086/8 "make sco86"
  3781. # for SCO Xenix/386 2.2.2 with development system 2.2 on 80386, "make sco386"
  3782.  . . .
  3783. #SCO Xenix/286 2.2.1, e.g. for IBM PC/AT, PS/2 Model 50, etc.
  3784. sco286:
  3785.     make wermit "CFLAGS= -DXENIX -DUXIII -DDEBUG -DTLOG -F 3000 -i -M2le" \
  3786.         "LNKFLAGS = -F 3000 -i -M2le"
  3787.  
  3788. #SCO Xenix 2.2.1 for IBM PC, XT, PS2/30, or other 8088 or 8086 machine
  3789. sco86:
  3790.     make wermit "CFLAGS= -DXENIX -DUXIII -DDEBUG -DTLOG -F 3000 -i -M0me" \
  3791.         "LNKFLAGS = -F 3000 -i -M0me"
  3792.  
  3793. #SCO Xenix/386 2.2.2
  3794. sco386:
  3795.     make wermit "CFLAGS= -DXENIX -DUXIII -DDEBUG -DTLOG -Otcl -i -M3e" \
  3796.         "LNKFLAGS = -i"
  3797.  . . .
  3798.  
  3799. I also changed cku[ft]io.c to print Xenix/86, Xenix/286, or Xenix/386,
  3800. instead of always printing Xenix/286:
  3801.  
  3802. [Ed. - Diffs omitted, but added to the CKUKER.BWR file.  Thanks!]
  3803.  
  3804. John Owens        SMART HOUSE Limited Partnership
  3805. john@jetson.UUCP    (old uucp) uunet!jetson!john
  3806. +1 301 249 6000        (internet) john%jetson.uucp@uunet.uu.net
  3807.  
  3808. ------------------------------
  3809.  
  3810. Date: 16 Jun 88 18:43:23 GMT
  3811. From: unido!IAO.FhG!news@uunet.UU.NET (News user)
  3812.  
  3813. INFO-KERMIT DIGEST V8 #7                                                Page 67
  3814.  
  3815. Subject: Parameters for Using Kermit over X.25?
  3816. Keywords: X.25, C-Kermit
  3817.  
  3818. I would like to transfer files from a VAX running VAX-Kermit (Don't know
  3819. which version) to a UNIX-VAX running C-Kermit.  I connect to the VMS Machine
  3820. via our external PAD and get into the VMS-Machine via their X.25 Interface
  3821. (PSI). For normal operation, that works like it should, but when I want to
  3822. copy files from the VMS to UNIX, nothing works. The Kermit on the UNIX side
  3823. prints some of its messages (nothing understandable) and exits because he
  3824. couldn't speak to the other machine (timeout).  Can someone give me please
  3825. some clues which parameters (Kermit and X.25) to tweak how.
  3826.  
  3827. Thanks in advance.
  3828.  
  3829. Dieter Woerz
  3830. Fraunhofer Institut fuer Arbeitswirtschaft und Organisation
  3831. Abt. 453
  3832. Holzgartenstrasse 17
  3833. D-7000 Stuttgart 1
  3834. W-Germany
  3835.  
  3836. BITNET: iaoobel.uucp!woerz@unido.bitnet
  3837. UUCP:   ...{uunet!unido, pyramid}!iaoobel!woerz
  3838.  
  3839. [Ed. - Beyond our stock answer of "set parity even" (or mark, or odd, or
  3840. space) to the Kermits on each end, there's not much else to say.  It is very
  3841. likely that VMS Kermit (any version -- Kermit-32 or C-Kermit/VMS) must have
  3842. explicit VAX-PSI support written into it in order to work over a direct X.25
  3843. connection (as opposed to an X.3/X.28 connection through a serial port).  In
  3844. fact, there seems to be a big demand for this in Europe.  Can anyone shed
  3845. more light?]
  3846.  
  3847. ------------------------------
  3848.  
  3849. Date: Tue, 8 Mar 88 00:13:01 est
  3850. From: Brendan Reilly <reilly@aqua.whoi.edu>
  3851. Site: Woods Hole Oceanographic Institution, Woods Hole Mass.
  3852. Subject: Problem with C-Kermit on VMS
  3853. Keywords: C-Kermit/VMS, VAX/VMS
  3854.  
  3855. After compiling KERMIT with the latest CC from Digital it appears that both
  3856. KERMIT and VMS/CC have a procedure named system.  I've changed the KERMIT
  3857. reference to vms_system, but I'm still having a problem with server mode.
  3858. When talking to the VMS side in server mode from a PC running the latest
  3859. KERMIT and I issue a REMOTE CWD [reilly.a] I'm asked for a password, to
  3860. which I reply with a return and then see some garbage characters typed on
  3861. the screen.  If I then do a REMOTE DIR it's a surprise to find that I'm
  3862. still in the original directory - not [REILLY.A] Has anyone else seen this?
  3863.  
  3864. [Ed. - There's definitely some kind of bug in VMS C-Kermit in this area.
  3865. When I try this, the VMS C-Kermit server goes into an infinite unstoppable
  3866. loop printing messages issued by its "ttinl()" function: "SYS$QIOW:
  3867. %SYSTEM-F-EXQUOTA, exceeded quota", and "%NONAME-W-NOMSG, Message number 0".
  3868. Another one for the bug list.]
  3869.  
  3870. ------------------------------
  3871.  
  3872. Page 68                                                INFO-KERMIT DIGEST V8 #7
  3873.  
  3874.  
  3875. Date: Tue, 19 Jul 88 13:54 CDT
  3876. From: <ARCHERB%UMKCVAX1.BITNET@CUVMB.CC.COLUMBIA.EDU> (BARRY ARCHER)
  3877. Subject: Fixes for VMS C-Kermit
  3878. Keywords: C-Kermit/VMS, VAX/VMS
  3879.  
  3880. In working with the VMS version of C-Kermit, I became frustrated with the
  3881. problems with hanging subprocesses after exiting the image, as well as the
  3882. problems I was having running C-Kermit while connected to a DECserver ( LAT
  3883. protocol ).  I finally figured out how to fix those problems and installed
  3884. the new version.
  3885.  
  3886. [Ed. - The remainder of this (long) message, plus the new ckv*.c files, have
  3887. been added to CKVKER.BWR for now.  Hopefully the various people working on
  3888. C-Kermit for VMS will be able to reconcile their changes and improvements.
  3889. Thanks, Barry!]
  3890.  
  3891. ------------------------------
  3892.  
  3893. Date: Wed, 08 Jun 88 10:04:12 EST
  3894. From: PIQUE%UCONNVM.BITNET@MITVMA.MIT.EDU
  3895. Subject: C-Kermit 4E(070) Under VAX/VMS Using Parity
  3896. Keywords: C-Kermit/VMS, VAX/VMS
  3897.  
  3898. I compiled C-Kermit 4E(070) under VAX/VMS, using version 1.0(013) of the VMS
  3899. tty I/O module, CKVTIO.C.  The fixes present to make it work properly when
  3900. parity is being used didn't work properly because ttprty was never set in
  3901. CKVTIO.C.
  3902.  
  3903. Here's a partial fix: [Ed. - Omitted, but added to KER:CKVKER.BWR.]
  3904.  
  3905. This doesn't fix things completely since the call to SYS$QIOW in ttinl()
  3906. waiting for an end-of-line character waits for an end-of-line character with
  3907. high bit 0, but if (say) end-of-line is 13 and parity is even it should
  3908. really be waiting for a character with high bit set (X'8D').  However you
  3909. can work around this by choosing the end-of-line character so that it is
  3910. sent with high bit reset (e.g., eol=15 for even parity).
  3911.  
  3912. Another problem is that data overrun errors occur periodically during file
  3913. transfer and (at times) during terminal emulation.  Kermit will also
  3914. sometimes lock up completely during terminal emulation.  I do not know what
  3915. is causing this.  We are running on a MicroVax II with BYTLM=20480.
  3916.  
  3917. David Moews                 PIQUE@UCONNVM.BITNET
  3918.  
  3919. ------------------------------
  3920.  
  3921. End of Info-Kermit Digest
  3922.  
  3923. INFO-KERMIT DIGEST V8 #8                                                Page 69
  3924.  
  3925. Info-Kermit Digest         Tue, 13 Sep 1988       Volume 8 : Number 8
  3926.  
  3927. Departments:
  3928.  
  3929.   ANNOUNCEMENTS -
  3930.         Submission of new version of Kermit-UCSD  (UCPECAN)
  3931.  
  3932.   MS-DOS KERMIT -
  3933.         MS-Kermit 2.31 Quirks
  3934.         Re: MS-Kermit v2.31 Eats Nulls
  3935.         MS-Kermit and the Novell File Server
  3936.         Re: MS-Kermit 2.31 Tabs Bug (Feature?)
  3937.         MS-Kermit: Five Bugs and a Whinge
  3938.         Goto Label Bug in 2.31 August 15?
  3939.         Problems with MS-Kermit 2.31 Graphics
  3940.         MS Kermit 2.31 File Dates
  3941.  
  3942.   MISCELLANY -
  3943.         Kermit Through Protocol Converters Under CMS
  3944.  
  3945. Send digest submissions to Info-Kermit@CUNIXC, requests for addition to or
  3946. deletion from the Info-Kermit subscriber list to Info-Kermit-Request@CUNIXC.
  3947.  
  3948. Kermit files may be obtained over networks and by mail order.  On the
  3949. Internetwork, use FTP to log in to host CUNIXC, CUNIXC.CC.COLUMBIA.EDU, a
  3950. VAX 8700 running UNIX (Ultrix).  The IP host number is 128.59.40.130.  You
  3951. should be able to to FTP to CUNIXC, login as user ANONYMOUS (any password)
  3952. and GET the desired files.  The Kermit files are in directories kermit/a,
  3953. kermit/b, kermit/c, kermit/d, and kermit/e.  You can also get Kermit files
  3954. over BITNET/EARN; to get started send a message with text HELP to KERMSRV,
  3955. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  3956. file k1/aanetw.hlp (AANETW HLP on KERMSRV).  To order by mail, request a
  3957. complete list of Kermit versions and an order form from Kermit Distribution,
  3958. Columbia University Center for Computing Activities, 612 West 115th Street,
  3959. New York, NY 10025 USA.
  3960.  
  3961. ----------------------------------------------------------------------
  3962.  
  3963. Date: Sat Aug 20 08:10:52 1988
  3964. From: portal!cup.portal.com!R_Tim_Coslet@sun.com
  3965. Subject: Submission of new version of Kermit-UCSD  (UCPECAN)
  3966. Keywords: UCSD Pascal Kermit, p-System, Pecan
  3967.  
  3968. I am submiting my version of Kermit-UCSD (Version 1.0, 17 Aug 88), currently
  3969. running on the Atari MEGA ST2 with the Pecan Software Systems Inc p-System,
  3970. and adaptable to Version IV p-Systems on other machines.
  3971.  
  3972. You should recieve 2 files: 1) a large (about 3300 lines) one containing the
  3973. source (UCPECA.PAS) and a smaller (about 300 lines) one containing the
  3974. documentation (UCPECA.DOC).
  3975.  
  3976. This version does not have everything I was planning on implementing and has
  3977. a few known bugs (listed in the .DOC file), but it does work (and is MUCH
  3978. better than any of the other versions of Kermit-UCSD I could find currently
  3979. in Kermit Distribution).
  3980.  
  3981.  
  3982. Page 70                                                INFO-KERMIT DIGEST V8 #8
  3983.  
  3984. I hope to be able to provide an upgrade on this within about 6 months to one
  3985. Year.
  3986.  
  3987. I also hope to be able to send an encoded KERMIT.CODE file and a decoding
  3988. program, containing the entire program and implementation parts of the Pecan
  3989. Library Units, for people with Version IV p-Systems that can't afford or
  3990. don't need to purchase Pecan's Library.  They will still have to go to USUS
  3991. to get a REMUNIT for their system though.  The Pecan Licence permits this for
  3992. "non-commercial purposes" (but I am still checking with them to make sure
  3993. that they are satisified with my attachment of their copyright notice, so
  3994. there will be no problems).
  3995.                                         R. Tim Coslet
  3996.  
  3997. Usenet: R_Tim_Coslet@cup.portal.com
  3998. BIX:    r.tim_coslet
  3999.  
  4000. [Ed. - Many thanks, R. Tim!  Those who are interested in bringing up this
  4001. program on other UCSD p-Systems besides the Atari are encouraged to do so,
  4002. and report back the results.  The .DOC file tells how to reach Pecan and
  4003. USUS.  The files are in kermit/c/ucpeca.* on CUNIXC, and in KER:UCPECA.* on
  4004. CU20B, and also available as UCPECA * from KERMSRV at CUVMA, and also can be
  4005. ordered on Kermit Tape C from Columbia.]
  4006.  
  4007. ------------------------------
  4008.  
  4009. Date: Wed, 17 Aug 88 10:30 EDT
  4010. From: <RWMIRA01%ULKYVX.BITNET@cuvmb.cc.columbia.edu>
  4011. Subject: MS-Kermit 2.31 Quirks
  4012. Keywords: MS-DOS Kermit 2.31
  4013.  
  4014. Thanks, Joe, Kermit 2.31 is wonderful.  I have noticed some minor problems
  4015. with 2.31 though.  First the set up:
  4016.  
  4017.   IBM-PS/2 Model 30, DOS 3.3, DEVICE=ANSI.SYS, FILES=20, BUFFERS=20
  4018.   Kermit 2.31
  4019.  
  4020. We have an IBM 3081 that we connect to via a 7171.  Our script file takes us
  4021. to the opening screen and leave us there.  The last 2 lines of our script
  4022. file are:
  4023.  
  4024.   TAKE IBM.INI    ; Our keybindings file
  4025.   CONNECT
  4026.  
  4027. The first problem has to do with connect.  After a script executes and it
  4028. does a connect it starts with a clear screen.  Apparently any data arriving
  4029. while we are defining keys is being stored somewhere besides the screen
  4030. buffer.  When it connects we get a clear screen.  To circumvent this
  4031. problem, the script file was changed to read:
  4032.  
  4033.   TAKE IBM.INI    ; Our keybindings file
  4034.   OUTPUT \7       ; ^G causes the 7171 to repaint the screen and make sure
  4035.                   ; that the keypad is in application mode.
  4036.   CONNECT
  4037.  
  4038. In 2.30 this did not seem like a problem.
  4039.  
  4040.  
  4041. INFO-KERMIT DIGEST V8 #8                                                Page 71
  4042.  
  4043. The second problem is really just an annoyance.  If you have set your screen to
  4044. a non-white on black and you get into kermit and take a script file, sometime
  4045. between the last input/output command and connect command the screen flickers
  4046. white on black and then after the connect command it is ok.  (Probably the
  4047. output \7 causes the 7171 to send an escape sequence to Kermit and it
  4048. reinstates the colors.  Up to this point, all local kermit i/o was in the
  4049. non-white color.  Upon exit of the script (ALT-X from connect) the local
  4050. kermit i/o is in white and any subsequent scripts/connects etc. are in white.
  4051. Sometimes if I immeditaly execute another script file the connect mode is in
  4052. non-white, but sometimes it isn't.  Regardless, when I return to DOS, it is
  4053. back to white again.  Like I said, not a problem just annoying.  It may be
  4054. tied to the first problem.
  4055.  
  4056. The third problem is a bit more severe, but may not be a Kermit problem as
  4057. much as a PS/2 model 30 problem, however I have only noticed it in Kermit.
  4058. In the unlikely event that I have CTRL, SHIFT down and I press Left arrow and
  4059. mis-hit and probably hit several keys with my short stubby fingers, all of the
  4060. lights flash on the keyboard and it locks up.  I am thinking that it is a BIOS
  4061. problem since it happened last night when I was in Turbo Pascal, but it didn't
  4062. lock the keyboard.  I don't know what keys I hit to cause it, but my left hand
  4063. was in the vacinity of the LEFT shift, LEFT CTRL area and my right hand was in
  4064. the area of the editing keypad.
  4065.  
  4066. One other note and question.  I was trying to transfer some stuff between my
  4067. PS/2 and my AT&T PC6300 at 115600 baud.  The AT&T could not handle it.  The
  4068. machine rebooted once and another time the SERVER on the AT&T would not respond
  4069. and we had to reset it.  Slowing it down to the next lower baud rate worked
  4070. (56Kbps).  Could it be the wierd AT&T bios?  (By the by, the AT&T has the same
  4071. processor as the PS/2 8086 8Mhz)
  4072.  
  4073. Thanks,
  4074.  
  4075. Rob Miracle              | Bitnet   : RWMIRA01@ULKYVX    CIS: 74216,3134
  4076. Programmer/Analyst-II    | INTERNET : rwmira01%ulkyvx.bitnet@cunyvm.cuny.edu
  4077. University of Louisville | UUCP     : ...psuvax1!ulkyvx.bitnet!rwmira01
  4078. "Incompetence is not an excuse." -- Anton Devious
  4079.  
  4080. [From jrd - Quote above is removing my favorite cover!  Scripts and Connect:
  4081. These share a common buffer for the serial port but once a script command
  4082. draws from it the data are no longer available for Connect displays.  The
  4083. OUTPUT and PAUSE commands read from that buffer to helpfully echo material
  4084. from the host and they keep doing it until there is 10 milliseconds of no
  4085. new incoming characters (10 millisec was needed for some modem responses).
  4086. Otherwise, Connect mode and scripts are not related.
  4087.  
  4088. Screen coloring: If the host sent an escape sequence such as ESC [ number m
  4089. then ANSI.SYS, or equivalent, if installed will see and react to it.  There
  4090. is nothing Kermit can do about such DOS level activities.
  4091.  
  4092. Multiple keys on PS/2 Model 30s: I agree, it seems like a machine problem.
  4093.  
  4094. Superspeed and AT&T 6300's. I can't say much about that machine because one
  4095. was available to me for only one evening some time ago. Selected IBM PC clones
  4096. can run Kermit at 115KB, and I've never encountered a system corruption
  4097. problem with those which can't make that speed. I second the motion that the
  4098. Bios may be unusual.]
  4099.  
  4100. Page 72                                                INFO-KERMIT DIGEST V8 #8
  4101.  
  4102.  
  4103. ------------------------------
  4104.  
  4105. Date: Fri, 19 Aug 88 17:35 MDT
  4106. From: Pete Klammer 303/556-3915 <PKLAMMER%CUDNVR@vaxf.colorado.edu>
  4107. Subject: Re: MS-Kermit v2.31 Eats Nulls
  4108. Keywords: MS-DOS Kermit 2.31
  4109.  
  4110. May I second that emotion?  It may be a misuse of KERMIT, but that's the way
  4111. life with tools is.  Many's the time I wish to get a good log of exactly
  4112. what's spitting out some serial path, nulls and all.  I respectfully request
  4113. that NULL and RUBOUT filtering be selectable.
  4114.  
  4115. --poko
  4116.  
  4117. [From jrd - Alas, there is no happy middle ground since most people use logs
  4118. to record worth while text for later editing and printing but many editors
  4119. cannot cope with NULLs. Also, these characters are frequently used as padding
  4120. so passing them quickly fills the serial port holding buffer; the reaction to
  4121. that was annoyance about performance when it was done as an experiment. We
  4122. have the idea that they should be displayed when DEBUG is ON, but not logged;
  4123. However, most releases have removed NULLs completely anyway. I'd suggest that
  4124. you make the simple changes to the sources to pass all the characters of
  4125. interest; see file msxibm.asm.]
  4126.  
  4127. ------------------------------
  4128.  
  4129. Date: Tue, 23 Aug 1988 16:27:47 LCL
  4130. From: David Boeshaa <BOESHAAR%SUVM.BITNET@cuvmb.cc.columbia.edu>
  4131. Subject: MS-Kermit and the Novell File Server
  4132. Keywords: MS-DOS Kermit 2.31, Novell File Server
  4133.  
  4134. Greetings! and Grief!
  4135.  
  4136. I have been working with KERMIT 2.3x for some time and I think it is GREAT!
  4137. but I have a small problem with 2.31 15 Aug 1988.
  4138.  
  4139. We are running KERMIT from a Novell 2.0a file server and every download
  4140. gives the message "Not Enough Disk Space for File" but then downloads the
  4141. file anyway. I think this may be related to the new file size checking
  4142. (we have not EVER had this problem before and we are still running
  4143. 2.31 17 July 1988 without grief).
  4144.  
  4145. Thanx and Thanx for a GREAT product!!!
  4146.  
  4147.    David K. Boeshaar    ACADEMIC COMPUTING SERVICES
  4148.               ______    SYRACUSE UNIVERSITY
  4149.              /      |         -------
  4150.             |       |   BITNET:    BOESHAAR@SUVM.BITNET
  4151.   _________/        |   NOISENET:  (315) 443-3166
  4152.   |         *       |   SNAILNET:  215 Machinery Hall
  4153.  /      SYRACUSE    |              Syracuse, NY 13244-1260 USA
  4154. |______________     |
  4155.                |_   |  Disclaimer: Any sufficiently advanced technology
  4156.                  |__|              is indistinguishable from magic.
  4157.  
  4158.  
  4159. INFO-KERMIT DIGEST V8 #8                                                Page 73
  4160.  
  4161. [From jrd - a known bug, fixed in the middle August update for IBM PCs.  There
  4162. is a similar problem if the file destination is the screen or printer; fixed
  4163. recently but not in time for the update above.  Sorry about that.]
  4164.  
  4165. ------------------------------
  4166.  
  4167. Date: Mon, 15 Aug 88 21:05 EST
  4168. From: "Frank Rens" <RENS%UBVMS.BITNET@CUVMB.CC.COLUMBIA.EDU>
  4169. Subject: Re: MS-Kermit 2.31 Tabs Bug (Feature?)
  4170. Keywords: MS-DOS Kermit 2.31, Tabs
  4171.  
  4172. I have the same problem with the TABS being cleared by a CLEAR.  Here's the
  4173. script...The tabs are OK before the clear but not after.
  4174.  
  4175. [Ed. - Script omitted.]
  4176.  
  4177. [From jrd - I finally found the bug: an uninitialized register at one place
  4178. which could (but never here!) clear the tab storage area when the CLEAR cmd
  4179. was issued. Fixed in the middle August maintence release for IBM PCs.]
  4180.  
  4181. ------------------------------
  4182.  
  4183. Date: 24-AUG-1988 12:38
  4184. From: IN%"SYSKERMIT@VAX1.LANCS.AC.UK"
  4185. Subject: MS-Kermit: Five Bugs and a Whinge
  4186. Keywords: MS-DOS Kermit 2.31
  4187.  
  4188. Syskermit & Columbia:
  4189.  
  4190.     I have recently made extensive use of MSKERMIT (various releases)
  4191. running on an XT-Clone while developing a new Kermit for a UK (Research
  4192. Machines) Micro.  Obviously, I have been putting it through quite a lot of
  4193. hoops while testing my own code's reaction to error-situations and various
  4194. sequences of events.  I have encountered the following five problems, which
  4195. are guaranteed not to be caused by misrunning of my new RM Kermit.  They
  4196. have been identified while running at various speeds, on a short direct
  4197. "null-modem" wire connecting the two micros, with the MSKermit in normal
  4198. (not server or back-port) mode.  Parity has in all cases been set to Even.
  4199. The clone is running MSDOS 3.20.
  4200.  
  4201.     The different errors have different incidences in different releases
  4202. of MSKERMIT.  See table below.
  4203.  
  4204. 1.  Unable to Resume Binary Mode.
  4205.  
  4206.     When first loaded, MSKermit will happily transfer files in binary
  4207. with 8th-bit quoting.  If a file is then transferred in 7-bit non-binary
  4208. mode, MSKermit will never revert to 8th-bit-quoting.  It simply sticks
  4209. an "N" into the spar/rpar packet, no matter what it has been sent or what
  4210. command you give it.  Main tests caried out while receiving, but I think
  4211. it is true in all circumstances.
  4212.  
  4213. [From jrd - I'm a little confused. MS Kermit lacks the concept of a 7-bit
  4214. non-binary mode transfer since there is no way within DOS to tag files as
  4215. binary or not. The conditions to use 8-bit quoting are the local side uses
  4216. parity plus the other side agrees on 8-bit quoting, or the other side requests
  4217.  
  4218. Page 74                                                INFO-KERMIT DIGEST V8 #8
  4219.  
  4220. the quoting. However, I ran a series of tests and found that indeed Kermit
  4221. could remember an old 8-bit quoting state and then later respond with the 'N'
  4222. above (for No Can Do). Tracked down and fixed this past weekend. Thanks.]
  4223.  
  4224. 2.  Not Displaying Early E-Message.
  4225.  
  4226.     If MSKermit is receiving and the other end sends an E-packet before
  4227. sending the F-packet, then this is not displayed in the "Last Error"
  4228. field.  The packet is received OK, and can be seen as "Rpack" if debug
  4229. is switched on.  (The error in question is, of course, "File not found";
  4230. this does not occur on most Kermits as they check for the presence of the
  4231. file before sending the S-packet.)
  4232.  
  4233. [From jrd - True, the packet type test had the code to do this but it was
  4234. not reached because an earlier test would fail to the file complete code
  4235. rather than the error handler code. Safe, but it omitted the error message.
  4236. Fixed now. Thanks.]
  4237.  
  4238. 3.  DEL *. Doesn't Work.
  4239.  
  4240.     If the default disk has on it files which have no extension to their
  4241. names, then although they are correctly displayed in response to the command
  4242. "dir *.",  the command "del *." gives the response "?Protected or no such
  4243. file(s).".  "del a:*." gets the same response, but "del x*." works OK
  4244. (assuming there are files whose names start with "x").
  4245.  
  4246. [From jrd - That's a DOS characteristic since Kermit merely composes a DOS
  4247. command line with DEL followed by the user's filespec and submits the works
  4248. to COMMAND.COM.  MS Kermit 2.28 and predecessors used FCBs and did deletion
  4249. file by file within Kermit; starting with 2.29 all FCB material was removed
  4250. and COMMAND.COM was employed.  Compare results of these commands between
  4251. giving them within Kermit and then at the DOS prompt.]
  4252.  
  4253. 4.  F3 Setting Gets Lost.
  4254.  
  4255.     When MSKermit is loaded and then cancelled by "Quit", if only Kermit
  4256. commands have been entered, then F3 causes MSDOS to repeat the command used
  4257. to load Kermit.  However, if a DOS-type command is given at the Kermit-MS>
  4258. prompt (e.g. dir or del), then, after quitting, F3 brings up a null command.
  4259.  
  4260. [From jrd - DOS again.  Kermit reads the keyboard at both DOS and Bios
  4261. levels.  When DOS does the work the DOS typeahead buffer can be flushed.
  4262. Kermit, and probably most programs, make no effort to restore the contents
  4263. of the typeahead buffer private to DOS. This is not a fault.]
  4264.  
  4265. 5.  Hitting RTN Kills the System.
  4266.  
  4267.     This is caused by hitting RTN [Return, Enter] unneccessarily and
  4268. repeatedly during normal running.  Tests were done during a receive, but I
  4269. think it happens during send as well.  You have to hit RTN quite frequently,
  4270. and MSKermit sends a NAK each time you do.  Eventually an unpredictable loss
  4271. of status occurs.  In the table below "Clean" means that you get back to a
  4272. Kermit-MS> prompt, and can continue (except you will find that there is a
  4273. packet or part of one in the input buffer).  "Bad" means that, if you do get
  4274. back to a prompt, then you cannot continue, and when you quit you find MSDOS
  4275. won't work; or the display may go wild.  If the display goes wild, then the
  4276.  
  4277. INFO-KERMIT DIGEST V8 #8                                                Page 75
  4278.  
  4279. keyboard is quite dead.  If you are able to quit, then the MSDOS disk-driver
  4280. has been overwritten in such a way that any attempt to use any disk gets the
  4281. "not ready error" response.  (The details of this may depend on the nature
  4282. of the Clone, since I have had the same trouble with a completely different
  4283. communications program on the same machine.)
  4284.  
  4285. [From jrd - that's not happened on any of my clones. I can hold down the
  4286. Enter key and let it auto-repeat with no ill effects except the packet
  4287. retry count goes the exhaustion limit. If a keyboard helper or an unusual
  4288. Bios does not preserve registers in a standard way (Kermit is reading keys
  4289. via DOS during packet transfers) then disaster can strike any program. DOS
  4290. stack overflow can also occur under some versions of MS DOS. Similarly,
  4291. 'adjusting' DOS's typeahead buffer can lead to wild results when that buffer
  4292. overflows. The problem has not been reported by others.]
  4293.  
  4294. 6.  Where's MSKERMIT.INI?
  4295.  
  4296.     This is really a whinge rather than a bug.  If MSKermit is not on the
  4297. default disk, i.e. you load it by a command such as "b:kermit", or get it
  4298. picked up by being in the PATH, then MSKERMIT.INI is not picked up even
  4299. though it is on the same disk as MSKermit has been loaded from.  This is
  4300. true of all versions; and for all I know it may be impossible in MSDOS to
  4301. find out what disk the program has been loaded from.
  4302.  
  4303. [From jrd - Kermit does not know from which disk or directory it has been
  4304. loaded.  The manual states that Kermit searches for MSKERMIT.INI, and Take
  4305. files, first in the CURRENT directory and then if necessary along the PATH=
  4306. string. DOS 2.X lacks the ability to inform a clever program about its
  4307. loading source.  Thus, if the current disk were say C: and Kermit were loaded
  4308. by C> b:kermit and then the search would not include drive B: unless it were
  4309. stated in the PATH= string.]
  4310.  
  4311.     Here is the incidence of bugs 1 - 5 in the different releases:-
  4312.  
  4313. \.
  4314.   \.  Bug:     Text/Bin     E-Message    DEL *.      RTNs       F3
  4315.     \.----     --------     ---------    -----       ----       --
  4316.       \.
  4317. Version \
  4318.  ------
  4319. 2.26             No           OK          OK        Clean      Lost
  4320. (undated)
  4321.  
  4322. 2.28             No           OK          OK        Clean      Lost
  4323. (undated)
  4324.  
  4325. 2.29c            No           OK          No         Bad       Lost
  4326. (23 Aug 87)
  4327.  
  4328. 2.30             No           OK          No         Bad       Lost
  4329. (8 Jan 88)
  4330.  
  4331. 2.31             No           No          No         Bad       Lost
  4332. (1 July 88)
  4333.  
  4334.  
  4335.  
  4336. Page 76                                                INFO-KERMIT DIGEST V8 #8
  4337.  
  4338.      I hope this is of some use to you.  The only one which seems to me to
  4339. be at all serious is the inability to revert to binary-quoting after running
  4340. in text-mode.
  4341.  
  4342.                                                Chris.
  4343.  
  4344. ------------------------------
  4345.  
  4346. Date: Fri 26 Aug 88 16:39:07-PDT
  4347. From: Ted Shapin <BEC.SHAPIN@ecla.usc.edu>
  4348. Subject: Goto Label Bug in 2.31 August 15?
  4349. Keywords: MS-DOS Kermit 2.31, GOTO
  4350.  
  4351. I can't get this version of Kermit to recognize labels in loops?
  4352.  
  4353. This doesn't work:
  4354. set count 3
  4355. :TOP
  4356. echo Hello\13
  4357. if count goto top
  4358. echo Goodbye\13
  4359.  
  4360. Ted.
  4361.  
  4362. [Ed. - Yup, it's another bug in the label-finding code.  A quick workaround
  4363. is to append a space to the label, followed by a comment, as in
  4364.  
  4365. :TOP ; (of loop)
  4366.  
  4367. Thanks also to B. Schneuwly, Joellen Windsor (an early Kermit developer!),
  4368. and others who reported this problem.  Joellen also discovered the following
  4369. interesting workaround:
  4370.  
  4371. define %b loop
  4372. :loop
  4373. echo test
  4374. goto \%b
  4375.  
  4376. The problem should be fixed in the next release.]
  4377.  
  4378. ------------------------------
  4379.  
  4380. Date: Wed, 31 Aug 88 14:38 GMT
  4381. From: <SCHNEUWL%CZHETH5A.BITNET@cuvmb.cc.columbia.edu>
  4382. Subject: Problems with MS-Kermit 2.31 Graphics
  4383. Keywords: MS-DOS Kermit 2.31, Olivetti
  4384.  
  4385. We have problems with Kermit 2.31, dated 15 Aug.
  4386.  
  4387. [Ed. - Loop label report omitted, see above.]
  4388.  
  4389. Graphics is another problem:
  4390. On a Olivetti M24 we don't get the same high(er) resolution display
  4391. as with MS-Kermit 2.30.
  4392.  
  4393. [Ed. - If Kermit can't automatically recognize your screen's resolution, use
  4394.  
  4395. INFO-KERMIT DIGEST V8 #8                                                Page 77
  4396.  
  4397. the SET TERMINAL GRAPHICS ATT command.  The next release should recognize
  4398. the Olivetti automatically.]
  4399.  
  4400. Furthermore I'd like to have the possibility to control whether the TEK4010
  4401. emulation does merge it's text with already existing pixels or does overwrite
  4402. as in 2.30.
  4403.  
  4404. [Ed. - Noted.]
  4405.  
  4406. ------------------------------
  4407.  
  4408. Date: Wed, 31 Aug 88 12:17 MST
  4409. From: <JWINDSOR%ARIZRVAX.BITNET@cuvmb.cc.columbia.edu>
  4410. Subject: MS Kermit 2.31 File Dates
  4411. Keywords: MS-DOS Kermit 2.31, Attributes, File Dates
  4412.  
  4413. I am running MSKERMIT 2.31 dated August 15, 1988, on a PS2 model 50 under
  4414. MS-DOS 3.30.  I have noticed the following 2 problems:
  4415.  
  4416. (1) [Ed. - Loop label problem report omitted, see above.]
  4417.  
  4418. (2) All files shipped to my PS2 from my VM/SP mainframe have the correct
  4419. creation date but they all have 12:00a for the creation time.  This appears
  4420. to have been a problem with the July 1 version, too.
  4421.  
  4422. [Ed. - This is because VM/CMS Kermit reports the creation date but not the
  4423. time of creation, so MS-Kermit uses midnight.]
  4424.  
  4425. Joellen Windsor
  4426. Systems Programmer
  4427.  
  4428. ------------------------------
  4429.  
  4430. Date: Fri, 29 Jul 88 22:40:53 EDT
  4431. From: (John F. Chandler)   PEPMNT@CFAAMP.BITNET
  4432. Subject: Kermit Through Protocol Converters Under CMS
  4433. Keywords: Kermit-370, Protocol Converters
  4434.  
  4435. Kermit-370, as released for CMS last winter, contained code designed to
  4436. allow file transfer through a variety of terminal emulators other than the
  4437. "usual" Yale ASCII type (Series/1, 7171, 4994, 9370 ASCII subsystem).  The
  4438. code was, in fact, modelled after similar code that was demonstrated under
  4439. TSO, but I haven't heard any reports of success under CMS, and the only
  4440. negative report was for a device that hadn't been tried under TSO.  I would
  4441. like to hear from anyone who has any kind of 3270 emulator other than the
  4442. Yale type (and other than the 3708, which is known to work with Kermit
  4443. already).  The particular types of emulators which might be expected to work
  4444. include the MICOM 7400, the Datastream/Leedata 8010/8030/874 models, the PCI
  4445. 1076 and 276, and possibly the Renex TMS-1.  Any others which claim to
  4446. support a "graphics passthrough" mode are also likely candidates.  I would
  4447. be interested in any results of trying Kermit with such emulators, whether
  4448. positive or negative (be sure to SET CONTROLLER GRAPH to the mainframe
  4449. Kermit before starting the transfer).  Actually, I now have an update that
  4450. automatically detects non-SERIES1-type terminals and also some other updates
  4451. that may help in diagnosing the problems if the existing code should prove
  4452. not to work "as is."
  4453.  
  4454. Page 78                                                INFO-KERMIT DIGEST V8 #8
  4455.  
  4456.  
  4457. [Ed. - John has done an amazing job with Kermit-370, but need help from
  4458. people who have equipment that he doesn't have access to.  Any cooperation
  4459. will be much appreciated!  Meanwhile, John is preparing a large batch of
  4460. updates to Kermit-370 for release -- watch Info-Kermit for announcements.]
  4461.  
  4462. ------------------------------
  4463.  
  4464. End of Info-Kermit Digest
  4465.  
  4466. INFO-KERMIT DIGEST V8 #9                                                Page 79
  4467.  
  4468. Info-Kermit Digest         Thu, 13 Oct 1988       Volume 8 : Number 9
  4469.  
  4470. Today's Topics:
  4471.  
  4472.   ANNOUNCEMENTS -
  4473.         Announcing MS-DOS Kermit 2.31/A, Beta Test
  4474.         Announcing a Beta Test Kermit for MS-Windows
  4475.         Announcing EASYK, a Menu-Driven Shell for MS-DOS Kermit
  4476.         Announcing Hebrew Kermit, a Test Version for the IBM PC
  4477.  
  4478.   MS-DOS KERMIT -
  4479.         MSDOS Kermit Script Files Available from SIMTEL20
  4480.         Automatic Exit from MS-Kermit?
  4481.         Missing the Bell When Using MS-Kermit
  4482.         Kermit Across Ethernet
  4483.         MS-Kermit with StarLAN
  4484.         PC-KERMIT ReGIS Support
  4485.         IBM 3174 AEA Ports and Kermit?
  4486.         MSDOS BOO Maker/unBOO Bug
  4487.         Kermit SET KEY Command and String Capability
  4488.         Solution Found to Problem with <ESC>[?3l
  4489.         Problem Compiling MSSSET.ASM
  4490.         AT&T Graphics in MS-KERMIT 2.31
  4491.         OS/2 Kermit Won't Work
  4492.         Problem with OS/2 Kermit
  4493.         Re: Kermit-MS version 2.31
  4494.         MS-Kermit v2.31 Inquiries
  4495.  
  4496. Send digest submissions to Info-Kermit@CUNIXC.CC.COLUMBIA.EDU, requests for
  4497. addition to or deletion from the Info-Kermit subscriber list to
  4498. Info-Kermit-Request@CUNIXC.CC.COLUMBIA.EDU.
  4499.  
  4500. Kermit files may be obtained over networks and by mail order.  On the
  4501. Internetwork, use FTP to log in to host CUNIXC, CUNIXC.CC.COLUMBIA.EDU, a
  4502. VAX 8700 running UNIX (Ultrix).  The IP host number is 128.59.40.130.  You
  4503. should be able to to FTP to CUNIXC, login as user ANONYMOUS (any password)
  4504. and GET the desired files.  The Kermit files are in directories kermit/a,
  4505. kermit/b, kermit/c, kermit/d, and kermit/e.  You can also get Kermit files
  4506. over BITNET/EARN; to get started send a message with text HELP to KERMSRV,
  4507. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  4508. file k1/aanetw.hlp (AANETW HLP on KERMSRV).  To order by mail, request a
  4509. complete list of Kermit versions and an order form from Kermit Distribution,
  4510. Columbia University Center for Computing Activities, 612 West 115th Street,
  4511. New York, NY 10025 USA.
  4512.  
  4513. ----------------------------------------------------------------------
  4514.  
  4515. Date: Wed, 2 Nov 1988 11:27:34 EDT
  4516. From: Christine M Gianone <cmg@cunixc.cc.columbia.edu>
  4517. Subject: Info-Kermit Digest V8 #9 is Sent a Second Time
  4518. Keywords: Mailer Problems
  4519.  
  4520. Due to mailer problems and reports of the many sites who did not receive this
  4521. digest the first time, Info-Kermit Digest V8 #9 is being sent a second time
  4522. to all members on the distribution list.  Apologies to those who have already
  4523. received this issue.
  4524.  
  4525. Page 80                                                INFO-KERMIT DIGEST V8 #9
  4526.  
  4527.  
  4528. ------------------------------
  4529.  
  4530. Date: Sun, 2 Oct 88 14:58 MDT
  4531. From: <JRD%USU.BITNET@cuvmb.cc.columbia.edu>
  4532. Subject: Announcing MS-DOS Kermit 2.31/A, Beta Test
  4533. Keywords: MS-DOS Kermit 2.31/A, IBM PC, DEC Rainbow, Rainbow, DECnet/DOS
  4534.  
  4535. This is to announce a beta test of a new maintenance release of MS-DOS
  4536. Kermit for the IBM PC and PS/2 families, and for the DEC Rainbow.  This
  4537. version is numbered 2.31/A.  It fixes a number of problems with 2.31, but
  4538. adds no major new functionality.
  4539.  
  4540. The most important fixes are in the following areas:
  4541.  
  4542. - Calculation of available disk space when receiving files with the size
  4543.   attribute.
  4544.  
  4545. - SET EOF {CTRL-Z | NOCTRL-Z} now applies also to the TRANSMIT command.
  4546.  
  4547. - Correct bug requiring spaces after formal labels and bug clearing VT102
  4548.   Connect mode tab stops when CLEAR command is issued.
  4549.  
  4550. - Correct long standing problem of local 8-bit quoting requirement not
  4551.   following current operating conditions.
  4552.  
  4553. - Add REMOTE LOGIN {username {password {account}}} command.
  4554.  
  4555. - Automatic identification of AT&T/Olivetti machines in Tektronix graphics
  4556.   mode for IBM-PC version of MS Kermit.
  4557.  
  4558. - Revise printing strategy within Connect mode to buffer 132 characters and
  4559.   to send XOFF/XON to the host while the buffer is written to DOS.  This helps
  4560.   prevent serial port overruns noted with slow PCs at high baud rates when the
  4561.   printer is attached to a network or similar.
  4562.  
  4563. In addition, it was verified that MS Kermit operates across DECnet using the
  4564. NetBios emulator supplied with DECnet-DOS 2.0 (aka PSCA).
  4565.  
  4566. These changes are in addition to the changes that were made between the July
  4567. initial release and the mid-August maintenance release, which included fixes
  4568. for the Ungermann-Bass Net/One support, the "stray interrupt" problem, and
  4569. a GOTO-label problem.
  4570.  
  4571. [Ed. - Thanks, Joe!  The new beta test release has been added to Kermit
  4572. Distribution as MSTIBM.BOO (IBM version) and MSTRB1.BOO (Rainbow version).
  4573. A more complete summary of the changes is in the file MSR23A.UPD.  Readers,
  4574. please get the new version and try it, report any problems to
  4575. Info-Kermit@CUNIXC.CC.COLUMBIA.EDU.  If all is well, this will become the
  4576. standard version, and sources will be made available.]
  4577.  
  4578. ------------------------------
  4579.  
  4580. Date: Thu, 6 Oct 1988 17:08:33 EDT
  4581. From: Christine M Gianone <cmg@cunixc.cc.columbia.edu>
  4582. Subject: Announcing a Beta Test Kermit for MS-Windows
  4583.  
  4584. INFO-KERMIT DIGEST V8 #9                                                Page 81
  4585.  
  4586. Keywords: MS Windows, Microsoft Windows
  4587.  
  4588. This is to announce a new version of Kermit for Microsoft Windows from Bill
  4589. Hall of the Olivetti Advanced Technology Center, who contributed the first
  4590. version in early 1987.  Unlike the earlier release, this one includes a
  4591. full-fledged Heath-19 terminal emulator, and it should run on any PC that has
  4592. version 2.0 or later of MS-Windows, Windows/286, and Windows/386.  Sources
  4593. will not be available during the testing period.  If you have comments,
  4594. suggestions, bugs to report, etc, you may correspond with Bill as
  4595. uunet!oliveb!whall (or oliveb!whall@uunet.uu.net).  The files replace the
  4596. old release on "Tape A" as WIN*.* (on CUNIXC, ~kermit/a/win*.*).  Many thanks
  4597. to Bill for all his excellent work.
  4598.  
  4599. ------------------------------
  4600.  
  4601. Date: Thu, 6 Oct 1988 17:08:33 EDT
  4602. From: Christine M Gianone <cmg@cunixc.cc.columbia.edu>
  4603. Subject: Announcing EASYK, a Menu-Driven Shell for MS-DOS Kermit
  4604. Keywords: MS-DOS Kermit 2.31, EASYK
  4605.  
  4606. This is to announce a "shell" for MS-Kermit 2.31, for use on the IBM PC and
  4607. compatibles in the IBM mainframe 3270 emulation environment, developed by
  4608. Prof. Richard A. Elnicki of the University of Florida College of Business,
  4609. $$$HUSG@NERVM.BITNET.  The shell is constructed from a collection of Kermit
  4610. command and script files, and provides menu-driven access to IBM CMS and TSO
  4611. systems via VTAM.  The package, called EasyK, was designed to provide users
  4612. with easy, full-screen access to CMS and TSO services in their homes in a mode
  4613. that looks similar to the access they have in their offices on micros with
  4614. coaxial cable boards.
  4615.  
  4616. Since there are numerous files, some of them containing 8-bit characters (e.g.
  4617. for special effects in the documentation), they have been archived using the
  4618. PKARC utility, and the resulting EASYK.ARC file was converted to .BOO format
  4619. for distribution over networks, on tape, etc, as MSIEAS.BOO (available on
  4620. CUNIXC as ~kermit/a/msieas.boo).
  4621.  
  4622. Thanks to Prof. Elnicki for contributing his work.  Although some of it is
  4623. specific to the University of Florida mainframe environment, it should be
  4624. easily customizable for other institutions with similar equipment.
  4625.  
  4626. ------------------------------
  4627.  
  4628. Date: Thu, 6 Oct 1988 17:08:33 EDT
  4629. From: Christine M Gianone <cmg@cunixc.cc.columbia.edu>
  4630. Subject: Announcing Hebrew Kermit, a Test Version for the IBM PC
  4631. Keywords: MS-DOS Kermit 2.30, Hebrew
  4632.  
  4633. This is to announce a beta-test (binaries only, no source) of MS-Kermit for
  4634. the IBM PC, adapted for use with the Hebrew Alphabet, complete with
  4635. right-to-left screen presentation, from Baruch Cochavy, IIT, Haifa, Israel,
  4636. baruchc@techunix.bitnet.  The work is based on version 2.30 of MS-Kermit.
  4637. Included are KERMITH.EXE, a "readme" file that explains some of the changes
  4638. that were made, and a Unix termcap file.
  4639.  
  4640. The program may also be used on non-Hebrew PC's, for those who enjoy reading
  4641. English (or German, or French, etc) right to left.  The default mode of
  4642.  
  4643. Page 82                                                INFO-KERMIT DIGEST V8 #9
  4644.  
  4645. operation is left to right, so say "SET TERMINAL DIRECTION RIGHT" to get the
  4646. new perspective.
  4647.  
  4648. The files are collected into a single file, MSTIBH.BOO, which in turn contains
  4649. an .ARC file which can be dearchived using PKXARC.  Thanks to Baruch for
  4650. sending this in.  Please send comments or questions directly to him at the
  4651. address noted above, and perhaps this new directional capability can become
  4652. an integral part of MS-Kermit.
  4653.  
  4654. ------------------------------
  4655.  
  4656. Date: Tue, 13 Sep 1988  01:59 MDT
  4657. From: Keith Petersen <W8SDZ@simtel20.army.mil>
  4658. Subject: MSDOS Kermit Script Files Available from SIMTEL20
  4659. Keywords: MS-DOS Kermit 2.31, Script Files
  4660.  
  4661. Now available via standard anonymous FTP from SIMTEL20.ARMY.MIL...
  4662.  
  4663. Filename            Type     Bytes     CRC
  4664.  
  4665. Directory PD1:<MSDOS.MODEM>
  4666. KER_SCP1.ARC.1            BINARY     17291  D6E5H
  4667.  
  4668. Here is the author's description:
  4669.  
  4670. KER_SCP1.ARC    - Useful Kermit (v2.29 or above) script files.
  4671.  
  4672.         Files include:
  4673.           Scripts to log on to a Unix VAX host and two common BBS's
  4674.           (TCOMM, ROS)
  4675.  
  4676.               A typical initialization file (MSKERMIT.INI) to set baud,
  4677.           port, function key macros, command macros, etc.
  4678.  
  4679.           An extract from the Kermit user's manual specifically
  4680.           discussing scripts.
  4681.  
  4682. Other Kermit users are encouraged to contribute scripts that help ease
  4683. THEIR communications.  Keith Petersen <W8SDZ@SIMTEL20.ARMY.MIL> will
  4684. collate and incorporate them in this archive.  (So send them to HIM,
  4685. not to me!)
  4686.  
  4687.           David Kirschbaum
  4688.           Toad Hall
  4689.           kirsch@braggvax.ARPA
  4690.  
  4691. --Keith Petersen
  4692. Maintainer of the CP/M and MSDOS archives at SIMTEL20.ARMY.MIL [26.0.0.74]
  4693. Arpa: W8SDZ@SIMTEL20.ARMY.MIL
  4694. Uucp: {att,decwrl,harvard,ucbvax,uunet,uw-beaver}!simtel20.army.mil!w8sdz
  4695.  
  4696. [From jrd - Thanks for the public service Keith. Users should be aware of
  4697. some small changes to MS Kermit from version to version. The Users Manual,
  4698. noted above, has a list of them.]
  4699.  
  4700. ------------------------------
  4701.  
  4702. INFO-KERMIT DIGEST V8 #9                                                Page 83
  4703.  
  4704.  
  4705. Date: Tue, 13 Sep 88 20:55:40 PDT
  4706. From: cnrdean2@violet.berkeley.edu
  4707. Subject: Automatic Exit from MS-Kermit?
  4708. Keywords: MS-DOS Kermit 2.31
  4709.  
  4710. Our administrative users utilize mskermit 2.31 exclusively as a vt102 terminal
  4711. emulator to connect their PCs to a Sun mini-computer.  Currently each user's
  4712. mskermit.ini file is set up to log them on to their own account.
  4713.  
  4714. QUESTION: Is it possible for Unix to send an escape sequence to mskermit
  4715.           which would automatically `quit' the mskermit session after the
  4716.       user types `logout'?  This sequence would be echoed from the
  4717.       user's .logout file.
  4718.  
  4719. Our goal is to make mskermit fully transparent to our administrative users.
  4720. (One less thing to teach them.)
  4721.  
  4722. Jim Bradley, College of Natural Resources Computer Facility, UC Berkeley
  4723.  
  4724. [From jrd - There is no such provision in MS Kermit, and the danger is that
  4725. TYPING/cat-ing a strange file could abruptly terminate Kermit. ALT-X is a
  4726. rather common key combination to exit a mode, and it is assigned the Kermit
  4727. keyboard verb \Kexit. Another keyboard combination can be assigned to the
  4728. same verb to satisfy local operating conditions. Both simplify the task
  4729. of remembering (and finding) Control-] C. If MS Kermit is started via a nice
  4730. batch file to simplify the startup and login process then the EXIT command
  4731. can be slipped into the script to exit Kermit whenever Connect mode is
  4732. exited.]
  4733.  
  4734. ------------------------------
  4735.  
  4736. Date: Tue, 13 Sep 88 21:24:58 PDT
  4737. From: Dallas_Hinton@mtsg.ubc.ca
  4738. Subject: Missing the Bell When Using MS-Kermit
  4739. Keywords: MS-DOS Kermit 2.31
  4740.  
  4741. RE: Info-Kermit Digest V8 #8
  4742.  
  4743. Re Kermit 2.31.....may I add my thanks and congratulations also on a superb
  4744. product.  I use it constantly at 2400 to talk with the UBC MTS system from
  4745. my home.
  4746.  
  4747. Problem (very minor):
  4748. using an Acer710 (IBM-clone) running at 10MHz with lots of TSRs and stuff(!)
  4749. or running vanilla DOS (3.1, 2. or 3) I find that after a file transfer in
  4750. either direction, I no longer have a bell.  It doesn't sound after the
  4751. transfer and can't be made to sound even after exit ( using a norton utility
  4752. called beep)
  4753.  
  4754. Rebooting solves this problem, but....
  4755. Not really important, but mildly annoying.
  4756. Thanks again.
  4757. Dallas (music teacher)
  4758.  
  4759. [From jrd - Must be an ACER bug of some subtle kind or even more likely a TSR
  4760.  
  4761. Page 84                                                INFO-KERMIT DIGEST V8 #9
  4762.  
  4763. has repeatedly grabbed the timer chip hardware. Kermit makes only very
  4764. standard use of the timer chip. Pretty common advice applies here: unload
  4765. TSRs etc and then retest with them added back in one at a time.]
  4766.  
  4767. ------------------------------
  4768.  
  4769. Date: Tue, 23 Aug 88 19:43:54 -0700
  4770. From: Alastair Milne <milne@ics.uci.edu>
  4771. Subject: Kermit Across Ethernet
  4772. Keywords: MS-DOS Kermit 2.31
  4773.  
  4774. We have recently installed PC-NFS running on Western Digital WD8003E ethernet
  4775. cards.  While in general it is running great, Sun's telnet not infrequently
  4776. experiences long suspensions of activity.  This is a known bug in telnet, but
  4777. it's very irritating.
  4778.  
  4779. Having observed that Kermit now allows a SET PORT NET command, I tried it, but
  4780. apparently it wants something different from our arrangement.  I've heard that
  4781. it's actually intended for a Token Ring net.
  4782.  
  4783. Is there any form of the command that would let Kermit work across ethernet?
  4784. It's a better terminal emulator than telnet, and it would be nice to be able
  4785. to move files without exiting and running ftp every time.  At the moment, the
  4786. alternative seems to be a different telnet implementation that runs better,
  4787. but requires that NFS not be installed -- which is a nuisance when you have
  4788. virtual drives mounted across the net.
  4789.  
  4790. Thanks for any help,
  4791.  
  4792. Alastair Milne
  4793.  
  4794. [From jrd - The SET PORT NET command requires NetBios support from your
  4795. network. With that it is independent of the access method (Token Ring,
  4796. Ethernet, or anything else). I use the same Western Digital Ethernet boards
  4797. with no ill effects. If NFS lacks NetBios perhaps it came with a serial port
  4798. Bios "interceptor" which simulates a hardware serial port. In that case SET
  4799. PORT BIOSn might work for you. No local copies of PC-NFS here so I'm guessing.]
  4800.  
  4801. ------------------------------
  4802.  
  4803. Date: Thu, 25 Aug 88 11:29:14 EDT
  4804. From: Caroline Arms <CA0L%CMUCCVMA.BITNET@cuvmb.cc.columbia.edu>
  4805. Subject: MS-Kermit with StarLAN
  4806. Keywords: MS-DOS Kermit 2.31
  4807.  
  4808. At the business school (GSB) at the University of Pittsburgh, the IBM PCs
  4809. (real IBMs, not compatibles) are on a StarLAN network, which is linked in turn
  4810. over an AT&T box called a SLIM-C to an AT&T ISN (9600 baud asynchronous
  4811. point-to-point connections) for terminal access to hosts, primarily the
  4812. school's 4381 (through a 7171 protocol converter).
  4813.  
  4814. The StarLAN was purchased after AT&T "assured" GSB that, in combination with
  4815. the SLIM-C (in beta test), this setup would provide good access to the 4381
  4816. from the PCs in the lab.  When AT&Ts own terminal emulators proved inadequate
  4817. to the task of providing enough special keys for 3270 emulation, someone at
  4818. Bell Labs. modified a version of the "clone" MS-Kermit 2.29b to work over
  4819.  
  4820. INFO-KERMIT DIGEST V8 #9                                                Page 85
  4821.  
  4822. StarLAN.  I assume that it was this work that later was incorporated into an
  4823. "official" release, since I remember an acknowledgement to AT&T.
  4824.  
  4825. We have been happily using this modified version of MS-Kermit 2.29b with
  4826. StarLAN (v 1.1), and getting adequate performance.  The screen does not redraw
  4827. as quickly as at 9600, but much faster than at 2400, after an initial hiccup
  4828. as the first few lines draw rather slowly, and then the rest is drawn almost
  4829. instantly.  This effect was marginally annoying, but you got used to it.
  4830. Input was fine, with the screen display keeping up with cursor movements and
  4831. deletions.
  4832.  
  4833. Then came StarLAN 2.0, and MS-Kermit 2.30.  MS-Kermit 2.30 now officially
  4834. supports StarLAN, but only version 2.0, so I can't do the upgrade in two
  4835. steps.  The screen re-display hiccup has disappeared, but at an unacceptable
  4836. price.  On input, the display gets several characters behind, which makes
  4837. using the cursor on a full-screen editor or to correct a typo at the beginning
  4838. of a line an exercise in frustration.  If we made the upgrade, there would be
  4839. a revolt by 300 MBA students, notoriously intolerant at the best of times!
  4840.  
  4841. I had the same problems with the "test" versions of 2.31 that I tried, but
  4842. have not yet had a chance to try out the recent 2.31 service release.
  4843.  
  4844. I have naturally asked AT&T where they think the problem might be, so far with
  4845. no luck.  Is anyone else out there using Kermit with StarLAN, and have they
  4846. experienced anything similar?
  4847.  
  4848. Caroline Arms
  4849. CA0L@CMUCCVMA   (or VMA.CC.CMU.EDU)
  4850.       [BITNET]       [Internet]
  4851.  
  4852. [From jrd - The STARLAN support in MS Kermit (as contrasted to the AT&T quicky
  4853. named STARKERM) is entirely non-AT&T code and unrelated in method or approach
  4854. to STARKERM. The AT&T item coupled to their proprietary terminal emulator
  4855. package whereas MS Kermit uses the industry standard NetBios API. Paul Fox,
  4856. then with AT&T, did a nice job with STARKERM. At his/their request and with
  4857. their material assistance I independently developed LAN support in MS Kermit,
  4858. using the industry standard NetBios session APIs, and added provisions to
  4859. support the published AT&T ISN extensions (if testing by Kermit reveals their
  4860. presence). So, any design faults in NetBios support are mine.
  4861.  
  4862.         I run STARLAN in my office and notice no such cursor lag time when
  4863. talking to locally available hosts (a Unix PC and a DOS AT) on my huge three
  4864. node network. If your host(s) or relay boxes attempt to form larger packets
  4865. from smaller ones, and wait until full, then lags will develop. Kermit sends
  4866. terminal emulation characters one per packet, except when some known escape
  4867. sequences occur which go out as one sequence per packet (also solves some
  4868. Emacs timing problems on hosts). File transfer material goes out in 256 byte
  4869. chunks. It requests reception well ahead of exhausting the buffer of
  4870. unprocessed characters and continues to process all available characters. All
  4871. this network packet sending and receiving is fully interrupt driven.  Delays
  4872. in exporting packets to MS Kermit can result in a slow response time. A simple
  4873. test of matters is to observe echos on the PC from typing single keys; lack of
  4874. timely echos is a host problem. I do notice a substantial delay when the DEL
  4875. key is pressed because my Unix host system imposes that delay; similarly with
  4876. XON/XOFF - a host problem. VI runs just fine here, with no noticable lagging.
  4877.  
  4878.  
  4879. Page 86                                                INFO-KERMIT DIGEST V8 #9
  4880.  
  4881.         If worse comes to worse we need to talk on the telephone: 300 irate
  4882. students is cruel and unusual, even for us Prof types. (801)-750-2982 work.]
  4883.  
  4884. ------------------------------
  4885.  
  4886. Date:  Wed, 31 Aug 88 15:00:31 EDT
  4887. From: Rick Wiggins <WIGGINS%AKRONVM.BITNET@cuvmb.cc.columbia.edu>
  4888. Subject: PC-KERMIT ReGIS Support
  4889. Keywords: ReGIS Color Graphics, MS-DOS Kermit 2.31
  4890.  
  4891. We (The University of Akron, Computer Center) would like to have VT241 ReGIS
  4892. color graphics support for the PCs we have attached to our VM and Ultrix
  4893. systems.  There are several commercial packages available, however, we have
  4894. been unable to find one that supports color hardcopy from the screen.  Has
  4895. anyone used a system that does support color hardcopy?  Does anyone know of a
  4896. PC-KERMIT that supports ReGIS?  If there isn't already a version of PC-KERMIT
  4897. that supports ReGIS, would anyone else be interested in one?  Please send
  4898. responses directly to me.  Thanx...
  4899.  
  4900. [From jrd - MS Kermit supports the DEC VT102 but does not support ReGIS
  4901. graphics. There is also the not insignificant problem of matching to the
  4902. particular color printer device.]
  4903.  
  4904. ------------------------------
  4905.  
  4906. Date: 31 August 88, 13:58:26 CST
  4907. From: ZZHS0008@UNLVM
  4908. Subject: IBM 3174 AEA Ports and Kermit?
  4909. Keywords: MS-DOS Kermit 2.31, IBM AEA Ports
  4910.  
  4911. I would like to know if there is any current developement in an enhancement
  4912. for MSVIBM or MSVGEN versions of Kermit for file transfer capabilities going
  4913. thru IBM 3174 AEA ports ?????  If you are unable to answer this question,
  4914. please tell me who I should ask.  Thank you for your time.
  4915.  
  4916. Jim Boeckman
  4917. University of Nebraska Medical Center
  4918. Omaha, NE
  4919. (402) 559-5679
  4920.  
  4921. [From jrd - For MS Kermit to function the host side needs to be able to cope
  4922. with async ASCII full duplex communications. MS Kermit recognizes the IBM
  4923. line turn character (typically XON) during file transfers if HANDSHAKE is
  4924. on, such as by saying DO IBM at the Kermit prompt. I am unfamiliar with 3174
  4925. ports so another person is needed to answer that part of your question.]
  4926.  
  4927. ------------------------------
  4928.  
  4929. Date: Sat, 10 Sep 1988  13:29 MDT
  4930. From: Keith Petersen <W8SDZ@simtel20.army.mil>
  4931. Subject: MSDOS BOO Maker/unBOO Bug
  4932. Keywords: MS-DOS Kermit 2.31, BOO Files
  4933.  
  4934. I don't know if it's MAKEBOO or UNBOO but if you make a BOO file from a binary
  4935. on MSDOS and then UnBOO it you end up with three binary zeros followed by a 1A
  4936. hex (control-Z) in the reconstitued binary file at the very end.  This
  4937.  
  4938. INFO-KERMIT DIGEST V8 #9                                                Page 87
  4939.  
  4940. shouldn't hurt most programs but I would recommend that you ARC your files
  4941. first and then BOO the ARC to send it to your friend.  This way the files he
  4942. ultimately extracts from the ARC will be *exactly* the same as your originals.
  4943. You will also have the added peace of mind of knowing they arrived error-free
  4944. because of the CRC checking that is done when extracting the files from the
  4945. ARC.
  4946.  
  4947. --Keith
  4948.  
  4949. [From jrd - Yes, archiving does preserve both exact lengths and applies a CRC
  4950. check as well. However, it now requires two utilities at each end, with the
  4951. archive program developer requesting payment. The trailing bytes from unBOOing
  4952. are cosmetic in almost all applications. However, Columbia has a long list of
  4953. possible improvements to the highly successful BOO process and they have now
  4954. seen your suggestion.]
  4955.  
  4956. ------------------------------
  4957.  
  4958. Date: Mon, 12 Sep 88 18:19 EST
  4959. From: SECHZER@pisces.rutgers.edu
  4960. Subject: Kermit SET KEY Command and String Capability
  4961. Keywords: MS-DOS Kermit 2.31
  4962.  
  4963. It would be nice to program combined command and strings using SET KEY.  For
  4964. example many systems do not have delete keys which delete the character
  4965. directly over the cursor.  In such cases one would like to:
  4966.  
  4967. SET KEY \339 \Krtarr \127
  4968.  
  4969. i.e. jump ahead 1 space then backspace over (thus deleting) the offeding
  4970. character.  My reading of the KERMIT manual (and trial and error) indicates
  4971. this is impossible in the current version.
  4972.  
  4973. Am I wrong?  Please advise. and ... Thanks for a great program!
  4974.  
  4975. [From jrd - the current design of the keyboard translator mechanism does not
  4976. allow multiple "items" in a key definition, despite the desirability in some
  4977. cases. Hosts working with standard terminal types, such as the VT100 series,
  4978. normally echo cursor commands to provide such customized Deletes. They
  4979. understand that just moving the cursor does not modify the characters on the
  4980. screen, but that moving the cursor plus sending spaces does simulate erasure.
  4981. It's the host which has to do all this in response to a simple command,
  4982. usually the byte BS or DEL, from the "terminal."]
  4983.  
  4984. ------------------------------
  4985.  
  4986. Date: 16 Sep 88 13:24:00 EST
  4987. From: "OAX::TURPINM" <turpinm%oax.decnet@stc10.ctd.ornl.gov>
  4988. Subject: Solution Found to Problem with <ESC>[?3l
  4989. Keywords: MS-DOS Kermit 2.31
  4990.  
  4991. I had sent a message earlier......
  4992.  
  4993. > I am using Kermit v2.31 dated 8-18-88 and have noticed a problem with
  4994. > character attributes in VT102 mode AFTER Kermit has received:
  4995.  
  4996.  
  4997. Page 88                                                INFO-KERMIT DIGEST V8 #9
  4998.  
  4999. > <ESC>[?3l
  5000.  
  5001. > This is the VT control code to go to 80 column mode.  Kermit does that
  5002. > fine, but it also turns the bold character attribute on as well.  A real
  5003. > VT doesn't do this.
  5004.  
  5005. > Has this already been reported?  If so, is there a fix?  If not, does
  5006. > anyone have any suggestions on a patch that would fix the problem?
  5007.  
  5008. I have located the problem.  I was using Kermit with it's default SET TERM
  5009. COLOR.  The default is high intensity foreground.  Apparently when <ESC>[?3l
  5010. resets the screen width to 80 columns, it also resets the foreground
  5011. intensity.
  5012.  
  5013. The solution is to EXPLICITLY set the foreground to low intensity before going
  5014. to connect mode (eg., SET TERM COLOR 0 37 44).  Then when <ESC>[?3l is
  5015. executed it will set the foreground to low intensity.
  5016.  
  5017. Mike Turpin
  5018. DDN:  turpindm%mrgate.decnet@stc10.ctd.ornl.gov
  5019.  
  5020. P.S. The host software that I'm communicating with is DEC's ALL-IN-1 and
  5021. with the minor change given above, Kermit works BEAUTIFULLY.
  5022.  
  5023. [From jrd - When entering terminal emulation MS Kermit attempts to use the
  5024. existing screen coloring and attributes, unless overridden by a SET TERM
  5025. command. If bold is on at that time then the screen width change attempts to
  5026. restore it. It is useful to realize that DEC and IBM differ about boldness:
  5027. DEC means make all displayed dots brighter but IBM hardware says make only
  5028. foreground dots brighter but not the surrounding background area of a
  5029. character.]
  5030.  
  5031. ------------------------------
  5032.  
  5033. Date: Wed 21 Sep 88 11:07:39-EDT
  5034. From: Charles Reader <READER-C@OSU-20.IRCC.OHIO-STATE.EDU>
  5035. Subject: Problem Compiling MSSSET.ASM
  5036. Keywords: MS-DOS Kermit 2.31
  5037.  
  5038.     I've been trying to built MS-KERMIT V2.31 for my Rainbow, but I get an
  5039. "OUT OF MEMORY" error when I try to compile the MSSSET.ASM module.  I'm
  5040. running version 2.11 of DOS on the Rainbow and the Rainbow has 704K of
  5041. memory.(My memory drive takes up 192K of memory, but I got the same result
  5042. when I removed the memory drive.)  Any suggestions on how I compile this
  5043. module?
  5044.  
  5045.                         Chuck
  5046.  
  5047. [From jrd - MASM is running out of symbol table space when it shows that
  5048. message. A later version of MASM is the only known cure.]
  5049.  
  5050. ------------------------------
  5051.  
  5052. Date: 10 Aug 88 21:44:00 CDT
  5053. From: "DAVID GAY" <gay@chewi.che.wisc.edu>
  5054. Subject: AT&T Graphics in MS-KERMIT 2.31
  5055.  
  5056. INFO-KERMIT DIGEST V8 #9                                                Page 89
  5057.  
  5058. Keywords: MS-DOS Kermit 2.31, Graphics
  5059.  
  5060. My original bug report:
  5061.  
  5062. > Version 2.30 of MS-Kermit supports the AT&T 6300 graphics card for the
  5063. > Tektronix emulation just fine, but version 2.31 does not. The setup shows
  5064. > that the terminal graphics is set to AT&T 640x400, but the plots are done in
  5065. > CGA 640x200 mode.
  5066.  
  5067. Reply from JRD:
  5068.  
  5069. > I wonder if we are encountering a Bios version number effect.
  5070. > For AT&T machines Kermit looks for the word 'OLIVETTI' in the Bios starting
  5071. > at locations FC00:0h or FC00:2014H.  It would be worthwhile using Debug to
  5072. > probe your Bios.
  5073.  
  5074. At your service! Here is what Debug has to say about my Bios, which is
  5075. ROM BIOS version 1.43 from AT&T, the last version for the 6300, I'm told,
  5076. which replaced the old version 1.21:
  5077.  
  5078. -D FC00:0000
  5079. FC00:0000  13 00 3C E2 E8 02 20 CB-E8 71 25 CB E8 7A 25 CB   ..<... ..q%..z%.
  5080. FC00:0010  E8 82 25 CB E8 8F 25 CB-E8 92 25 CB E8 A2 25 CB   ..%...%...%...%.
  5081. FC00:0020  E8 B3 25 CB E8 B7 25 CB-E8 3B 25 CB E8 36 21 CB   ..%...%..;%..6!.
  5082. FC00:0030  E8 A1 21 CB E8 68 10 CB-E8 A3 15 CB CA D9 00 00   ..!..h..........
  5083. FC00:0040  43 4F 50 59 52 49 47 48-54 20 28 43 29 20 20 20   COPYRIGHT (C)
  5084. FC00:0050  4F 4C 49 56 45 54 54 49-20 31 39 38 34 20 20 20   OLIVETTI 1984
  5085. FC00:0060  00 00 00 00 00 00 00 00-00 00 00 00 00 00 00 00   ................
  5086. FC00:0070  00 00 7E 81 A5 81 81 BD-99 81 7E 00 00 00 00 00   ..~.......~.....
  5087.         ...
  5088. I don't see "OLIVETTI" at either FC00:0h or FC00:2014H, but it is there
  5089. at FC00:0050. After FTPing MSGIBM.ASM from Columbia, I found this bit of
  5090. code which checks if we have an AT&T or Olivetti machine.
  5091.         ...
  5092. Now, I'm no whiz at assembler, but I think I understand why the AT&T isn't
  5093. sensed correctly. Do you perhaps need a compare at FC00:0050 ?
  5094. But the wierd thing is that it works with 2.30 but NOT with 2.31. Were
  5095. there any AT&T-related changes implemented between the two releases?
  5096. I cut the following out of the header of MSGIBM.ASM, perhaps it will jog
  5097. your memory, or lead to a solution. Unfortunately, I don't have the
  5098. source to 2.30 around here.
  5099.         ...
  5100. I hope this will enable you to track down this problem. Thanks for all your
  5101. time and efforts on this project, by the way. MSKERMIT has improved immensely
  5102. in the last several releases. The script capability is dynamite, and the
  5103. extended length packets make file transfer go faster. Keep up the good work.
  5104.  
  5105. If you want somebody to try out possible solutions, send me a message and
  5106. I'll gladly be a beta-tester.
  5107.  
  5108. David Gay
  5109. Gay@CheWi.ChE.Wisc.Edu
  5110.  
  5111. [From jrd - Thanks David. A number of people (those with that Bios and a
  5112. network connection) have commented upon this item, and it has been fixed in
  5113. the forth coming version 2.31/A (in progress for rapid release).]
  5114.  
  5115. Page 90                                                INFO-KERMIT DIGEST V8 #9
  5116.  
  5117.  
  5118. ------------------------------
  5119.  
  5120. Date: Thu, 11 Aug 88 09:29:55 ADT
  5121. From: PWHITE@unb.ca
  5122. Subject: OS/2 Kermit Won't Work
  5123. Keywords: OS/2 Kermit
  5124.  
  5125. The advent of OS/2 Kermit, even in beta form, is very welcome.  But, when I
  5126. retrieved the .BOO file from KERMSRV and de-boo-ed it in the same way as I do
  5127. for MS-Kermit releases, the program refused to run.  The error indicates that
  5128. Kermit is trying to dynamically load a portion of code and cannot find it.  I
  5129. suspect that the problem is in the BOO/unBOO process, do you know if anyone
  5130. has succesfully implemented the program in this fashion?  I am running OS/2
  5131. from Zenith (Version 1.00.5) on a Z286.
  5132.  
  5133. Peter White
  5134. Department of Chemistry
  5135. University of New Brunswick
  5136.  
  5137. [From jrd - The message here and below on OS/2 C Kermit relate to the same
  5138. fault in the original CKOKER.EXE file shipped to Columbia. It happens to me as
  5139. well, on two machines. The problem needs to be solved by the developer of that
  5140. version, Chris Adie C.Adie@edinburgh.ac.uk, since the source code has not been
  5141. made available. Here's the status of sources from Chris: "The sources for the
  5142. program and the Scribe source for the documentation will be made available
  5143. when beta testing is over (ie when bug reports slow to a trickle?)."]
  5144.  
  5145. ------------------------------
  5146.  
  5147. Date: Tue, 20 Sep 88 09:38:20 EDT
  5148. From: "Pierre Goyette" <PIERRE%MCGILL1.BITNET@cuvmb.cc.columbia.edu>
  5149. Subject: Problem with OS/2 Kermit
  5150. Keywords: OS/2 Kermit
  5151.  
  5152. I ftp'd the new Kermit for the OS/2 system for PC's and I'm sad to say it
  5153. doesn't work at all. The program fails to startup. I have had friends at other
  5154. universities try it and get the same result.
  5155.  
  5156. I get a SYS2070 when trying to load OS/2 Kermit. The message said: "The system
  5157. could not demand load the application's segment. WERMIT is in error. For
  5158. additional information, see error message SYS0202. The software diagnostic
  5159. code (trap number is: 13."
  5160.  
  5161. I'm running this on an 8 Mhz IBM AT with IBM VGA Adapter and 8513, using
  5162. Microsoft's SDK Version 1.03 (but without the PM shell). 3 Meg memory.
  5163.  
  5164. The size of my KERMIT.EXE is 126,220.
  5165.  
  5166. I hope that some other folks have reported this problem. Please let me know if
  5167. you need any other info.
  5168.  
  5169. Thanks,
  5170.  
  5171. Pierre Goyette
  5172. Network Systems Programmer
  5173.  
  5174. INFO-KERMIT DIGEST V8 #9                                                Page 91
  5175.  
  5176. McGill University
  5177.  
  5178. ------------------------------
  5179.  
  5180. Date: Tue, 20 Sep 88 09:57:16 BST
  5181. From: J. M. Hicks Coventry 523262 <CUDAT@sky.warwick.ac.uk>
  5182. Subject: Re: Kermit-MS version 2.31
  5183. Keywords: MS-DOS Kermit 2.31
  5184.  
  5185. Kermit-MS for the IBM PC seems to have had a curious flaw since version 2.28
  5186. at least.  The flaw seems to still be there in version 2.31.  If I have a file
  5187. called SKY.CON which has "connect" as the last command, I can use "do sky"
  5188. in Kermit-MS to invoke the commands in this file, including the "connect"
  5189. command at the end.  Everything is fine until I want to get back to Kermit-MS,
  5190. whereupon I need to type "ctrl/] c ctrl/] c" instead of just "ctrl/] c".
  5191.  
  5192.    If the "connect" command is typed at the keyboard, there is no problem.
  5193.  
  5194.    I would be grateful if you'd pass this on to the author of Kermit-MS
  5195. for his attention.  I should see any reply in the Info-Kermit Digest, as
  5196. I can obtain copies of it here in the U.K.
  5197.  
  5198. J. M. Hicks, (a.k.a. Hilary)
  5199. Computing Services, Warwick University, Coventry, England. CV4 7AL
  5200. On JANET: CUDAT at UK.AC.WARWICK.SKY
  5201. On BITNET: CUDAT@SKY.WARWICK.AC.UK
  5202. On the Internet: try  CUDAT%SKY.WARWICK.AC.UK@cunyvm.cuny.edu  (untested)
  5203.  
  5204. [From jrd - I think you might have two CONNECT commands in there somewhere.
  5205. If commands were really done from a file one would say TAKE SKY.CON rather
  5206. than invoking a macro as DO SKY. So, SKY.CON probably defines a macro and
  5207. one can view it via SHOW MACRO.
  5208. Here is an example which worked fine here a minute ago:
  5209.         File x.tak holds the single line
  5210.         define x echo starting,connect
  5211.  
  5212.         Then I said
  5213.         Kermit-MS> Take x.tak
  5214.         Kermit-MS> Do x
  5215.         see Connect mode screen ...
  5216.         type ALT-X or Control-] C
  5217.         Kermit-MS> prompt reappears.]
  5218.  
  5219. ------------------------------
  5220.  
  5221. Date: Fri, 12 Aug 88 13:38 N
  5222. From: <VDVELDE%HLERUL53.BITNET@cuvmb.cc.columbia.edu>
  5223. Subject: MS-Kermit v2.31 Inquiries
  5224.  
  5225. I have some peculiar problem with Kermit v2.31. After the Kermit> prompt
  5226. appears, my PC says: "Unexpected HW interrupt xx at yy. Type R to (R)eboot,
  5227. any other key to continue". The first time this happens, xx= 0C, and yy=
  5228. 0070:0470, all successive times xx= 0B and yy= 0070:03F6.  The message keeps
  5229. appearing when going into and from connect mode.  Thing is, that it doesn't
  5230. happen with Kermit v2.30. Still, first I blaimed my Smartlink 1200BS modem,
  5231. but after removing it (and all other cards except the display adapter) I still
  5232.  
  5233. Page 92                                                INFO-KERMIT DIGEST V8 #9
  5234.  
  5235. got the HW interrupt message.  So it is either my PC (Commodore PC20-II) which
  5236. up until now, has shown to be a perfect IBM-PC compatible, or something
  5237. special in the new MS-Kermit. Any ideas?
  5238.  
  5239. Enno Van Der Velde
  5240. University Hospital Leiden, The Netherlands
  5241. vdvelde@hlerul53.bitnet
  5242.  
  5243. [From jrd - Interrupts 0ch and 0bh refer to the serial port hardware which
  5244. your machine says it has. The indications are of a very early version of
  5245. the Phoenix Bios where interrupts were not fielded as gracefully as in most
  5246. machines. The mid-August maintenance release, and hence the newer 2.31/A
  5247. currently in the works, attempts to bypass that problem.]
  5248.  
  5249. ------------------------------
  5250.  
  5251. End of Info-Kermit Digest
  5252.  
  5253. INFO-KERMIT DIGEST V8 #10                                               Page 93
  5254.  
  5255. Info-Kermit Digest         Fri, 11 Nov 1988       Volume 8 : Number 10
  5256.  
  5257. Today's Topics:
  5258.  
  5259.         Announcing a Partial Re-release of TSO Kermit
  5260.            Announcing a new version of Kermit for the Z100
  5261.                 New OS/2 Kermit Files
  5262.             Utilities for Conversion of BOO files
  5263.                  EasyK Kermit Package
  5264.          Mac Kermit Settings Files for Mac Plus with Series/1
  5265.              File Transfer using SIM3278
  5266.         Apparent Bug in Long Packet Size on IBM 370 Mainframe
  5267.           Using EPSON PX8 without BASIC interpreter
  5268.                   EPSON PX8 Support
  5269.         Echo Bug in Kermit-80 ver 4.09 for NorthStar Advantage
  5270.               Kermit-80 for a Xerox 16/8
  5271.             8th-bit Stripping VAX-VMS->PC
  5272.                VMS Kermit with an X.29 Module?
  5273.                Need Compiled Kermit for VMS 4.3
  5274.              Kermit for MVS/Roscoe ETSO?
  5275.        VT100 Terminal EMulation for TI Professional Computers?
  5276.              A Question about the Kermit Protocol
  5277.               Kermit and SysVR3 layers.
  5278.  
  5279. Send digest submissions to Info-Kermit@CUNIXC.CC.COLUMBIA.EDU, requests for
  5280. addition to or deletion from the Info-Kermit subscriber list to
  5281. Info-Kermit-Request@CUNIXC.CC.COLUMBIA.EDU.
  5282.  
  5283. Kermit files may be obtained over networks and by mail order.  On the
  5284. Internetwork, use FTP to log in to host CUNIXC, CUNIXC.CC.COLUMBIA.EDU, a
  5285. VAX 8700 running UNIX (Ultrix).  The IP host number is 128.59.40.130.  You
  5286. should be able to to FTP to CUNIXC, login as user ANONYMOUS (any password)
  5287. and GET the desired files.  The Kermit files are in directories kermit/a,
  5288. kermit/b, kermit/c, kermit/d, and kermit/e.  You can also get Kermit files
  5289. over BITNET/EARN; to get started send a message with text HELP to KERMSRV,
  5290. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  5291. file k1/aanetw.hlp (AANETW HLP on KERMSRV).  To order by mail, request a
  5292. complete list of Kermit versions and an order form from Kermit Distribution,
  5293. Columbia University Center for Computing Activities, 612 West 115th Street,
  5294. New York, NY 10025 USA.
  5295.  
  5296. ----------------------------------------------------------------------
  5297.  
  5298.  
  5299. Date: Thu, 1988 Sep 8 23:21 EDT
  5300. From: (John F. Chandler) PEPMNT@CFAAMP.BITNET
  5301. Subject: Announcing a Partial Re-release of TSO Kermit
  5302. Keywords: IBM 370 Kermit, 370 Kermit, TSO Kermit, MVS/TSO
  5303.  
  5304. This is to announce a number of modifications to release 4.0 of TSO
  5305. Kermit-370.  The most noticeable change is in the installation procedure --
  5306. the revised IKTKER.INS contains a batch job that can be submitted to install
  5307. Kermit from scratch all in one step (after allocating a pair of datasets by
  5308. hand).  One by-product of the new procedure is a version of the update
  5309. utility GUPI-TSO that can be used either interactively or in batch.  There
  5310. is now a HELP file for GUPI.  Numerous bugs in Kermit have been fixed (as
  5311.  
  5312. Page 94                                               INFO-KERMIT DIGEST V8 #10
  5313.  
  5314. promised in a recent edition of IKTKER.BWR), and several new features have
  5315. been added.  Among them are:
  5316.  
  5317. 1. A subcommand SET LONGLINE, to select the desired processing of lines
  5318.    received by TSO that exceed the LRECL -- the options are to fold,
  5319.    truncate or simply halt.
  5320.  
  5321. 2. LRECL is now treated by Kermit exactly the way MVS does -- it must be
  5322.    four greater than the actual data length for RECFM=V files.
  5323.  
  5324. 3. File allocation error messages are now saved in Kermit and reported
  5325.    as part of the STATUS text.
  5326.  
  5327. 4. Kermit-370 automatically detects three types of terminal connections
  5328.    and sets CONTROLLER accordingly -- to TTY for line-mode terminals
  5329.    (3705, 3708, etc.), to SERIES1 for terminals hooked through protocol
  5330.    converters of the Yale ASCII family (Series/1, 7171, 4994, etc.), and
  5331.    to GRAPHICS for any other kind of full-screen device.
  5332.  
  5333. By the way, items 1 and 4 will soon be available for CMS as well.
  5334.  
  5335. This revision of Kermit requires not only a fresh copy of IKTKER.UPD, but
  5336. also replacements for four of the source components: IKTDYN.ASM, IKTGUP.ASM,
  5337. IK0GUP.ASM, and IKTMAC.ASM.  Note: if you try to install Kermit without
  5338. these new sources, the batch job will issue an error message in the update
  5339. step.
  5340.  
  5341. [Ed. - Many thanks, John.  The new files have been installed in Kermit
  5342. Distribution, along with a new file, IK0CON.HLP, that describes the
  5343. technical aspects of supporting (or adding support for) the various 370
  5344. console devices.]
  5345.  
  5346. And now a question: Kermit-CMS and, more recently, Kermit-370 for both CMS and
  5347. TSO have offered the option of setting the assumed parity of the incoming data
  5348. from protocol emulators (such as the 7171).  Has anyone ever taken advantage
  5349. of this option by setting the PARITY to NONE in the mainframe Kermit?  If so,
  5350. what hardware was involved?
  5351.  
  5352. ------------------------------
  5353.  
  5354. Date: Thu, 3 Nov 88 20:23:53 EST
  5355. From: Drew Derbyshire <ahd@omnigate.clarkson.edu>
  5356. Subject: Announcing a new version of Kermit for the Z100
  5357. Keywords: MS-Kermit Z-100 MSTZ10, Zenith Z100, Heath/Zenith Z100
  5358.  
  5359. You will find following as separate files my latest test copy of Kermit-MS
  5360. 2.31 for the Z-100, internally dated 16 October 1988.  Its documentation also
  5361. follows, no additional charge.  Please add the program to the Columbia Kermit
  5362. archives as MSTZ10.BOO (replacing my earlier test copy of MSTZ10 from January)
  5363. and the documentation as MSTZ10.DOC.
  5364.  
  5365. These files are also available from clutx.clarkson.edu via anonymous ftp as
  5366. pub/mstz10.exe and pub/mstz10.doc.
  5367.  
  5368. Very quickly, the highlights of the program include:
  5369.  
  5370.  
  5371. INFO-KERMIT DIGEST V8 #10                                               Page 95
  5372.  
  5373.         All the new features of base Kermit-MS 2.31
  5374.         Improved printer support (in other words, it has some!)
  5375.         Improved key pad handling
  5376.         Various bug fixes
  5377.  
  5378. It does NOT include VT-102 emulation, nor do I have any plans to add it soon.
  5379. I write (and retrofit) operating systems for a living, and while retrofitting
  5380. the IBM-PC Kermit VT-102 emulation is possible, I have no personal use for it
  5381. and only one outside request to do so; thus putting the 100+ hours required
  5382. into such a project is not reasonable.
  5383.  
  5384. I will be adding VT-52 print through (escape code) processing and making other
  5385. changes as listed in the documentation, but not for a few weeks.  In the
  5386. meantime, I eagerly (?) await bug reports on the current version.
  5387.  
  5388. As for other comments about Z-100 Kermit-MS on various mailing lists in recent
  5389. months, Shift F12 (nor BREAK) are processed by Kermit because the BIOS
  5390. intercepts the keys.  For additional comments on this and related topics, see
  5391. the documentation file.
  5392.  
  5393. Drew Derbyshire
  5394.  
  5395. [Ed. - Thanks Drew!  The new MSTZ10.BOO has replaced the old one and
  5396. MSTZ10.DOC has been added.]
  5397.  
  5398. ------------------------------
  5399.  
  5400. Date: 27-SEP-1988 09:50:34 GMT +01:00
  5401. From: SYSKERMIT%vax1.central.lancaster.ac.uk@nss.cs.ucl.ac.uk
  5402. Subject: New OS/2 Kermit Files
  5403. Keywords: OS/2 Kermit
  5404.  
  5405. Sending the following new OS/2 Kermit files to replace the older ones:
  5406. CKOKER.DOC
  5407. CKOKER.BOO
  5408. CKOKER.BWR
  5409.  
  5410. [Ed. - Thanks Steve!  These files have been added to the distribution.
  5411. Reportedly, this version has some problems with VT100 terminal emulation
  5412. when parity is being sent by the host -- the problem is that CR, LF, ESC, and
  5413. other control characters are not recognized when they have the parity bit on,
  5414. and so do not cause the intended screen formatting to occur.  The typical
  5415. symptom is that after the first line reaches "column 80", all subsequent
  5416. characters are displayed in the same cell.  Reportedly, the workaround is to
  5417. get the host to not send parity (if this is possible).  Sources are not yet
  5418. available.  The executable program is encoded in "boo" format, and may be
  5419. decoded using any of the MSBPCT programs.  The C version has been included
  5420. with the CKO files as CKOPCT.C, but may need some conversion to work under
  5421. OS/2.  Please report any problems to Info-Kermit]
  5422.  
  5423. ------------------------------
  5424.  
  5425. Date: Mon, 24 Oct 1988 22:16 EDT
  5426. From: <WGCP0086@GWUVM.BITNET>
  5427. Subject: Utilities for Conversion of BOO files
  5428. Keywords: .BOO Files
  5429.  
  5430. Page 96                                               INFO-KERMIT DIGEST V8 #10
  5431.  
  5432. Xref: BOO Files, see .BOO Files
  5433.  
  5434. I am sending two ASM files for your evaluation which perform the conversion
  5435. of BOO to EXE and EXE to BOO files far faster than the utilities you
  5436. currently distribute on KERMSRV@CUVMA.  For example, MSBPCT took 6 mins 32
  5437. secs to convert MSVIBM.BOO, while the new program MSPC2 took under 16
  5438. seconds on a PC XT.
  5439.  
  5440. The file names are:
  5441.       MSBPC2 ASM  to convert from BOO to EXE or binary
  5442.       MSBMKB ASM  to convert from EXE or binary to BOO
  5443.  
  5444. William C. Parke
  5445. Physics Department, GWU
  5446.  
  5447. [Ed. - Many thanks!  Now we have EXE-to-BOO programs in assembler (your new
  5448. one), C, Fortran, and Pascal, and BOO-to-EXE programs in Basic, C, Fortran,
  5449. Pascal, and assembler (two of them, one of them yours).]
  5450.  
  5451. ------------------------------
  5452.  
  5453. Date: Tue, 30 Aug 88 12:07 MDT
  5454. From: Joe Doupnik <JRD@cc.usu.edu>
  5455. Subject: EasyK Kermit Package
  5456. Keywords: EasyK Kermit
  5457.  
  5458. The people at the University of Florida, Gainsville have put together an
  5459. extensive shell for MS Kermit to IBM mainframe communications. I've taken
  5460. their draft floppy disk and archived it with PKARC and boo'd the result as
  5461. EASYK.BOO, which follows.
  5462.  
  5463.         The package is pretty slick in its way.
  5464.         Regards,
  5465.         Joe D.
  5466.  
  5467. [Ed. - Thanks Joe!  The file has been put in kermit/a/msieas.boo, which after
  5468. unbooing (via any of the MSBPCT programs), is an archive of many files which
  5469. can be de-archived using PKXARC.]
  5470.  
  5471. ------------------------------
  5472.  
  5473. Date: Fri, 02 Sep 88 15:57:58 EDT
  5474. From: Joe McMahon <XRJDM@SCFVM.BITNET>
  5475. Subject: Mac Kermit Settings Files for Mac Plus with Series/1
  5476. Keywords: MacKermit 0.9(40), 3270 Emulation, Series/1, 7171
  5477.  
  5478. Here's a Mac Plus settings file. It provides a setup useful for IBM 3270
  5479. emulation with Series/1-style protocol converters (7171, 4994, etc), along
  5480. with a MacDraw file for the keyboard map.
  5481.  
  5482. [Ed. - Thanks!  Your files have been put in ~kermit/b/ckmker.set.  Any Mac
  5483. Kermit settings files to be contributed in the future will be appended to the
  5484. same file.  The ckmker.set file is in Unix mail format, with the settings
  5485. files and keyboard maps embedded as BinHex 4.0 text files.]
  5486.  
  5487. ------------------------------
  5488.  
  5489. INFO-KERMIT DIGEST V8 #10                                               Page 97
  5490.  
  5491.  
  5492. Date: Thu, 11 Aug 88 13:22:42 ITA
  5493. From: Angelo De Florio <ADF0%ICINECA.BITNET@CUVMB.CC.COLUMBIA.EDU>
  5494. Subject: File Transfer using SIM3278
  5495. Keywords: Kermit-CMS, SIM3278
  5496.  
  5497. Has someone tried file-transfer using SIM3278/VM or SIM3278/GCS with
  5498. Kermit-CMS ?
  5499.  
  5500. I didn't be able to obtain it, using the Kermit-CMS command SET SERIES1 ON,
  5501. although the SIMWARE software should be compatible with IBM Series/1 YALE
  5502. IUP emulation system. Do I need to change some parameters ?
  5503.  
  5504. Thanks
  5505. Angelo De Florio
  5506.  
  5507. [Ed. - To our knowledge, so far no one has succeeded in getting Kermit to work
  5508. with SIM3278.  If anyone out there knows the secret, please share it!]
  5509.  
  5510. ------------------------------
  5511.  
  5512. Date: Tue, 20 Sep 88 20:08:28 EDT
  5513. From: Peter Jones <MAINT@UQAM>
  5514. Subject: Apparent Bug in Long Packet Size on IBM 370 Mainframe
  5515. Keywords: Kermit-370, Kermit-CMS
  5516.  
  5517. There seem to be a small problem with SET RECEIVE PACKSIZE and SET SEND
  5518. PACKSIZE. SET RECEIVE seems to be OK, but changes both RECEIVE and SEND size.
  5519. SET SEND gives a diagnostic, but the SHOW PROTOCOL command shows that both
  5520. sizes have been altered.  These remarks apply to Version 4.0 of mainframe
  5521. KERMIT.
  5522.  
  5523. When running large packets on a PC with SET DEBUG on, the packets displayed
  5524. are so long that they cause the screen to scroll, ruining the status display.
  5525.  
  5526. [Ed. - Normally, the SET SEND PACKET command is never needed.  The size of
  5527. the packets is controlled by the receiver, via the SET RECEIVE PACKET command.
  5528. SET SEND PACKET is used only to override the negotiated value, and then only
  5529. downwards.  And you're right, long packets are not suitable for display on the
  5530. PC screen during file transfer.  It might be possible to create separate
  5531. scrolling regions for them, but this would destroy the system-independence of
  5532. this part of the program.  If you need to see long packets on the PC, you
  5533. can LOG PACKETS, rather than SET DEBUG ON.]
  5534.  
  5535. ------------------------------
  5536.  
  5537. Date: Thu, 15 Sep 88 11:14:35 EDT
  5538. From: Peter Jones <MAINT@UQAM.BITNET>
  5539. Subject: Using EPSON PX8 without BASIC interpreter
  5540. Keywords: EPSON PX8
  5541.  
  5542.      I note with interest that there exists a file CPVPX8 that presumably
  5543. allows KERMIT to run on an EPSON PX-8. I have a two professors at UQAM who
  5544. would be interested in transferring files from their EPSONs to an IBM PC.
  5545. These EPSONs have the Travelling Wordstar PROMs, and do not have BASIC, as
  5546. there is not enough room for both.  I wonder how one can get KERMIT onto such
  5547.  
  5548. Page 98                                               INFO-KERMIT DIGEST V8 #10
  5549.  
  5550. a machine?
  5551.  
  5552.      It is possible to use EPSON's TERM program to send and receive files in
  5553. 8-bit ASCII, but I'm not sure that every ASCII character is in fact
  5554. transferred.  I have been able to transfer EPSON Wordstar files to the PC,
  5555. pass them through WORDPERFECT's CONVERT program, and read them with
  5556. WORDPERFECT.  The formatting commands seem to be transferred correctly.  The
  5557. user, however, finds the procedure too complicated and error-prone.  There are
  5558. too many manually-initiated steps, each of which must be done a certain way.
  5559.  
  5560.      EPSON has given us an QMODEM-compatible program to run on the EPSON, and
  5561. a copy of QMODEM to run on the PC. The transfer procedure is as complicated as
  5562. using TERM because, without documentation, I have been unable to discover how
  5563. to send a batch of files from the EPSON to the IBM PC without manually
  5564. starting and stopping each file on both machines.  Moreover, the EPSON
  5565. software takes 15 minutes to load from the microcassette --- far too long
  5566. because the software has too many features which are useless in this
  5567. particular application.
  5568.  
  5569.      EPSON does have a built-in program called FILSND, which allows PX-8's to
  5570. send files to one another and to EPSON PX-10's.  However, I have been unable
  5571. to get specifications from EPSON Canada on this protocol.  Do you have any
  5572. idea where this information can be obtained?  The difficulties of loading
  5573. executable programs into a PX-8 lead me to believe that any communication
  5574. program must be built-in.  I conclude with two methods of transferring files
  5575. from a PX-8 to a PC:
  5576.  
  5577. 1) Write a PC program to emulate EPSON's PX-8 PX-10 protocol
  5578. 2) Modifiy the EPSON PROMs to hold KERMIT instead on FILSND.
  5579.  
  5580. Does anyone have any comments on these ideas?
  5581.  
  5582. ------------------------------
  5583.  
  5584. Date: Thu, 29 Sep 88 12:40:49 EDT
  5585. From: Peter Jones <MAINT@UQAM.BITNET>
  5586. Subject: EPSON PX8 Support
  5587. Keywords: EPSON PX8
  5588.  
  5589.      I would like to draw your attention to a letter published in the "Chaos
  5590. Manor Mail" section of "Byte" magazine, July 1988, Volume 13, Number 7,page
  5591. 36, column 1, entitled "Calling All PX-8 Orphans".  This letter gives a couple
  5592. of places where one can still get PX-8 support:
  5593.  
  5594. 1)   Synderscope International
  5595.      1594 Hilltop Dr., El Cajon, CA 92020
  5596.      (619)-442-3789 (bulletin board)
  5597.  
  5598. 2)   DAK Industries
  5599.      8200 Remmet Ave., Canoga Park, CA 91304
  5600.  
  5601. 3)   Bulletin board: (313)-538-6968
  5602.  
  5603.      I believe people ordering the PX-8 version of KERMIT would appreciate
  5604. receiving this information with their documentation.  I personally have not
  5605. tried contacting any of these people yet, so I don't know how useful these
  5606.  
  5607. INFO-KERMIT DIGEST V8 #10                                               Page 99
  5608.  
  5609. addresses are.
  5610.  
  5611. [Ed. - Thanks, Peter.  This message has been put with the Epson Kermit hex
  5612. file as cpvpx8.hlp.]
  5613.  
  5614. ------------------------------
  5615.  
  5616. Date: Wed, 2 Nov 1988  00:12 MST
  5617. From: Keith Petersen <W8SDZ@wsmr-simtel20.army.mil>
  5618. Subject: Echo Bug in Kermit-80 ver 4.09 for NorthStar Advantage
  5619. Keywords: CP/M Kermit, NorthStar Advantage, Kermit-80
  5620.  
  5621. Relayed from GEnie's CP/M RoundTable:
  5622.  
  5623.  Message 17        Mon Oct 31, 1988
  5624.  J.LABUTE [JP]                at 21:42 EST
  5625.  
  5626. There seems to be a bug in CP409ADV.COM, the version 4.09 of Kermit-80
  5627. for the NorthStar Advantage. This version echos back to the remote
  5628. everything it receives.  I have had no problems with the versions for
  5629. the Apple II.
  5630.  
  5631. Replys to Keith Petersen <W8SDZ@WSMR-SIMTEL20.ARMY.MIL> please.
  5632. I will relay back to GEnie.
  5633.  
  5634. [Ed. - Bertil, are you out there?]
  5635.  
  5636. ------------------------------
  5637.  
  5638. Date: Sat, 22 Oct 88 22:41:36 EDT
  5639. From: olmiller@ludwig.scc.com (Otto Miller)
  5640. Subject: Kermit-80 for a Xerox 16/8
  5641. Keywords: Kermit-80, Xerox 16/8
  5642.  
  5643. A little background info...  I am posting this request to both info-kermit and
  5644. info-cpm since I am not sure where my problem is.  I have installed several
  5645. variations of unix c-kermit and now I am trying to install cpm-80 kermit.
  5646.  
  5647. The specifics; I am using a Xerox 16/8 (5.25" floppy + HD).  Since the heart
  5648. of the 16/8 is a Xerox 820 (like) board I used the sys-dependent overlay for
  5649. the Xerox 820 (cpvxer.hex) with the sys-indep. (cpsker.hex) hex file to
  5650. generate the .com file.  All went well up to this point.  The problem appears
  5651. when I connect...  I get nothing out the serial port!  I looked at the source
  5652. to see if port B was used, but everything points to port A.  I have been
  5653. unable to hang a protocol analyzer to confirm no output, but when I hang a
  5654. known good modem on the port I get nothing... not even garbage.  WHAT's
  5655. WRONG?
  5656.  
  5657. Any and all help, pointers, etc. with constructive flames are solicited.
  5658. PLEASE email directly to me in that I am not on regular distribution for this
  5659. group; Thanks.  Thank you in advance for all the help.
  5660.  
  5661. #include std_disclaim.h             In real(?) life: Otto L. Miller
  5662.                                                      Contel FSS/ASD
  5663. email olmiller@ludwig.scc.com                        Fairfax, VA 22033
  5664.                                                      703-359-7500
  5665.  
  5666. Page 100                                              INFO-KERMIT DIGEST V8 #10
  5667.  
  5668.  
  5669. [Ed. - Maybe the actual port addresses are different for the two Xerox
  5670. machines.  If anybody knows the answer, please send it in!  In a pinch, you
  5671. should be able to use the "generic" version of CP/M-80 Kermit.]
  5672.  
  5673. ------------------------------
  5674.  
  5675. Date: Thu, 6 Oct 88 10:16:03 EDT
  5676. From: A.W.F. Huggins <bhuggins@LABS-N>
  5677. Subject: 8th-bit Stripping VAX-VMS->PC
  5678. Keywords: VAX/VMS Kermit, MS-DOS Kermit 2.31
  5679.  
  5680. IBM-PC Kermit-MS 2.31: 15 Aug 88
  5681.  
  5682. Since we installed the latest Kermit, we have had problems transferring binary
  5683. files between a Compaq 286 or 386 to a VAX 785 running VMS.  Either sending or
  5684. receiving from the PC, the msb is stripped.  We thought at first that it was
  5685. some problem with the serial line card in the Compaqs, since the problem did
  5686. not arise running the same version of Kermit on an XT.  Then we tried IC, and
  5687. 8-bit files transferred without problems.  After some poking around, we found
  5688. that setting the parity bit to SPACE (instead of NONE) resulted in correct
  5689. transfers on the Compaqs, although the transfers were inefficient because all
  5690. 8-bit bytes were transmitted by quoting.  Finally, we noticed that the
  5691. MSKERMIT.INI on the Compaqs had CRC block-checking enabled, whereas the XT did
  5692. not.  It appears that CRC checksumming is incompatible with 8-bit
  5693. transmissions.  There is no statement to this effect in our copy of the
  5694. documentation, and there should be.
  5695.  
  5696. [Ed. - 8th bit prefixing is totally independent of CRC block checking.
  5697. However, there was a bug in that version of Kermit which would make it
  5698. sometimes use 8th-bit prefixing when it was not negotiated.  Try the current
  5699. beta-test prerelease version of 2.31A, which is in kermit/a/mstibm.boo on
  5700. CUNIXC or MSTIBM BOO from KERMSRV at CUVMA.]
  5701.  
  5702. VMS Kermit-32:
  5703.  
  5704. And while we are at it, what is the use/purpose of SET FIL TYP BIN on
  5705. VMS-KERMIT-32?  If you use this mode instead of FIX, you get CR-LF inserted
  5706. after every block.  Can this ever be beneficial?  Why not have BIN map onto
  5707. FIX, so people who apply common sense rather than consulting the documentation
  5708. win too?
  5709.  
  5710. [Ed. - Kermit-32 is no longer actively maintained, and the original authors
  5711. are not available to answer your question.  To our knowledge, FIX does not
  5712. insert CRLFs, rather it pads the file out with nulls to a multiple of 512
  5713. bytes.  Binary files transferred to a VAX from a PC and back using BINARY
  5714. come back identical to the original, whereas those sent using FIX come back
  5715. with extra nulls on the end.]
  5716.  
  5717. Bill
  5718.  
  5719. ------------------------------
  5720.  
  5721. Date: Thu, 21 Jul 88 14:02 GMT
  5722. From: "Eoin Meehan, EuroKom" <EMEEHAN@vms.eurokom.ucd.ie>
  5723. Subject: VMS Kermit with an X.29 Module?
  5724.  
  5725. INFO-KERMIT DIGEST V8 #10                                              Page 101
  5726.  
  5727. Keywords: VAX/VMS Kermit, X.29 Module
  5728.  
  5729. Does anyone know of a version of Kermit for VMS which has the ability to
  5730. connect over PSI?
  5731.  
  5732. What I'm looking for is a command something of the form
  5733.  
  5734.                 "CONNECT/X29 0272412345678"
  5735.  
  5736. We use Kermit extensively for file transfer. Our users with PC's are ok
  5737. but some of our users have VAXen and they use  the SET HOST/X29 to connect
  5738. to us. This obviously precludes using Kermit file transfer! If a user really
  5739. needs kermit file transfer then we suggest using the "kermit loop" whereby
  5740. they loop two async ports together, "CONNECT" to one port and log in again,
  5741. and then use the SET HOST/X29 command -- kinda messy.
  5742.  
  5743. Thanks,
  5744.  
  5745. Eoin Meehan
  5746.  
  5747. Eoin Meehan                      Bitnet/EARN: emeehan@vms.eurokom.ucd.ie
  5748. EuroKom                          Voice:       +353 1 697890
  5749. Computer Centre                  Telex:       91178 UCDEI
  5750. University College Dublin        FAX:         +353 1 838605
  5751. Belfield                         PSImail:     PSI%027243159000737::emeehan
  5752. Dublin 4
  5753. Republic of Ireland
  5754.  
  5755. "Why is there only one Monopolies Commission?"
  5756. "Who builds DIY stores?"
  5757.  
  5758. [Ed. - This is a very popular request, especially from Europe.  Kermit needs
  5759. to have special code added to use VAX/PSI.  So far, no one has volunteered to
  5760. do this.]
  5761.  
  5762. ------------------------------
  5763.  
  5764. Date: Fri, 22 Jul 88 14:00:48 CDT
  5765. From: moore@ncsc.ARPA (Moore)
  5766. Subject: Need Compiled Kermit for VMS 4.3
  5767. Keywords: C-Kermit, VAX/VMS Kermit
  5768.  
  5769. Is there anyone out there who would be willing to send me a tape of a
  5770. compiled Kermit for a VAX 11/780 running VMS 4.3?  We don't have a C
  5771. compiler so we can't grab the source from CU20B.
  5772.  
  5773. Thanks for any assistance.
  5774.  
  5775. Jim
  5776. moore@NCSC.arpa
  5777.  
  5778. ------------------------------
  5779.  
  5780. Date: Thu, 25 Aug 88 15:09:50 CST
  5781. From: "Subir Prasad" <CCDOC%UMSVM.BITNET@cuvmb.cc.columbia.edu>
  5782. Subject: Kermit for MVS/Roscoe ETSO?
  5783.  
  5784. Page 102                                              INFO-KERMIT DIGEST V8 #10
  5785.  
  5786. Keywords: MVS/Roscoe ETSO
  5787.  
  5788. To Anyone With Ideas,
  5789.  
  5790. We have an Amdahl 470V/8 on which we run MVS/Roscoe ETSO.  We tried
  5791. implementing the MVS/TSO version of Kermit-CMS Version 4.0, but we could never
  5792. get a file to transfer to or from an IBM PC.  We were, however, able to put
  5793. MVS Kermit into server mode and successfully execute the FINISH and HOST DIR
  5794. command from the PC's Kermit (which by the way was Kermit-MS Version 2.30).
  5795.  
  5796. When we tried file transfer, MVS seemed to acknowledge and receive the file
  5797. header, but it never could read or write the actual contents of a file.  It
  5798. would give a "BAD COMMAND" error.
  5799.  
  5800. We believe the problem is a disk allocation problem that Kermit-MVS has with
  5801. ROSCOE ETSO.
  5802.  
  5803. Any suggestions? Has anyone successfully implemented a version of Kermit on
  5804. ROSCOE ETSO?
  5805.  
  5806. One other possible avenue: Does anyone know where we could get XMODEM, YMODEM
  5807. or ZMODEM for ROSCOE??
  5808.  
  5809. Thanks.
  5810. Subir Prasad
  5811. The University of Mississippi/Computing & Information Systems
  5812. Acknowledge-To: <CCDOC@UMSVM>
  5813.  
  5814. [Ed. - Can anyone help?  It is unlikely that Xmodem or any of its derivatives
  5815. will work with MVS/ROSCOE or any IBM mainframe operating system, because it
  5816. is an 8-bit transparent protocol, and IBM mainframes always do 7-bit i/o with
  5817. ASCII devices.]
  5818.  
  5819. ------------------------------
  5820.  
  5821. Date: Thursday, 26 May 1988  08:29-MDT
  5822. From: "Wayne J. Hauber" <GA.WJH%ISUMVS.BITNET@CUNYVM.CUNY.EDU>
  5823. Subject: VT100 Terminal EMulation for TI Professional Computers?
  5824. Keywords: TI Professional Kermit
  5825.  
  5826. Is anyone aware of a communication program for the Texas Instrument
  5827. Professional computer that has a reliable VT100 emulation?  I am aware of a
  5828. Kermit that is a couple years old but which has a poor emulation.
  5829.  
  5830. The TI Pro is a MS-DOS micro that is not fully IBM compatible and does not
  5831. work with the usual MS-DOS communication programs.  My university has a number
  5832. of TI Pros.  We were wondering if there were alternatives to an aging locally
  5833. written communication program.
  5834.  
  5835. Please send responses to me directly rather than posting them.
  5836.  
  5837. Wayne Hauber
  5838. Programming Consultant
  5839. Iowa State University
  5840. Computation Center
  5841. ga.wjh@isumvs.bitnet    <-- preferred addresss
  5842.  
  5843. INFO-KERMIT DIGEST V8 #10                                              Page 103
  5844.  
  5845. gawjh@vaxd.iastate.edu  <-- a good alternate
  5846. hauber@atanasoff.cs.iastate.edu  <-- if you are on CSNET
  5847.  
  5848. [Ed. - The TI Pro version of Kermit no longer has anybody working on it.]
  5849.  
  5850. ------------------------------
  5851.  
  5852. Date: Thu, 22 Sep 88 08:58:39 +0100
  5853. From: J M Hicks <cudat@warwick.ac.uk>
  5854. Subject: A Question about the Kermit Protocol
  5855. Keywords: Kermit Protocol
  5856.  
  5857. This is a question about the Kermit protocol itself.
  5858.  
  5859. I have been working with two old implementations of Kermit.  When a CR (or, in
  5860. one implementation, XON) character arrives in between packets, they seem to
  5861. think that a bad packet has arrived and reply with a negative acknowledgement
  5862. for the expected packet or another acknowledgement of the previous packet.
  5863.  
  5864. I was under the impression that characters between packets should be
  5865. completely ignored.
  5866.  
  5867. What ought Kermit to do?  If such characters between packets appear often,
  5868. one end can give up due to all the bad packets it thinks it has received.
  5869.  
  5870. [Ed. - It should ignore any characters that appear between packets.]
  5871.  
  5872. J. M. Hicks (a.k.a. Hilary),
  5873. Computing Services, Warwick University, Coventry, England. CV4 7AL
  5874. On JANET:       cudat@UK.AC.WARWICK.CU (in the U.K.)
  5875. >From BITNET:    cudat@CU.WARWICK.AC.UK
  5876. >From Internet: try  cudat%cu.warwick.ac.uk@cunyvm.cuny.edu   (untested)
  5877. On Usenet:      ...!mcvax!ukc!warwick!cudat
  5878.                                 It helps if you spell "cudat" in lower case.
  5879.  
  5880. ------------------------------
  5881.  
  5882. Date: Sat, 13 Aug 88 00:28:13 EDT
  5883. From: Greg Woods <woods%gpu.utcs.toronto.edu@RELAY.CS.NET>
  5884. Subject: Kermit and SysVR3 layers.
  5885. Keywords: SysVR3 Layers
  5886.  
  5887. Is anyone doing, or interested in doing, something about getting kermit
  5888. to talk to Sys V R3 layers?  I would find that an EXTREMELY important
  5889. feature.
  5890.  
  5891. Since a lot of PC's have graphics and mice these days, a graphical
  5892. version would be even better.  It would sure cut the entry-level cost of
  5893. using layers (for those of us with access to a PC).
  5894.  
  5895.                         Greg Woods.
  5896.  
  5897. UUCP: utgpu!woods, utgpu!{ontmoh, ontmoh!ixpierre}!woods
  5898. VOICE: (416) 242-7572 [h]        LOCATION: Toronto, Ontario, Canada
  5899.  
  5900. ------------------------------
  5901.  
  5902. Page 104                                              INFO-KERMIT DIGEST V8 #10
  5903.  
  5904.  
  5905. End of Info-Kermit Digest
  5906.  
  5907. INFO-KERMIT DIGEST V8 #11                                              Page 105
  5908.  
  5909. Info-Kermit Digest         Thu, 15 Dec 1988       Volume 8 : Number 11
  5910.  
  5911. Today's Topics:
  5912.         Announcing a New Kermit for Microsoft Windows
  5913.         Announcing Version 3.1 of Turbo Pascal Kermit
  5914.               New xxu.c Utility Program
  5915.             Some responses to V8#10 and Kermit-370
  5916.              Kermit 370 Trailing Blanks Trimming
  5917.                 VMS C-Kermit Dialing?
  5918.         C-Kermit Problems on SUNs with SUNOS Version 4
  5919.             Directory Locking in C-Kermit?
  5920.                 VAX/VMS Kermit via x25
  5921.        Response to Request of vt100 Setting Macro for MacKermit 0.9(40)
  5922.              MacKermit vt100 Initialization Files
  5923.                MacKermit Settings File
  5924.               Kermit for Honeywell DPS-6/GCOS6?
  5925.           Kermit wanted for Burroughs B21 (Not B25)?
  5926.  
  5927. Send digest submissions to Info-Kermit@CUNIXC.CC.COLUMBIA.EDU, requests for
  5928. addition to or deletion from the Info-Kermit subscriber list to
  5929. Info-Kermit-Request@CUNIXC.CC.COLUMBIA.EDU.
  5930.  
  5931. Kermit files may be obtained over networks and by mail order.  On the
  5932. Internetwork, use FTP to log in to host CUNIXC, CUNIXC.CC.COLUMBIA.EDU, a VAX
  5933. 8700 running UNIX (Ultrix).  The IP host number is 128.59.40.130.  You should
  5934. be able to to FTP to CUNIXC, login as user anonymous (lower case user name
  5935. with any password) and GET the desired files.  The Kermit files are in
  5936. directories kermit/a, kermit/b, kermit/c, kermit/d, and kermit/e.  You can
  5937. also get Kermit files over BITNET/EARN; to get started send a message with
  5938. text HELP to KERMSRV, the Kermit file server, at host CUVMA.  For detailed
  5939. instructions, read the file k1/aanetw.hlp (AANETW HLP on KERMSRV).  To order
  5940. by mail, request a complete list of Kermit versions and an order form from
  5941. Kermit Distribution, Columbia University Center for Computing Activities, 612
  5942. West 115th Street, New York, NY 10025 USA.
  5943.  
  5944. ----------------------------------------------------------------------
  5945.  
  5946. Date: Fri, 9 Dec 1988 16:40:00 EST
  5947. From: Christine M. Gianone <cmg@cunixc.cc.columbia.edu>
  5948. Subject: Announcing a New Kermit for Microsoft Windows
  5949. Keywords: IBM PC, MS-Windows, JAR files
  5950.  
  5951. This is to announce a new version of Kermit for Microsoft Windows.  The
  5952. program is written in Microsoft C and should run on any computer that has
  5953. Microsoft Windows version 2.03 or later, or Windows/286, or Windows/386.  It
  5954. was written by Bill Hall of Santa Clara, CA, and is a complete replacement for
  5955. the earlier, primitive WINKERM contributed by Bill a couple years ago.  This
  5956. version includes a Heath/Zenith-19 terminal emulator, fully integrated with
  5957. the Windows environment (so you can cut & paste, etc).  This Kermit program
  5958. can run in one window while you are doing some other work in another window.
  5959. Thus, file transfers can proceed in the background.  In fact, multiple
  5960. instances of MS-Windows Kermit can be active at the same time (but of course,
  5961. each requires a separate communication port).
  5962.  
  5963. As submitted by Bill, the files have names that don't fit into the Kermit
  5964. Distribution naming conventions -- He renamed the program from WINKRM to
  5965.  
  5966. Page 106                                              INFO-KERMIT DIGEST V8 #11
  5967.  
  5968. WKERM, but unfortunately there already is a WKERMIT (the sliding windows
  5969. version of Kermit for PC).  But since the names are important (e.g. the
  5970. program itself must be called WKERM.EXE to allow code sharing for multiple
  5971. instances), the files have been concatenated into "text archives", which we
  5972. call "jar" files ("join archive").  The explanation of the format of the jar
  5973. file is found at the beginning of the file, along with a C program for taking
  5974. it apart again (you can also use a text editor).
  5975.  
  5976. Here's a brief description of the archives and the files within them:
  5977.  
  5978. WINKRM.JAR:
  5979.  
  5980.   README.1ST - A "read me first" file.
  5981.  
  5982.   WKERM.EXE - A Windows terminal emulator for the Heath/Zenith H/Z-19 terminal
  5983. with the Kermit file transfer protocol added.  (This has been encoded in "boo"
  5984. format as WKERM.BOO, and can be decoded back into .EXE form using any of the
  5985. MSBPCT programs.)
  5986.  
  5987.   HEATH19.FON - An optional ASCII font also containing the original H-19
  5988. graphics characters which can be loaded with the Windows control panel and
  5989. used with the H-19 Windows terminal emulator.  (This has been encoded in "boo"
  5990. format as HEATH19.BOO, and can be decoded back into .EXE form using any of the
  5991. MSBPCT programs.)
  5992.  
  5993.   DEMO.C - a C-program which can be used to illustrate some terminal functions
  5994. and the H-19 font.  Please read the comments in that file for more
  5995. information.
  5996.  
  5997.   WINH19.DOC - Documentation for the H-19 terminal emulator portion of Windows
  5998. Kermit.
  5999.  
  6000.   WKERM.DOC - Supplementary information on how to use the Kermit feature of
  6001. the emulator.
  6002.  
  6003.   WINH19.APP - An appendix to the WINH19.DOC file giving detailed technical
  6004. information about keyboard usage.
  6005.  
  6006.   TERMCAPS.H19 - a suggested UNIX TERMCAPS file for the H-19 emulator.  Both
  6007. 24 and 30 line versions are illustrated for ANSI and HEATH modes.
  6008.  
  6009. WINSRC.JAR:
  6010.  
  6011.   Source files for WKERM in C, plus header files, compile and link procedures.
  6012.  
  6013. Thanks to Bill for all the work that went into this package and for
  6014. contributing it to Kermit Distribution.  Bill's network mail address (as seen
  6015. from Columbia) is:
  6016.  
  6017. rochester!ames!olivej.ATC.OLIVETTI.COM!whall@columbia.edu
  6018.  
  6019. ------------------------------
  6020.  
  6021. Date: Wed, 14 Dec 1988 17:40:00 EST
  6022. From: Christine M. Gianone <cmg@cunixc.cc.columbia.edu>
  6023. Subject: Announcing Version 3.1 of Turbo Pascal Kermit
  6024.  
  6025. INFO-KERMIT DIGEST V8 #11                                              Page 107
  6026.  
  6027. Keywords: IBM PC, Turbo Pascal, Tektronix 4100, QK Kermit
  6028.  
  6029. Victor Lee, VIC@QUCDN.BITNET, of Queens University, Kingston, Ontario, has
  6030. submitted a new release of QK-Kermit, the Turbo Pascal that runs under MS-DOS
  6031. on the IBM PC and PS/2 and compatibles.  The new release, 3.1, now emulates
  6032. the Tektronix 4100 rather than the 4010, so that it is capable of color
  6033. graphics.  Many thanks to Vic for this submission.
  6034.  
  6035. The files are in kermit/a/qk3*.* on CUNIXC, available via anonymous FTP, and
  6036. as QK3* * from KERMSRV at CUVMA on BITNET.
  6037.  
  6038. ------------------------------
  6039.  
  6040. Date: Tuesday, 15 November 1988 09:22-MST
  6041. From: steinmetz!crdos1!davidsen@uunet.UU.NET (Wm E. Davidsen)
  6042. Subject: New xxu.c Utility Program
  6043. Keywords: xxu.c, Unix, Filenames
  6044.  
  6045. I was looking at the "xxu" program you posted recently, and I added a few
  6046. changes for SysV and Xenix. I also changed the logic to turn blanks into
  6047. underscores rather than keep or delete them. I could probably find the info on
  6048. ftp upload, but this is small enough to mail and I have to ftp via another
  6049. machine anyway.
  6050.  
  6051. I don't have problems with my ftp names, but I do get programs which create
  6052. really bad filenames at times. This is an easy way to clear them out.
  6053.  
  6054. bill davidsen        (wedu@ge-crd.ARPA -or- davidsen@crdos1.uucp)
  6055.             {uunet | philabs}!steinmetz!crdos1!davidsen
  6056.  
  6057. [Ed. - Thanks for the update Bill.  And thanks to Keith Petersen for
  6058. submitting the program to Info-Kermit.  The file has replaced the old one in
  6059. kermit/b/xxu.c]
  6060.  
  6061. ------------------------------
  6062.  
  6063. Date: 1988 Dec 5   22:46 EST
  6064. From: (John F. Chandler) PEPMNT@CFAAMP.BITNET
  6065. Subject: Some Responses to V8#10 and Kermit-370
  6066. Keywords: Kermit-370
  6067.  
  6068. Angelo De Florio wrote concerning SIM3278:
  6069.  
  6070. > Has someone tried file-transfer using SIM3278/VM or SIM3278/GCS with
  6071. > Kermit-CMS ?
  6072.  
  6073. Yes, it has been tried, but I don't believe it is possible, since SIM3278
  6074. doesn't have any transparent READ capability, only a WRITE.
  6075.  
  6076. Peter Jones wrote concerning Long Packet Size on IBM 370 Mainframe:
  6077.  
  6078. > There seem to be a small problem with SET RECEIVE PACKSIZE and SET SEND
  6079. > PACKSIZE. SET RECEIVE seems to be OK, but changes both RECEIVE and SEND size.
  6080. > SET SEND gives a diagnostic, but the SHOW PROTOCOL command shows that both
  6081. > sizes have been altered.  These remarks apply to Version 4.0 of mainframe
  6082. > KERMIT.
  6083.  
  6084. Page 108                                              INFO-KERMIT DIGEST V8 #11
  6085.  
  6086.  
  6087. Beyond the editorial response in the digest, I must point out a couple things.
  6088. First, SET SEND PACK has only one function in Kermit-370 -- it provides the
  6089. default in case the other Kermit leaves the packet size blank in its
  6090. negotiations.  If you wish to restrict the size of packets to be sent from a
  6091. given Kermit, you may tell the *other* Kermit to insist on a smaller packet
  6092. size.  Second, SHOW SEND PACK (like SHOW for other parameters that do nothing
  6093. but set defaults) reveals just the latest value defined -- whether by the SET
  6094. subcommand or by negotiating with another Kermit.  The diagnostic message for
  6095. SEND PACK > 94 is just a reminder that the only functional values are 20-94.
  6096.  
  6097. Subir Prasad wrote about Kermit for MVS/Roscoe ETSO:
  6098.  
  6099. > When we tried file transfer, MVS seemed to acknowledge and receive the file
  6100. > header, but it never could read or write the actual contents of a file.  It
  6101. > would give a "BAD COMMAND" error.
  6102. >
  6103. > We believe the problem is a disk allocation problem that Kermit-MVS has with
  6104. > ROSCOE ETSO.
  6105.  
  6106. However, I have heard from John Lundberg that "TSO Kermit actually seems to
  6107. work rather well in ROSCOE/ETSO as long as one doesn't try to create a host
  6108. data set name that violates ACF2's rule."
  6109.  
  6110. J M Hicks asked A Question about the Kermit Protocol:
  6111.  
  6112. > I was under the impression that characters between packets should be
  6113. > completely ignored.
  6114. >
  6115. > What ought Kermit to do?  If such characters between packets appear often,
  6116. > one end can give up due to all the bad packets it thinks it has received.
  6117.  
  6118. While the editor pointed out (correctly) that Kermits *should* ignore
  6119. characters between packets, half-duplex systems (such as IBM mainframes)
  6120. receive RECORDS, rather than PACKETS.  In such circumstances, an extra CR is
  6121. evidence of a lost or garbled packet, and, considering that some IBM mainframe
  6122. Kermit applications are unable to time out, it is clearly risky to ignore a
  6123. bare CR.
  6124.                                    John
  6125.  
  6126. ------------------------------
  6127.  
  6128. Date: Tue, 22 Nov 1988 10:11:55 +0100
  6129. From: Andre' PIRARD <A-PIRARD%BLIULG11.BITNET@cuvmb.cc.columbia.edu>
  6130. Subject: Kermit 370 Trailing Blanks Trimming
  6131. Keywords: Kermit-370
  6132.  
  6133. A proposal was accepted someday not to trim trailing blanks off variable
  6134. length text records sent by 370 Kermit.  I just found a little hitch to this
  6135. otherwise sensible decision.
  6136.  
  6137. God knows why, the CMS file system hates null-length records (as well as empty
  6138. files). CMS Kermit has consequently no other way to store a null line than to
  6139. make it a single blank. On its way back, this blank remains.
  6140.  
  6141. Software on micros can be sensitive to a record size being null and I used to
  6142.  
  6143. INFO-KERMIT DIGEST V8 #11                                              Page 109
  6144.  
  6145. be pleased to see my long dated first tests of CMS Kermit get a file back
  6146. correctly, as opposed to what we used before.
  6147.  
  6148. It's an occasional nuisance only, but I wonder if it wouldn't be nice to
  6149. consider a record containing a single blank as the result of the above and, by
  6150. exception, send it with null length.  I can't think of another way to bypass
  6151. this "feature".  The "other way round" nuisance (single blanks converted to
  6152. nulls) should be much less frequent in my opinion.
  6153.  
  6154. Andr).
  6155.  
  6156. ------------------------------
  6157.  
  6158. Date: Mon, 21 Nov 88 23:35:46 EST
  6159. From: Mark F. Proudman <mfp%sobeco.UUCP@larry.mcrcim.mcgill.edu>
  6160. Subject: VMS C-Kermit Dialing?
  6161. Keywords: C-Kermit 4E(070), setjmp(), C-Kermit/VMS, VAX/VMS
  6162.  
  6163. I have been trying to get C-Kermit's dialer (ckudia.c) working under VMS.
  6164. VMS does not like the setjmp() call within a setjmp that handles failures
  6165. during an exception (ie a failure to close the line after a timeout).  (This
  6166. is documented in the VAXC RTL manual).  Has anyone else re-implemented this
  6167. for VMS?
  6168.  
  6169. Thanks in advance.
  6170.  
  6171. Mark Proudman    uunet!attcan!sobeco!mfp
  6172.         (514) 878 9090
  6173. "It is agreed that the ice is thin here" - K&R.
  6174.  
  6175. [Ed. - Not us...  Various people have started working on VMS support in
  6176. C-Kermit, and no one -- to our knowledge -- has been able to complete the
  6177. job.  Is there anyone we don't know about who's working in this area???]
  6178.  
  6179. ------------------------------
  6180.  
  6181. Date: Tue, 1 Nov 88 20:26:31 EST
  6182. From: ksmith@cubmol (Kenneth Smith)
  6183. Subject: C-Kermit Problems on SUNs with SUNOS Version 4
  6184. Keywords: C-Kermit 4E(070), SUN, SUNOS 4.0
  6185.  
  6186. Hardware:  Sun 3/160, connected to IBM-PC/AT and IBM-PC.
  6187.  
  6188. Software:  SunOS 4.0, C-Kermit 4E(70).
  6189.        MSKERMIT 2.31 on the IBM-PC/AT,
  6190.        MSKERMIT 2.29 on the IBM-PC.
  6191.  
  6192. Problem: When I attempt to send a file from either PC to the Sun3, the
  6193. transfer hangs at a random point.  When I send a file from the Sun3 to
  6194. either PC, transfer is normal.  The problem occurs when the PC's are
  6195. connected to the MTI multiplexer card.  The problem DOES NOT occur for a PC
  6196. connected to one of the Sun3 hardwired serial ports.  For a PC connected to
  6197. the multiplexer, file transfer was tested for failure at baud rates from
  6198. 2400 to 19200. On the hardwired serial line, file transfer is successful at
  6199. 19200 baud.
  6200.  
  6201.  
  6202. Page 110                                              INFO-KERMIT DIGEST V8 #11
  6203.  
  6204. Comment: On a SUN3 running SUN UNIX version 3.4, and using an identical
  6205. multiplexer, we have no problems with C-Kermit.  It is possible that the
  6206. problem may be something on our multiplexer board, but I would like to know
  6207. if you have received any similar complaints about problems under SunOS 4.0,
  6208. or perhaps have even found a cure.
  6209.  
  6210.                         Ken Smith
  6211.                         Dept. of Biology
  6212.                         Computer Graphics Facility
  6213.                         854-6122
  6214.                         ksmith@cubmol.bio.columbia.edu
  6215.  
  6216. [Ed. - Complaints galore, including from ourselves!  There are bugs in the
  6217. MTI and ALM handlers in SUNOS 4.0.  On our SUN-4's, we could only use Kermit
  6218. through Ethernet terminal servers.  Release 4.0.1 of SUNOS is supposed to
  6219. fix these problems, but we haven't tested it.]
  6220.  
  6221. ------------------------------
  6222.  
  6223. Date: 28 Nov 88 22:21:12 GMT
  6224. From: execu!dewey@gatech.edu (Dewey Henize)
  6225. Subject: Directory locking in C-Kermit?
  6226. Keywords: C-Kermit 4E(070), Directory Locking
  6227.  
  6228. Anyone out there have kermit installed on a Sun system, running SunOS4.0.1?
  6229.  
  6230. we are currently doing so, and things are basically working well, with
  6231. kermit version: C-Kermit, 4E(070) 29 Jan 88, 4.2 BSD
  6232.  
  6233. except for one small problem.  We seem to get the actual lock file built
  6234. just fine when we select a device [set line /dev/ttyx], but we still get
  6235. the message: Warning, write access to lock directory denied
  6236.  
  6237. I've checked the protections as well as I know to, and (at least temporarily)
  6238. have it set up exactly like tip (which connects fine, but without the
  6239. warning message), ie "-rws--x--x  1 uucp     daemon".
  6240.  
  6241. I would prefer not to have it suid anyhow, obviously.  Anyone out there
  6242. got a pointer for me?  All help appreciated, and thanks in advance.
  6243.  
  6244. Dewey Henize
  6245. execu!dewey
  6246.  
  6247. [Ed. - Like it says in the manual, the lock file directory must be publicly
  6248. readable and writable.  The only alternative is the suid business, which in
  6249. these days of worms and viruses, is to be avoided.]
  6250.  
  6251. ------------------------------
  6252.  
  6253. Date:  Thu, 24 Nov 88 13:44 MST
  6254. From: kieffer@UNCANET.BITNET (Rom Kieffer, Academic Computing Services)
  6255. Subject: VAX/VMS Kermit via x25
  6256. Keywords: VAX/VMS Kermit, X25
  6257.  
  6258. I am faced with a remote site trying to use kermit over a packet network, and
  6259. not having any idea about how to route traffic to the packet interface card
  6260.  
  6261. INFO-KERMIT DIGEST V8 #11                                              Page 111
  6262.  
  6263. (DUP11) instead of the standard DZ11.
  6264.  
  6265. It is easy to connect via the carrier to the VAX, to start kermit and iniate
  6266. transfers, but the reverse has me stumped.  How does one start kermit, then
  6267. allocate a DUP11/PSI virtual circuit, connect via the packet carrier to the
  6268. remote system?
  6269.  
  6270. (decnet is out, the partner system is a NOS/VE Cyber 835, hence the kermit
  6271. approach)
  6272.  
  6273. If anybody out there has a better idea, we are not proud.
  6274.  
  6275. I have tried finding out whether there was a file transfer utility common to
  6276. VMS and NOS/VE, but to no avail so far. I have heard of RHF (Remote Host
  6277. Facility) but I cannot get any details about cost, availability or
  6278. performance.
  6279.  
  6280. any pointers are greatly appreciated.
  6281.  
  6282.  Rom Kieffer (403)-220-6210
  6283.  Manager, datacommunications
  6284.  Academic Computing Services
  6285.  The University of Calgary
  6286.  Calgary, Alberta
  6287.  CANADA T2N 1N4
  6288.  
  6289.  Kieffer%ucnet.unca.adhocnet.ca@uncaedu.bitnet
  6290.  
  6291. ------------------------------
  6292.  
  6293. Date: 11 Aug 1988 00:09:13 EST
  6294. From: Eric.Yang@EWT.MIT.EDU
  6295. Subject: Response to Request of vt100 Setting Macro for MacKermit 0.9(40)
  6296. Keywords: MacKermit 0.9(40), VT100 Settings
  6297.  
  6298. The following is a Binhexed version of the setting on my SE with extended
  6299. keyboard.  The VT100 keypad and cursors are mapped correctly.  In addition, I
  6300. have implemented the six-key feature that exists in VT200 mode.  F9-F12
  6301. on the top row is encoded as <esc>, <BS>, <LF>, and (ESC>[26~, respectively.
  6302. Although this setting violates DEC convention by mixing VT100 and VT200
  6303. codes, I found it more powerful than a true VT220 when I am using EVE or EDT.
  6304.  
  6305. [Ed. - Thanks!  Your settings file has been added to kermit/b/ckmker.set,
  6306. which is a collection of settings files for MacKermit 0.9.]
  6307.  
  6308. ------------------------------
  6309.  
  6310. Date: Thu, 11 Aug 88 16:14:36 PDT
  6311. From: zz1ml%sdcc3@ucsd.edu (Mick Laver)
  6312. Subject: MacKermit vt100 Initialization Files
  6313. Keywords: MacKermit 0.9(40), Initialization File
  6314.  
  6315. Here are two MacKermit 0.9(040) vt100 initialization files to remap the
  6316. MacPlus and Mac II keypads, and a short MacWrite doc that explains how the
  6317. keys are mapped. All are in Binhex 4.0 format.  The files are:
  6318.  
  6319.  
  6320. Page 112                                              INFO-KERMIT DIGEST V8 #11
  6321.  
  6322.     macvtdoc.hqx    Macwrite doc (11.2K)
  6323.     maciivt.hqx     Mac II vt100 init file (1.3K)
  6324.     macplvt.hqx     Mac Plus vt100 init file (1.3K)
  6325.  
  6326. Since the files are small I've stuck them one after the other below. I hope
  6327. someone finds them useful.
  6328.  
  6329. Mick Laver, C-010                 Internet:         mlaver@ucsd.edu
  6330. UCSD Academic Computing Center    UUCP:             ...!ucsd!mlaver
  6331. La Jolla, CA.92093                BITNET:        mlaver@ucsd.BITNET
  6332.  
  6333. [Ed. - Thanks!  These settings files have also been added to
  6334. kermit/b/ckmker.set.]
  6335.  
  6336. ------------------------------
  6337.  
  6338. Date: Fri, 12 Aug 88 16:58:25 EDT
  6339. From: Bob Rahe <CES00661%UDACSVM.BITNET@CUVMB.CC.COLUMBIA.EDU>
  6340. Subject: MacKermit Settings File
  6341. Keywords: MacKermit 0.9(40), VT100 Settings
  6342.  
  6343.   Someone asked for a settings file that would work with MacKermit 0.9(40) and
  6344. IBM mainframes as a VT100 through some kind of protocol convertor simulating a
  6345. 327x terminal.  Here's one that I use to a VM/CMS system through a series-1
  6346. front end.  PF1 thru 9 are the keypad 1 thru 9 keys on a Mac+.  The Clear key
  6347. (and ENTER key) also work.  Probably not complete but does the (minimal) stuff
  6348. I need it to do.
  6349.  
  6350.   One source of info I remember using to get the codes right was the sample
  6351. init file for the PC version of Kermit.  See what he maps things to and you
  6352. can add whatever else you need.  This will give you a head start tho.
  6353.  
  6354.      Enjoy,
  6355.        Bob
  6356.  
  6357. [Ed. - Thanks!  These settings files have been added to kermit/b/ckmker.set
  6358. too.]
  6359.  
  6360. ------------------------------
  6361.  
  6362. Date: Thu, 15 Dec 1988 09:54:00 EST
  6363. From: Christine M. Gianone <cmg@cunixc.cc.columbia.edu>
  6364. Subject: Kermit for Honeywell DPS-6/GCOS6?
  6365. Keywords: Honeywell, DPS-6, GCOS6
  6366.  
  6367. For several years, we have been distributing a version of Kermit for the
  6368. Honeywell DPS-6 GCOS6 operating system.  It was donated anonymously and
  6369. came without source code or documentation.  Does anyone have, or know of,
  6370. the source to this program, or any other Kermit program for the DPS-6?
  6371.  
  6372. ------------------------------
  6373.  
  6374. Date: Thu, 1 Sep 88 10:16:35 EDT
  6375. From: ciaraldi@cs.rochester.edu
  6376. Subject: Kermit wanted for Burroughs B21 (Not B25)?
  6377. Keywords: Burroughs B21 Kermit
  6378.  
  6379. INFO-KERMIT DIGEST V8 #11                                              Page 113
  6380.  
  6381.  
  6382. I'm looking for a Kermit that would run on the Burroughs B21 series.  This was
  6383. made for them by Convergent Technologies and runs an operating system called
  6384. BTOS.
  6385.  
  6386. I have used the B20 Kermit already available on a B25, but I don't think this
  6387. will work on the B21.
  6388.  
  6389. The B20 series is divided into two families, the older B21 and the newer B25.
  6390. They both run BTOS, but different versions, because the CPU is different.  I
  6391. have heard that the two families are mostly source-code compatible, but not
  6392. object-code compatible.
  6393.  
  6394. Does anyone know of a Kermit or any other terminal emulation program that will
  6395. run on the B21's?  Thanks.
  6396.  
  6397. Mike Ciaraldi
  6398. University of Rochester
  6399. ARPA:  ciaraldi@cs.rochester.edu
  6400. uucp:  ...rutgers!rochester!ciaraldi
  6401.  
  6402. ------------------------------
  6403.  
  6404. End of Info-Kermit Digest
  6405.  
  6406. Page 114                                              INFO-KERMIT DIGEST V8 #11
  6407.  
  6408.  
  6409. Index                                                                  Page 115
  6410.  
  6411. .BOO Files, 96
  6412. 3270 Emulation, 96
  6413. 370 Kermit, 93
  6414. 3COM EtherSeries, 39
  6415. 7171, 55, 96
  6416. Amiga, 59
  6417. Apple II, 37, 52
  6418. Archimedes, 17
  6419. AT&T 6300, 25
  6420. Attributes, 77
  6421. BOO Files, 86
  6422. BSD 2.10, 60
  6423. BTOS, 5
  6424. Burroughs B20, 5
  6425. Burroughs B21 Kermit, 113
  6426. C-Kermit, 8, 13, 27, 58, 60, 62, 67, 101
  6427. C-Kermit 4E, 60
  6428. C-Kermit 4E(070), 61, 64, 109-110
  6429. C-Kermit 4E(070) Performance, 63
  6430. C-Kermit/Amiga, 59
  6431. C-Kermit/VMS, 67-68, 109
  6432. C64 Kermit, 12
  6433. CGA, 20
  6434. CMS Kermit, 53
  6435. Color Graphics, 23
  6436. COM3, 45
  6437. Commodore Amiga, 59
  6438. Convergent NGEN, 5
  6439. CP/M Kermit, 99
  6440. Cray X-MP, 62
  6441. CTOS, 5
  6442. CU20B, 26, 36, 48
  6443. CUNIXC, 48
  6444. CX-80, 53, 55
  6445. DEC Rainbow, 80
  6446. DECnet/DOS, 80
  6447. DG/1, 16-17
  6448. Directory Locking, 110
  6449. DPS-6, 112
  6450. EASYK, 81
  6451. EasyK Kermit, 96
  6452. EPSON PX8, 97-98
  6453. Escape Character, 21
  6454. File Dates, 77
  6455. Filenames, 107
  6456. FTP, 26, 36, 48
  6457. Fujitsu FMR-60, 50
  6458. Future Kermits, 27
  6459. GCOS6, 112
  6460. Generic Kermit, 18
  6461. GNU EMACS, 62
  6462. GOTO, 76
  6463. Graphics, 89
  6464. Heath-Zenith/100, 15
  6465. Heath/Zenith Z100, 94
  6466. Hebrew, 81
  6467.  
  6468. Page 116                                                                  Index
  6469.  
  6470. Honeywell, 112
  6471. IBM 370 Kermit, 93
  6472. IBM AEA Ports, 86
  6473. IBM PC, 49, 80, 105, 107
  6474. Initialization File, 111
  6475. Initialization Files, 22
  6476. Internet, 36
  6477. IP Host Numbers, 36
  6478. Japan, 50
  6479. JAR files, 105
  6480. Kana, 50
  6481. Kanji, 50
  6482. Kermit Distribution, 48
  6483. Kermit Protocol, 13, 103
  6484. Kermit-370, 6, 53, 56, 77, 97, 107-108
  6485. Kermit-65, 52
  6486. Kermit-80, 99
  6487. Kermit-CMS, 97
  6488. Lock Files, 60
  6489. MacBinary, 46
  6490. MacKermit, 8, 10
  6491. MacKermit 0.9(36)b4, 9
  6492. MacKermit 0.9(40), 7-9, 11-12, 46-47, 96, 111-112
  6493. Mailer Problems, 79
  6494. Masscomp, 60
  6495. Microsoft Windows, 81
  6496. MS Windows, 81
  6497. MS-DOS Kermit, 19, 21-22, 45
  6498. MS-DOS Kermit 2.29d, 15
  6499. MS-DOS Kermit 2.30, 16-18, 30, 81
  6500. MS-DOS Kermit 2.31, 1, 19-25, 28-31, 33-34, 38-44, 49-51, 70, 72-73, 76-77,
  6501.  81-84, 86-89, 91, 100
  6502. MS-DOS Kermit 2.31 Manual, 38
  6503. MS-DOS Kermit 2.31/A, 80
  6504. MS-Kermit, 19
  6505. Ms-Kermit 2.31, 20
  6506. MS-Kermit Z-100 MSTZ10, 94
  6507. MS-Windows, 105
  6508. MVS/Roscoe ETSO, 102
  6509. MVS/TSO, 93
  6510. NEC-PC9801, 50
  6511. Network, 48
  6512. NorthStar Advantage, 99
  6513. Novell File Server, 72
  6514. Okstate, 26, 37
  6515. Olivetti, 76
  6516. OS/2 Kermit, 27, 90, 95
  6517. p-System, 69
  6518. Pecan, 69
  6519. Portable 370 Kermit, 53
  6520. Printer, 39
  6521. ProDOS, 37
  6522. Protocol Converters, 53, 55-56, 77
  6523. QK Kermit, 107
  6524. Rainbow, 80
  6525. Rainbow Kermit, 45
  6526.  
  6527. Index                                                                  Page 117
  6528.  
  6529. RedRyder, 10
  6530. ReGIS Color Graphics, 86
  6531. RTU, 60
  6532. SAS, 51
  6533. Scan Codes, 16
  6534. SCO Xenix, 65-66
  6535. Screen Rollback, 20-21
  6536. Script Files, 82
  6537. Scripts, 34
  6538. Series/1, 96
  6539. setjmp(), 109
  6540. Shell, 62
  6541. SIM3278, 97
  6542. Smartcom, 23
  6543. SmartLink., 40
  6544. SUN, 109
  6545. SUNOS 4.0, 109
  6546. Super Serial Card, 52
  6547. SysVR3 Layers, 103
  6548. Tabs, 41, 73
  6549. Take Files, 19
  6550. Tandy 3000HD, 24
  6551. Tektronix 4100, 107
  6552. Tektronix Emulation, 19, 25
  6553. Terminal Emulation, 23
  6554. TI Professional Kermit, 102
  6555. Transparent Printing, 20
  6556. TSO, 55
  6557. TSO Kermit, 93
  6558. Turbo Pascal, 107
  6559. UCSD Pascal Kermit, 69
  6560. Ungermann-Bass, 31, 33, 49
  6561. UNICOS, 62
  6562. Unix, 107
  6563. UUCP, 37
  6564. UUCP Lock Files, 60
  6565. VAX/VMS, 67-68, 109
  6566. VAX/VMS Kermit, 51, 100-101, 110
  6567. Victor 9000, 18
  6568. VT100 PF Keys, 19
  6569. VT100 Settings, 111-112
  6570. VT200 Support, 42
  6571. Wang PC, 51
  6572. X.25, 67
  6573. X.29 Module, 101
  6574. X25, 110
  6575. Xerox 16/8, 99
  6576. xxu.c, 107
  6577. Z-100, 44
  6578. Zenith Z-248, 45
  6579. Zenith Z100, 94
  6580. ZPC, 44
  6581.  
  6582. Page 118                                                                  Index
  6583.  
  6584.