home *** CD-ROM | disk | FTP | other *** search
/ Columbia Kermit / kermit.zip / e / imail.90b < prev    next >
Text File  |  2020-01-01  |  188KB  |  4,180 lines

  1.  
  2.  
  3.  
  4.  
  5.  
  6.  
  7.  
  8.  
  9.  
  10.  
  11.  
  12.  
  13.  
  14.  
  15.  
  16.  
  17.  
  18.  
  19.  
  20.  
  21.  
  22.          Columbia University Center for Computing Activities
  23.  
  24.  
  25.  
  26.                        
  27.                   INFO-KERMIT DIGEST
  28.  
  29.                    VOLUME 12
  30.  
  31.              Christine M. Gianone, Editor
  32.  
  33.                  July - December 1990
  34.  
  35.  
  36.  
  37. Table of Contents
  38.  
  39.  
  40. Volume 12, Number 1                                                           1
  41.   Updated MS-DOS 3.02 Test Version Available                                  1
  42.   New Release of Kermit for HP-3000                                           2
  43.   Announcing NCR 9800 Kermit                                                  3
  44.   New Release of Kermit for UNISYS (Burroughs) A-Series                       4
  45.   New Kermit Programs for Luxor ABC Series and FACIT DTC/DTC2                 4
  46.   New MS Kermit REMOTE PRINT interface with Kermit-370                        5
  47. Volume 12, Number 2                                                          10
  48.   Another Update of the MS-DOS Kermit 3.02 Test Release                      10
  49.   Version 8.12 of PRIME Kermit Now Available                                 13
  50.   New Documentation Available for VAX/VMS Kermit-32 3.3.126                  13
  51.   New Test Release of DEC Rainbow Kermit                                     13
  52. Volume 12, Number 3                                                          18
  53.   Proposed Kermit Extension for SET FILE TYPE                                18
  54.   And ANOTHER New Test Release of MS-DOS Kermit 3.02                         22
  55.   A New Version of Kermit for OS/2 Presentation Manager                      26
  56.   New Release UCPECAN Kermit V1.1                                            27
  57. Volume 12, Number 4                                                          30
  58.   Next Test Release of MS-DOS Kermit 3.02                                    30
  59. Volume 12, Number 5                                                          41
  60.   Announcing IBM Mainframe VM/CMS Kermit-370 Version 4.2.1                   41
  61.   Announcing  KERMIT-12 Version 10g                                          43
  62.   Kermit Proposal SET FILE TYPE                                              45
  63.   Re: A New Version of Kermit for OS/2 Presentation Manager                  47
  64. Volume 12, Number 6                                                          51
  65.   New Honeywell DPS-6 Superkermit                                            51
  66.   Announcing Kermit for the Tandy 100                                        53
  67. Volume 12, Number 7                                                          62
  68.   Announcing IBM Mainframe MVS/TSO Kermit-370 Version 4.2.1                  62
  69.   Announcing CP/M Kermit Version 4.10 for Testing                            64
  70.   New Test Release of Microsoft Windows Kermit                               65
  71.   Another New Prerelease of MS-DOS Kermit 3.02                               66
  72.   New Kermit for Gould/SEL 32/77 MPX                                         70
  73.  
  74.  
  75. INFO-KERMIT DIGEST V12 #1                                                Page 1
  76.  
  77. Info-Kermit Digest         Wed, 18 Jul 1990        Volume 12 : Number 1
  78.  
  79. Departments:
  80.  
  81.   ANNOUNCEMENTS -
  82.         Updated MS-DOS 3.02 Test Version Available
  83.         New Release of Kermit for HP-3000
  84.         Announcing NCR 9800 Kermit
  85.         New Release of Kermit for UNISYS (Burroughs) A-Series
  86.         New Kermit Programs for Luxor ABC Series and FACIT DTC/DTC2
  87.  
  88.   IBM MAINFRAME KERMIT -
  89.         New MS Kermit REMOTE PRINT interface with Kermit-370
  90.         Kermit-370 support for IBM 3174 AEA
  91.  
  92.   VAX/VMS KERMIT -
  93.         VMS Kermit-32 Hints
  94.         How to Create a Kermit Connection over X.25 from VAX/VMS?
  95.  
  96.   MISCELLANY -
  97.         Kermit for Microsoft Windows
  98.         External Kermit Driver
  99.         Kermit for Atari Portfolio?
  100.  
  101. Digest submissions may be sent to Info-Kermit@WATSUN.CC.COLUMBIA.EDU,
  102. requests for addition to or deletion from the Info-Kermit subscriber list to
  103. Info-Kermit-Request@WATSUN.CC.COLUMBIA.EDU or to KERMIT@CUVMA.BITNET.
  104.  
  105. Kermit files may be obtained over networks and by mail order.  On the
  106. Internetwork, use FTP to log in to host WATSUN.CC.COLUMBIA.EDU, a SUN-4/280
  107. running UNIX (SUNOS 4.1), IP host number 128.59.39.2.  Login as user anonymous
  108. (note, lower case), any password, and GET or MGET (MULTIPLE GET) the desired
  109. files.  The Kermit files are in directories kermit/a, kermit/b, kermit/c,
  110. kermit/d, and kermit/e.  Test versions are in kermit/test.  Binaries are in
  111. kermit/bin (use ftp in binary mode).  You can also get Kermit files over the
  112. BITNET/EARN network; to get started send a message with text HELP to KERMSRV,
  113. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  114. file kermit/a/aanetw.hlp (AANETW.HLP on KERMSRV).  To order by mail, request a
  115. complete list of Kermit versions and an order form from Kermit Distribution,
  116. Columbia University Center for Computing Activities, 612 West 115th Street,
  117. New York, NY 10025 USA.
  118.  
  119. ----------------------------------------------------------------------
  120.  
  121. Date: Fri Jul 13 18:16:23 1990
  122. >From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
  123. Subject: Updated MS-DOS 3.02 Test Version Available
  124. Keywords: MS-DOS Kermit 3.02
  125.  
  126. Additions to Joe Doupnik's MSR302.UPD file since the previous announcement
  127. in Info-Kermit V11 #8:
  128.  
  129. 22. Add more character set corrections from Matthias Reichling.
  130.  
  131. 23. Correct problem of placement of CR/LF in text screen dump (broken in
  132.     recent v3.02 changes), from Brian Holley.
  133.  
  134. Page 2                                                INFO-KERMIT DIGEST V12 #1
  135.  
  136.  
  137. 24. Make small change to Self Test code, DECTST, CSI 2 ; Ps y  (VT100) and
  138.     CSI 4 ; Ps...; Ps y  (VT320). Do a terminal reset only if a Ps is 0 or 1.
  139.     This is in the VT terminal emulator. Ps values greater than 1 are for
  140.     testing the terminal serial port, printer, and repeating the tests; they
  141.     are ignored here. All cases move the cursor back to the main display.
  142.  
  143. 25. Minor change to ensure modem status is correct even if the port has not
  144.     yet been used. File msxibm.asm.
  145.  
  146. 26. Modify the way characters are placed on Tektronix graphics screens from
  147.     starting them at intervals of 8 PC screen dots across the screen to
  148.     placing the lower left at the current drawing position (1 dot resolution).
  149.     Characters are 8x8 dots. The effect is better positioning accuracy along
  150.     the x-axis; the y-axis already behaved the better way. File msgibm.asm.
  151.  
  152. 27. Tiny tweak to Tek graphics code to overcome a deficiency in some Bios'
  153.     on storing video mode values if the high bit is set (means do not clear
  154.     the display buffer). Brian Holley found such a case and GRAPHICS.COM in
  155.     DOS 4 became confused.  The Bios on most machines clears that bit.  msgibm.
  156.  
  157. 28. Follow the lead of C-Kermit 5A and attempt simple correction of Parity
  158.     on the communications line during packet transfers.  Parity will be
  159.     shifted to EVEN, ODD, or MARK by the detector, in addition to the older
  160.     warning about removing the high bit on characters.  The first two packet
  161.     characters determine the parity, though the first four are examined for
  162.     unexpected setting of the high bit.  msscom, mssset, mssrcv, msssen.
  163.  
  164. 29. Adjust Tektronix emulator slightly so that SET TERM CLEAR-SCREEN does
  165.     a screen clear using the existing character coloring to write all spaces,
  166.     rather than doing a full reinitialization of the emulator. The text
  167.     emulator is not affected. Since ESC [ 30's; 40's m screen coloring command
  168.     sets the fore- and background color palette registers (7 and 0, resp) a
  169.     screen clear manually or by receipt of ESC [ 2 J will use this palette.
  170.     A terminal reset restores the built-in default palette.
  171.  
  172.     More changes are expected.
  173.  
  174. The files are in the Kermit test area, kermit/test/mstibm.boo and
  175. kermit/bin/mstibm.exe on watsun, and T:MSTIBM.BOO on CUVMA.
  176.  
  177. ------------------------------
  178.  
  179. Date: 15 June 1990
  180. >From: Tony Appelget, General Mills Inc, Minneapolis, MN, USA
  181. Subject: New Release of Kermit for HP-3000
  182. Keywords: HP-3000 Kermit
  183.  
  184. The HP3000 SPL Kermit seems to be badly neglected.  The listing in "Kermit
  185. News number 4" shows your most recent version to be v1.1 dated 85/06/24.  I am
  186. enclosing what I am assuming to be a much more modern version.  It is the
  187. product of a lot of hours put into the v1.0 version over the past five or six
  188. years and works very well here.
  189.  
  190. All my changes  are listed in comments at the beginning of the program.  A few
  191. of the big changes are:
  192.  
  193. INFO-KERMIT DIGEST V12 #1                                                Page 3
  194.  
  195.  
  196. - Added a HELP function.
  197. - Implementated a very versatile command name abbreviation scheme.
  198. - Implemented 3-byte checksums in addition to the old standard one-byte
  199.   checksums.
  200. - Implemented long packets with a maximum of 2000 bytes.  Decreasing packet
  201.   size in response to packet errors is not implemented.
  202. - This Kermit sets a JCW (program completion code).
  203. - Cleaned up a plethora of small bugs.
  204. - At the insistance of our security types, a file validation scheme was
  205.   built into the program.  No one uses it.  Leaving it in place will cause
  206.   no harm.  It should be removable without too much trouble.
  207.  
  208. The version number, 2.7, that I have stuck on the program is totally arbitrary
  209. as far as I am concerned.  If my enclosed Kermit meets your standards for
  210. distibution, it would be a feather in my cap for you to do so.
  211.  
  212. (signed)
  213. Tony Appelget
  214. K0DCF/AFA3AO/AFF3MN
  215.  
  216. [Ed. - Thanks, Tony, it does, apply feather to cap!  It looks like Kermit News
  217. #4 is shaking a lot of Kermit updates out of the woodwork.  This one replaces
  218. the old 1.1 version from 1985 in the Kermit "D" area as hp3000.*.]
  219.  
  220. ------------------------------
  221.  
  222. Date: Fri Jul 13 13:53:17 1990
  223. >From: Christine M. Gianone <cmg@watsun.cc.columbia.edu>
  224. Subject: Announcing NCR 9800 Kermit
  225. Keywords: NCR 9800 Kermit
  226.  
  227. This is to announce a Kermit program for the NCR 9800-4 computer, contributed
  228. by Paul E. Gladden of NCR Corporation, San Diego, CA, USA, for the NCR VE4.0
  229. operating system.  There are separate variations for the IVS and MCS
  230. environments.
  231.  
  232. Both products are written in the C language, based on C-Kermit 4E with
  233. features added or deleted depending on VRX system capabilities, and NCR
  234. selected product content.  Module names reflect the same usage as in C-Kermit
  235. 4E.  New modules have a short explanation.
  236.  
  237. IVS-Kermit and MCS-Kermit share common source code. MCS-Kermit has two
  238. additional files to interface to the MCS telecommunications facility.  MCS
  239. modules are labeled MCS ONLY.
  240.  
  241. A large number of files are included.  These have been installed in the "D"
  242. area of Kermit Distribution as ncrker.*.  The file ncrker.doc contains user
  243. documentation and installation instructions.  The file ncrker.jar is a "join"
  244. text archive of the numerous source files, which contains instructions at the
  245. top on how to break it apart.
  246.  
  247. Many thanks to Paul for submitting this new Kermit implementation.
  248.  
  249. ------------------------------
  250.  
  251.  
  252. Page 4                                                INFO-KERMIT DIGEST V12 #1
  253.  
  254. Date: 22 June 1990
  255. >From: Dave Squire, University of California at Davis
  256. Subject: New Release of Kermit for UNISYS (Burroughs) A-Series
  257. Keywords: UNISYS A-Series, Burroughs A-Series
  258.  
  259. Enclosed is the 1.040 release of our version of Kermit for Unisys A-Series
  260. systems.  This is an upgrade of "B78" Kermit 1.019 from 1986.  The important
  261. new features of this release are: long packets, alternate block-checks, and
  262. a "take" command.  The program has also been modified to use the A-Series
  263. hardware more efficiently.
  264.  
  265. This release contains the Kermit symbol (written in Algol) and a documentation
  266. file.  We will continue to distribute Kermit directly to A-Series sites who
  267. contact us.  We will also consult with new Kermit users to help them get
  268. started.  There is no fee for either service.  You may refer any Unisys
  269. A-Series users to us.  Also, please feel free to contact me if I can help with
  270. any A-Series questions.
  271.  
  272. Dave Squire
  273. Computing Services
  274. University of California
  275. Davis, CA  95616-8563  USA
  276. Phone 916-752-1927
  277.  
  278. [Ed. - Thanks to Dave and to Larry Johnson and Katie Stevens at UC Davis for
  279. their work on this program and for sending in this new version!  It's in the
  280. Kermit Distribution "D" area as uasker.*.  The previous release has been kept
  281. as b78ker.*, in case the new release does not run on the B7800 or on older
  282. Burroughs A-Series machines or operating system releases.  If it is not
  283. necessary to keep the old version, please let us know: does this new version
  284. run on the B7800?  Also, does anybody know the name of the operating system
  285. for these machines?]
  286.  
  287. ------------------------------
  288.  
  289. Date: Sun, 8 Jul 90 0:30:08 MET
  290. Subject: New Kermit Programs for Luxor ABC Series and FACIT DTC/DTC2
  291. >From: Bo Kullmar <bk@kullmar.se>
  292. Keywords: Luxor, ABC-80/800
  293.  
  294. I am sending you updates for the ABC80 and the ABC800 Kermit. Two members of
  295. the ABC-Klubben have done the work.
  296.  
  297. Joergen Westman has fixed the repeat prefix handling in the ABC80 Kermit.  It
  298. did not work against Kermit programs that could not do run-length encoding.
  299. (Jorgen with dots over the o.)
  300.  
  301. Mikael Johansson has updated the ABC800/ABC802/ABC806 Kermit to version 4.11
  302. and included support for FACIT DTC and FACIT DTC2. FACIT DTC/DTC2 is a ABC800
  303. clone.
  304.  
  305. --Bo Kullmar
  306.  
  307. [Ed. - Thanks, Bo!  The lux*.* files in the Kermit "C" area have been updated
  308. to the new versions.]
  309.  
  310.  
  311. INFO-KERMIT DIGEST V12 #1                                                Page 5
  312.  
  313. ------------------------------
  314.  
  315. Date: Tue, 1990 Jun 26   16:58 EDT
  316. >From: "John F. Chandler"   <PEPMNT@CFAAMP.HARVARD.EDU>
  317. Subject: New MS Kermit REMOTE PRINT interface with Kermit-370
  318. Keywords: MS-DOS Kermit, REMOTE PRINT Command, IBM 370 Kermit
  319.  
  320. I noticed from reading Info-Kermit V11 #8 that MS-DOS Kermit 3.02 has a REMOTE
  321. PRINT command.  For those not familiar with the guts of Kermit protocol, I'd
  322. like to remark that this command is essentially a slight modification of the
  323. MAIL command (you can see traces of that history on the screen when you issue
  324. the command), and support exists for the mainframe ends of both in release
  325. 4.2.0 of Kermit-370.  A slight modification of the KERMAIL command, renamed
  326. KERMPRT, will do the job (see IKxKER.DOC or IKxKER.PS for descriptions), as I
  327. have already demonstrated on my CMS system.  If there is sufficient interest,
  328. I could add samples of these mainframe commands to the Kermit-370 area, but
  329. the implementations will likely vary widely from site to site.
  330.  
  331. One of these days, I'll see about implementing a REMOTE PRINT subcommand in
  332. Kermit-370 so a mainframe user connected with a PC as a terminal can make hard
  333. copies on an attached desktop laser printer.  Right now, though, I'm
  334. concentrating on implementing Kermit support for the transparency mode in the
  335. new B2 microcode for the IBM 3174 (about which, more later).
  336.                                     John
  337.  
  338. ------------------------------
  339.  
  340. Date: Tue, 1990 Jun 26   17:34 EDT
  341. >From: "John F. Chandler"   <PEPMNT@CFAAMP.HARVARD.EDU>
  342. Subject: Kermit-370 support for IBM 3174 AEA
  343. Keywords: IBM 370 Kermit, IBM 3174 AEA
  344.  
  345. I now have word from IBM that the new release of 3174 microcode is out.  It
  346. is called B2 (that is, Level 2 of Configuration B) and is supposedly
  347. available to all customers with the "B" configuration of the 3174.  As I
  348. understand it, "B" costs extra and requires a bit more hardware (minimum
  349. requirement is model 11L or equivalent), while "A" comes with the box.
  350. Apparently, there are no plans for upgrading "A".  If you have "B"
  351. configuration installed, though, this upgrade should finally make it possible
  352. to run Kermit through the 3174.  I have some documentation about the new
  353. "ASCII Graphics Mode", and I think the necessary Kermit code is pretty
  354. simple.  Indeed, I have already made a stab at it (pending the answers to a
  355. few questions).  I'm looking for volunteers to try out the new feature and
  356. troubleshoot the code.  If you think you want to try it, the first thing to
  357. do is make sure that you (1) have the "B" configuration and (2) have received
  358. and installed the "B2" update from IBM.  I have no idea what the shipping
  359. schedule is like for things like that.  I can send you the Kermit update as
  360. soon as I check on a few details or even sooner if you're really eager.
  361. Meanwhile, let me know if you're interested.
  362.                                   John
  363.  
  364. ------------------------------
  365.  
  366. Date: Fri Jul 13 19:13:27 1990
  367. >From: Christine M. Gianone <cmg@watsun.cc.columbia.edu>
  368. Subject: VMS Kermit-32 Hints
  369.  
  370. Page 6                                                INFO-KERMIT DIGEST V12 #1
  371.  
  372. Keywords: VAX/VMS Kermit
  373.  
  374. Several people have complained that Kermit-32 has been sending files
  375. incorrectly when the file type is set to binary.  This can happen when the VMS
  376. file has a carriage control attribute; for example, VMS Lotus files are
  377. created this way.  The recent release of Kermit-32 has a new command SET FILE
  378. TYPE BLOCK which seems to get around this difficulty.
  379.  
  380. Also, judging from many messages received in the past month, many people who
  381. have complained that Kermit-32 did not pick up their terminal name correctly
  382. ("No default terminal line for transfers") have not obtained version 3.3.126,
  383. announced in Info-Kermit V11 #8, which fixes that problem, as well as
  384. problems related to spawning and batch operation.
  385.  
  386. ------------------------------
  387.  
  388. Date: 22 Jun 90 13:33:16 GMT
  389. >From: RE_Sonneveld@pttrnl.nl (R.E. Sonneveld, +31 70 332 5362, PTT Research)
  390. Subject: How to Create a Kermit Connection over X.25 from VAX/VMS?
  391. Keywords: VAX/VMS Kermit, X.25 Network
  392.  
  393. Wanted: Kermit communication over X.25 (with for example EuroKom) with my
  394. 'own' VMS system as my home base.  In other words: is it possible to start
  395. Kermit on the VAX, then SET HOST/X29 to e.g. EuroKom, then start Kermit at the
  396. remote side and establish the connection?
  397.  
  398. We have tried the following:
  399.  
  400.  1) loopback one DECserver port-X to port-Y
  401.  2) define the DECserver ports to be 'transparent'
  402.  3) define DECserver port-Y to autoconnect
  403.  4) with LCP, create LTA port pointing to the DECserver port-X
  404.  5) define default port for Kermit (KER$COMM), pointing to the LTA device
  405.  
  406. Starting Kermit and using the default port will result in a connect via the
  407. DECserver loopback to the original VMS-system, thus creating the possibility
  408. of establishing a X.25 connection.  This works great! The major problem and
  409. the reason why we won't release this 'facility' is SECURITY: whenever a
  410. connection is established by one person and this connection aborts
  411. unexpectedly, than the next user who tries to make a connection using this
  412. default port gets the session of the first user!
  413.  
  414. Any help greatly appreciated!
  415.  
  416. Rolf Sonneveld        E-mail: RE_Sonneveld@pttrnl.nl
  417.  
  418. PTT Research          *      Tele phone : +31 70 332 5362
  419. Neher Laboratories,   *      Tele  fax  : +31 70 332 6477
  420. The Netherlands       *      Tele   x   : 31236 prnl nl
  421.  
  422. [Ed. - This is a common question from Europe.  As yet, neither C-Kermit/VMS
  423. nor VMS Kermit-32 contain any explicit support for VAX/PSI or X.25 in any of
  424. its forms.  We are hoping that someone with an X.25-connected VAX/VMS system
  425. will add this support.  Volunteers?]
  426.  
  427. ------------------------------
  428.  
  429. INFO-KERMIT DIGEST V12 #1                                                Page 7
  430.  
  431.  
  432. Date: Fri, 29 Jun 90 05:27:31 EDT
  433. >From: Erez Zadok <ezk@cunixf.cc.columbia.edu>
  434. Subject: Kermit for Microsoft Windows
  435. Keywords: Microsoft Windows Kermit
  436.  
  437. I'm running now WIN100 V2.0 and it's a definite advantage over older
  438. versions.  I have a few comments and questions.
  439.  
  440. COMMENTS:
  441.  
  442. (1) Good job!  I like it in the GUI environment.  I actually downloaded
  443. files while playing Solitaire.  Multitasking on my desk.  I liked the fact
  444. that there was a running number of packets on the minimized kermit icon so I
  445. can always keep a watch over the file transfer.  I liked the beep at the end
  446. of the transfer and the Alert that popped up for me at the end.
  447.  
  448. (2) This is definitely the wave of the future.  I think windows kermit
  449. *will* be a major communication package to use on ms-dos platforms and will
  450. bite a large chunk of the regular ms-kermit.  But there's a lot to be done.
  451. With Windows 3.0 out now, you have all the reasons in the world to make this
  452. package at the same level of MS-Kermit 3.02.
  453.  
  454. (3) The first things that struck me as in need of improvement were (and bear
  455. in mind that I know MS-Kermit 3.02 *very* well):
  456.  
  457. - Not enough variety of fonts to choose.  My VGA does not fit 80
  458.   columns with this font (courier) and the other font (preview) is
  459.   way too small.  Maybe the fonts are windows dependent and not
  460.   depend on kermit?
  461. - No apparent way to map the keyboard,  Esp. swapping ` and ESC, and
  462.   mapping CAPSLOCK to CTRL.
  463. - No dialup scripts to use.
  464. - Not enough colors to choose from (esp. the high intensity ones.)
  465. - Win100 does not use the new builtin help feature in Windows 3.0.
  466.  
  467. (4) Then, upon closer look, I found that the next few items should be
  468. improved (lower priority in my opinion):
  469.  
  470. - Support for long packets.
  471. - When changing a font or its size, I would like not to have to
  472.   restart kermit.
  473. - I would like to see a way to set the number of lines as an input
  474.   from the keyboard and not only by resizing the window.
  475. - I would like to cut from the kermit screen by using the mouse only
  476.   and marking and cutting.  The current way is not as clean.
  477. - Screen refreshes are not as fast (maybe that's due to my
  478.   hardware configuration and lack of memory.)
  479.  
  480. (5) The vt100 emulator works well with all kind of applications including
  481. EMACS and curses-based applications.
  482.  
  483. (6) Possible serious bug:  At one point I got an alert "out of token memory"
  484. error or something, and kermit aborted.  My line was not disconnected.  I am
  485. running Windows 3.0 in real mode (no extended or expanded memory yet on my
  486. 286) and it seems that kermit ran out of memory of some sort.  In either
  487.  
  488. Page 8                                                INFO-KERMIT DIGEST V12 #1
  489.  
  490. case it should not have aborted but only gave an alert.  Maybe this error is
  491. unrecoverable...
  492.  
  493. [Ed. - This is one of several messages about Bill Hall's new VT100-emulating
  494. Kermit for Microsoft Windows.  All of them contained similar questions and
  495. favorable comments.  Bill replies as follows:
  496.  
  497. Most of the suggestions are on my worklist to do - someday!.  Life if pretty
  498. busy since I work a regular job and also write the Windows column for PC
  499. Magazine.  But, I do want to improve the version I have put up as a beta.
  500.  
  501. Please be patient.  There is no help because it is a 2.0 application.  The
  502. font is terrible, but don't blame me for this.  I would have to design a
  503. complete series of fonts for a range of displays if I could not use the ones
  504. supplied by Windows.  Maybe some kind soul will eventually help me with this
  505. really non-trivial effort.
  506.  
  507. The screen update is slow.  As you know, everything in Windows is done in
  508. software so this is a problem.  However, I'm sure I can do something to
  509. improve it a bit, and I'll give it a try.
  510.  
  511. The implementation of the Kermit part is confined at the present to the
  512. standard stuff.  It may be archaic, but it takes a lot of time to get all the
  513. new features working.  Also, making Kermit work smoothly in Windows is quite a
  514. chore.  It is so hard that I'm doing an article on it using the level I have
  515. developed up to now as the example.  When the article is done, I will then
  516. take up the task of upgrading the program.]
  517.  
  518. [Ed. again - By the way, Bill has sent in a slightly updated version of the
  519. program, and some preliminary documentation: win100.msp (to be viewed in
  520. Microsoft Paint) and win100.wri (for Microsoft Write).  They document just the
  521. emulator part of the program, not the file transfer part.]
  522.  
  523. ------------------------------
  524.  
  525. Date: Tue, 26 Jun 90 23:43:20 CST
  526. >From: BOB SHERMAN <SHERMAN%ETSUADMN@cuvmb.cc.columbia.edu>
  527. Subject: External Kermit Driver
  528. Keywords: MS-DOS Kermit 3.0
  529.  
  530. Greetings:
  531.   I am running a BBS on an IBM PC.  Do you have any suggestions
  532.   for a external kermit driver?  Thanks for any help.
  533.  
  534.                                          Bob Sherman
  535.  
  536. ------------------------------
  537.  
  538. Date: Mon, 25 Jun 90 11:33:39 CDT
  539. >From: phil@wubios.wustl.edu (J. Philip Miller)
  540. Subject: Kermit for Atari Portfolio?
  541. Keywords: Atari Portfolio Kermit
  542.  
  543. In a review of the Atari Portfolio - a pocket size, almost DOS PC - there was
  544. mention of a modified version of Kermit that would work on the Portfolio.
  545. Could some point me to where it is available?
  546.  
  547. INFO-KERMIT DIGEST V12 #1                                                Page 9
  548.  
  549.  
  550. This machine runs the DIP Operating System, which is very close to DOS 2.11.
  551. I understand it is a small patch to the regular PC version.
  552.  
  553. -phil
  554.  
  555. J. Philip Miller, Professor, Division of Biostatistics, Box 8067
  556. Washington University Medical School, St. Louis MO 63110
  557. phil@wubios.WUstl.edu - Internet  (314) 362-3617
  558. uunet!wuarchive!wubios!phil - UUCP (314)362-2693(FAX)  C90562JM@WUVMD - bitnet
  559.  
  560. [Ed. - Yes, and please point us at it too!]
  561.  
  562. ------------------------------
  563.  
  564. End of Info-Kermit Digest
  565.  
  566. Page 10                                               INFO-KERMIT DIGEST V12 #2
  567.  
  568. Info-Kermit Digest         Fri, 3 Aug 1990        Volume 12 : Number 2
  569.  
  570. Departments:
  571.  
  572.   ANNOUNCEMENTS -
  573.         Another Update of the MS-DOS Kermit 3.02 Test Release
  574.         Version 8.12 of PRIME Kermit Now Available
  575.         New Documentation Available for VAX/VMS Kermit-32 3.3.126
  576.         New Test Release of DEC Rainbow Kermit
  577.  
  578.   MS-DOS KERMIT -
  579.         MS-DOS Kermit 3.0x and Microsoft Windows 3.0
  580.         MS-DOS Kermit 3.0 Hints
  581.         MS-Kermit Server and Hayes 2400bps Modems?
  582.  
  583.   KERMIT-370 -
  584.        IBM Mainframe Kermit and Trailing Blanks
  585.        IBM Kermit 4.2 and tn3270
  586.  
  587. Digest submissions may be sent to Info-Kermit@WATSUN.CC.COLUMBIA.EDU,
  588. requests for addition to or deletion from the Info-Kermit subscriber list to
  589. Info-Kermit-Request@WATSUN.CC.COLUMBIA.EDU or to KERMIT@CUVMA.BITNET.
  590.  
  591. Kermit files may be obtained over networks and by mail order.  On the
  592. Internetwork, use FTP to log in to host WATSUN.CC.COLUMBIA.EDU, a SUN-4/280
  593. running UNIX (SUNOS 4.1), IP host number 128.59.39.2.  Login as user anonymous
  594. (note, lower case), any password, and GET or MGET (MULTIPLE GET) the desired
  595. files.  The Kermit files are in directories kermit/a, kermit/b, kermit/c,
  596. kermit/d, and kermit/e.  Test versions are in kermit/test.  Binaries are in
  597. kermit/bin (use ftp in binary mode).  You can also get Kermit files over the
  598. BITNET/EARN network; to get started send a message with text HELP to KERMSRV,
  599. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  600. file kermit/a/aanetw.hlp (AANETW.HLP on KERMSRV).  To order by mail, request a
  601. complete list of Kermit versions and an order form from Kermit Distribution,
  602. Columbia University Center for Computing Activities, 612 West 115th Street,
  603. New York, NY 10025 USA.
  604.  
  605. ----------------------------------------------------------------------
  606.  
  607. Date: Wed Aug  1 19:56:09 1990
  608. >From: Christine M. Gianone <cmg@watsun.cc.columbia.edu>
  609. Subject: Another Update of the MS-DOS Kermit 3.02 Test Release
  610. Keywords: MS-DOS Kermit 3.02
  611.  
  612. >From Joe Doupnik.  Changes since the version announced in Info-Kermit V12 #1
  613. are listed below.  Items 38 and 39 will be of interest to Honeywell users for
  614. supporting connections to the Honeywell DPS-6 (see article by Frank Dreano in
  615. Kermit News #4) -- they remove the need for a custom version of MS-DOS Kermit
  616. to communicate with the the Honeywell systems.  Please report any problems to
  617. Joe at JRD@USU.BITNET or JRD@CC.USU.EDU.
  618.  
  619. 30. Add correction to ensure NUL and DEL are passed through when printing
  620.     transparently. msxibm
  621.  
  622. 31. Revise REPLAY filespec a little. Previously at the end of the file any
  623.     key would exit Connect mode. Now all keys remain active so that screen
  624.  
  625. INFO-KERMIT DIGEST V12 #2                                               Page 11
  626.  
  627.     dumping and screen rollback function; ALT-X and Control-C (emergency
  628.     bailout) exit Connect mode. msxibm
  629.  
  630. 32. Internal revisions to details of handling 75/1200 baud split speeds.
  631.     Previously, filling the receiver buffer would cause an XOFF to be sent
  632.     at 1200 baud rather than at the correct 75 baud. Now the XOFF character
  633.     is sent at 75. A pseudo-queue is present to prevent conflicts of using
  634.     the timer chip for beep or regular 75 baud output when the receiver
  635.     interrupt occurs (and resulted in the XOFF being sent at interrupt level
  636.     rather than at task level). The queue makes the XOFF be sent at task level
  637.     and at the proper 75 baud rate. Dan Norstedt had the idea, this queueing
  638.     stuff is my solution. 75 baud output is done as a software UART. msxibm
  639.  
  640. 33. Details here and there to save code segment space, from Dan Norstedt.
  641.  
  642. 34. Add terminal type of VT100. This performs identically to a VT102 except
  643.     the response to the type inquiry from a host is ESC [ ? 1 c (a VT100).
  644.  
  645. 35. Found and fixed a protocol problem present in MSK 3.0 to date. If an I
  646.     packet is sent to a server Kermit which does not understand I packets
  647.     the resulting Error packet from the server is to be taken as a forgivable
  648.     error, rather than otherwise being a fatal one. The version 3.0 and 3.01
  649.     code forgot to forgive. I packets are sent by GET and REMOTE commands.
  650.     Fixed on 19 July. mssser.
  651.  
  652. 36. Remove VTxxx response to CSI [ .. R. That should not have been present.
  653.     Thanks to Ted Medin for discovering its presence.
  654.  
  655. 37. Add 80/132 column support for Paradise Autoswitch EGA Mono board, thanks
  656.     to David Zielke of Duke Univ. msyibm
  657.  
  658. 38. Add two new commands
  659.  
  660.     SET SEND DOUBLE-CHAR <number between 0 and 255>
  661.         which sends this character twice in a row in outgoing packets.
  662.         The doubling occurs as packet bytes are being sent to the
  663.         communications channel and hence does not occur during packet
  664.         construction. This accomodates communications devices which
  665.         need doubling to let through a single copy. The doubling
  666.         will not appear in the packet log. Padding characters are
  667.         not affected by this command since padding is to be discarded
  668.         at the other end anyway.
  669.  
  670.     SET RECEIVE IGNORE-CHAR <number between 0 and 255>
  671.         which ignores this character in received packets. This
  672.         accomodates communications devices which may insert a
  673.         character into a stream, such as a Line Feed inserted before a
  674.         Carriage Return. The ignored character will not appear in the
  675.         packet log. Note that NUL (0) and DEL (127) are normally
  676.         ignored even without this command.
  677.  
  678. 39. Add terminal type of HONEYWELL to the emulator. This is a VT100 with
  679.     two special responses. A received ENQ character, decimal 5, causes the
  680.     emulator to send the 14 characters
  681.  
  682.         '7813  P GC  A' 03h
  683.  
  684. Page 12                                               INFO-KERMIT DIGEST V12 #2
  685.  
  686.  
  687.     and receipt of  ESC y  causes the emulator to send the 27 characters
  688.  
  689.        ESC '[8p  OT' 03h ESC '[y7813  P GC  A ' 03h
  690.  
  691.     where the quote marks (') above are not sent but are present so we can
  692.     see the spaces which are sent; other spaces are for reading convenience.
  693.     03h is Control-C, a binary value 3. The terminal ident response is VT100.
  694.     These Honeywell VIP7809 emulation features are from Frank Dreano,
  695.     dreano@trout.nosc.mil.
  696.  
  697. 40. Modify the internals for SET KEY LK250 to result in checking for an
  698.     active external LK250 driver and if found activate the LK250 usage in
  699.     Kermit. If the driver is not found the message "?LK250 keyboard external
  700.     driver is not active." is displayed. Note that to obtain the default
  701.     LK250 key mappings the user must say SET KEY CLEAR after the above
  702.     command, and then add any local definitions. Tnx to Terry Kennedy. msuibm
  703.  
  704. 41. In command WAIT <timeout> \CD \CTS \DSR   let the backslash character
  705.     be optional, to merge syntax with C Kermit 5A. mssscp
  706.  
  707. 42. Add further Honeywell terminal type modifications from Frank Dreano:
  708.     - change invocation ESC y to be ESC [ y     (see item 39 above)
  709.       This will cause loops if the host echoes the response.
  710.     - add  ESC `  (accent grave) clears screen and homes the cursor.
  711.  
  712. 43. Revise slightly the query for the active Code Page, and the table of
  713.     character sets for use by SET FILE CHARACTER-SET, to be system dependent
  714.     so that MS-DOS in Japan can be used more effectively. That MS-DOS lacks
  715.     the concept of Code Pages. Thanks to Hirofumi Fujii for the suggestions.
  716.  
  717. 44. Add another host invocation control sequence to the text terminal emulator.
  718.     Reception of  CSI Pn;...Pn ~  invokes a macro named PRODUCT, if it
  719.     exits, and sets variables \%1, \%2, and so forth up to \%9 to be the ascii
  720.     version of the numeric parameters Pn above. Omitted parameters are taken
  721.     to be zeros. Only as many variables are set as numeric parameters, and
  722.     script item ARGC is one larger than this quantity. If macro PRODUCT does
  723.     not exist only the \%x items are changed. Connect mode is not reentered
  724.     automatically if the macro exists. This is very similar to TERMINALR and
  725.     TERMINALS, except for the variable setting part.
  726.     The invocation sequence may change so don't write it into applications
  727.     until Kermit is released again. It is not invocable from the keyboard,
  728.     because of the needed parameters. mssset.asm, msyibm.asm, mszibm.asm.
  729.  
  730. 45. Correct a keyboard translation problem. Control codes could be translated
  731.     to printable characters for a few National Replacment Character sets,
  732.     because that was where the special graphical symbols are located on the
  733.     IBM PC display adapter. This has been corrected to never translate control
  734.     codes.
  735.  
  736. 46. Slightly revise script INPUT and REINPUT command sensing of timeouts to
  737.     do the time reading after each serial port character reading operation.
  738.     This might cure complaints of REINPUT 0 reporting premature FAIL status.
  739.  
  740. 47. If disk reading fails while sending a file tell the receiver to dispose
  741.     of the file with a Delete status, rather than the previous status of
  742.  
  743. INFO-KERMIT DIGEST V12 #2                                               Page 13
  744.  
  745.     just end-of-file. There is no other simple way of informing the receiver
  746.     of such a local problem. Other files in a group will be processed normally.
  747.  
  748.   More changes are expected.
  749.  
  750. ------------------------------
  751.  
  752. Date: Fri Aug  3 10:46:17 1990
  753. >From: Christine M. Gianone <cmg@watsun.cc.columbia.edu>
  754. Subject: Version 8.12 of PRIME Kermit Now Available
  755. Keywords: PRIME Kermit
  756.  
  757. >From Matthew Sutter of Lincoln National Corporation in Fort Wayne, Indiana,
  758. USA, comes version 8.12 of PRIME Kermit, an upgrade of version 8.00 that was
  759. prepared by John Horne of Polytechnic South West in Plymouth, England.
  760.  
  761. The major feature of version 8.12 is support for local-mode (dialout)
  762. operation using the AMLC lines.  CONNECT, GET, FINISH, and BYE commands have
  763. been added, as well as the script commands INPUT, OUTPUT, CLEAR, and PAUSE.
  764.  
  765. This version also contains fixes to repeat character processing and sliding
  766. windows, but the sliding window feature still needs more work in order to
  767. interoperate with MS-DOS Kermit 3.0 or C-Kermit 5A.  This work is underway,
  768. watch Info-Kermit for further announcements.
  769.  
  770. The new files are in kermit/d/prime8.* on watsun, and PRIME8.* on CUVMA.
  771. There is, as yet, no new documentation.  Thanks to Matt for his work and for
  772. contributing it to the Kermit collection!
  773.  
  774. ------------------------------
  775.  
  776. Date: Wed Aug 1 18:29:49 1990-EDT
  777. >From: Christine M. Gianone <cmg@watsun.cc.columbia.edu>
  778. Subject: New Documentation Available for VAX/VMS Kermit-32 3.3.126
  779. Keywords: VAX/VMS Kermit
  780.  
  781. The VAX/VMS Kermit-32 chapter of the Kermit User Guide has been updated to
  782. include all the information about the new release, and the help text has also
  783. been updated.  Thanks to Jonathan Welch of the University of Massachusetts for
  784. his help in clarifying the actions and interaction of the SET FILE BLOCKSIZE
  785. and SET FILE TYPE commands, and for proofreading the manual.
  786.  
  787. The files for the new manual are VMSHDR.MSS and VMSMIT.MSS (Scribe source),
  788. VMSMIT.DOC (plain ASCII text) and VMSMIT.PS (Postscript).  The help text is in
  789. VMSMIT.RNH (Runoff source).  Tape B (kermit/b, etc etc).
  790.  
  791. ------------------------------
  792.  
  793. Date: Thu Aug 2 13:01:34 1990-EDT
  794. >From: Christine M. Gianone <cmg@watsun.cc.columbia.edu>
  795. Subject: New Test Release of DEC Rainbow Kermit
  796. Keywords: MS-DOS Kermit, Rainbow Kermit, DEC Rainbow
  797.  
  798. >From Robert Weiner, a new version of Kermit for the DEC Rainbow, based on
  799. the system-independent parts of the July 22nd MS-DOS Kermit 3.02 test release.
  800. The Rainbow version lacks many of the IBM-specific enhancements that appeared
  801.  
  802. Page 14                                               INFO-KERMIT DIGEST V12 #2
  803.  
  804. in MS-DOS Kermit 3.0, like international character sets, instantaneous screen
  805. rollback, etc, but these MAY be added before the final release of this
  806. program.  The system-independent features -- script programming enhancements,
  807. sliding windows, long packets, VT200/300 emulation (it's still at the VT102
  808. level), etc.
  809.  
  810. Thanks to Rob for keeping the Rainbow up to date!  The files are in
  811. kermit/a/mstrb1.boo ("BOO" file to be un-boo'd into an .EXE file using any of
  812. the MSBPCT files) on the Internet or MSTRB1 BOO on BITNET KERMSRV at CUVMA.
  813. For internet binary-mode FTP access only, also kermit/bin/mstrb1.exe.
  814.  
  815. ------------------------------
  816.  
  817. Date: Thu, 2 Aug 90 11:18:30 PDT
  818. >From: william@xlnvax.novell.com (William Hall)
  819. Subject: MS-DOS Kermit 3.0x and Microsoft Windows 3.0
  820.  
  821. Just tried MS-DOS Kermit 3.0 in enhanced mode Windows.  I began by running a
  822. DOS shell then running Kermit.  Did not bother with a PIF file.  I changed the
  823. settings to run from exclusive to background.  You can get to these setting
  824. with ALT-SPACE, then moved it to a Window.  For fun, I ran vttest on UNIX and
  825. checked the character set, especially the one for DEC special graphics.  They
  826. looked as expected.
  827.  
  828. Then I tried a long file transfer.  I made the window iconic, and the file
  829. transfer ran in the background to completion with no errors.  At the end, the
  830. bell beeped to let me know it was done.
  831.  
  832. So it works and quite well to the extent that I tested it.
  833.  
  834. --Bill
  835.  
  836. ------------------------------
  837.  
  838. Date: 1-AUG-1990 20:33:04.67
  839. >From: Joe Doupnik <jrd@cc.usu.edu>
  840. Subject: MS-DOS Kermit 3.0 Hints
  841. Keywords: MS-DOS Kermit 3.0 Troubleshooting
  842.  
  843. [Ed. - This is extracted from a reply by Joe to a user who reported a
  844. nonfunctioning Kermit on a Compaq computer.]
  845.  
  846. One common item is an internal modem which fails to follow the normal rules of
  847. how to behave like a regular serial port.  Another is two devices are having a
  848. small battle for an IRQ line or a memory (port) address.
  849.  
  850. MS-DOS Kermit 3.02 (not 3.2) is strictly development software, as indicated in
  851. the note file MSR302.UPD and on the startup banner.  The REMOTE command
  852. problems you report fall into the category of fixed, broken, and then fixed
  853. again (sigh).  In all of the broken cases Kermit would retry many times and
  854. that takes up to a minute or so to finish.  While that is happening one side
  855. can send an XOFF to the other and forget to send a matching XON later; Kermit
  856. waits 10 seconds to figure out this situation when it needs to send and breaks
  857. through regardless.  So, this "flow control deadlock" can add to the waiting
  858. interval, but it's better than being told to never speak ever again.
  859.  
  860.  
  861. INFO-KERMIT DIGEST V12 #2                                               Page 15
  862.  
  863. Another situation, which happened to me a couple of years ago, was the poor
  864. hard disk controller had a conflict with another board and every once in a
  865. while the machine would hang waiting on the hard disk for 10 minutes or more.
  866. Usually I had to push the big Red switch. When Kermit has to go to the hard
  867. disk then these kinds of things can happen.
  868.  
  869. Generally, I try to sort out the problem by trying Connect mode between
  870. machines right away. If that works then I know the communications line part is
  871. basically ok.  If packets fail then I say SET DEBUG ON and SET DISPLAY SERIAL
  872. to watch what's happening during a file transfer or REMOTE command (REMOTE WHO
  873. is my favorite "protocol" quicky, no disk access).
  874.  
  875. Let's suppose the Compaq is in sad shape.  My suggestion is to backup the hard
  876. disk contents and ensure they are correct (say to DOS VERIFY ON to help).
  877. Then do a low level format of the drive (yow!) and a DOS FORMAT with the
  878. correct version of DOS.  Put things back on the hard disk and keep your
  879. fingers crossed.  Double check the settings of the boards too.  Unload the
  880. likely mountain of favorite TSR's and see if things work.  If they do then
  881. start adding them back one at a time and retesting.  Make sure that CONFIG.SYS
  882. has BUFFERS=20 or so and FILES=20 or so.  Watch out for Desqview because it
  883. has troubles remembering that a communications program needs to service
  884. interrupts from the serial port.  Don't mix DOS files from different
  885. releases/manufacturers.
  886.  
  887. Ensure that parity is the same on both machines, NONE being better than any
  888. other flavor (Kermit does it's own checking in a much more secure way).  Don't
  889. use MODE COM1:....,P with that ",P" (for Printer waiting) on the same port as
  890. the communications channel because DOS will grab the port and won't release
  891. it.  Watch for mice drivers nibbling on the same serial port.  In short, take
  892. the machine down to bare metal and rebuild it slowly from a consistent set of
  893. original DOS floppies (put write-protect tabs on them).
  894.  
  895. Did I forget to mention virii?
  896.  
  897. ------------------------------
  898.  
  899. Date: Tue, 02 May 89 16:54:34 CDT
  900. >From: Barry Pederson  <UD092096@VM1.NoDak.EDU>
  901. Subject: MS-Kermit Server and Hayes 2400bps Modems?
  902. Keywords: MS-DOS Kermit Server Mode
  903.  
  904. Hi, we've been using MS-Kermit for a while now, and it's been incredibly
  905. useful, but we're having a slight problem.  We have some IBM model 30-286
  906. machines connected to Hayes 2400bps external modems.  They work fine except
  907. when we try to have the modems answer the phone while Kermit is in server
  908. mode.
  909.  
  910. We have a script set up to set the speed to 2400, parity to none, then we
  911. output ATS0=1\13 to the modem, and go into server mode.  Sometimes it works,
  912. but a lot of the time, when the phone rings the modem will drop down to
  913. 1200bps.
  914.  
  915. In a previous Kermit Digest, I thought I saw something about C-Kermit having
  916. problems with Hayes modems, does MS-Kermit have a known incompatibility
  917. between it's server mode and a Hayes modem?
  918.  
  919.  
  920. Page 16                                               INFO-KERMIT DIGEST V12 #2
  921.  
  922. I suspect that Kermit is working fine, but the Hayes is doing something screwy
  923. (we've tried MS-Kermit 2.31 and 2.32/a).  I'd really appreciate any
  924. suggestions on how to keep the Hayes from slowing down.
  925.  
  926. Thanks.
  927.  
  928. Barry Pederson    University of North Dakota School of Medicine
  929.      ud092096@ndsuvm1.bitnet     or    ud092096@vm1.nodak.edu
  930.  
  931. [Ed. - When MS-Kermit is in server mode, it times out every 30 seconds or so
  932. and transmits a negative acknowledgement packet (the familiar "^A# N3").  It
  933. is likely that the NAK packet is confusing the modem.  The workaround is to
  934. give MS-DOS Kermit the SET TIMER OFF command before putting it in server mode,
  935. which is safe as long as the other Kermit will do timeouts (some, like IBM
  936. mainframe CMS Kermit, cannot).  In MS-DOS Kermit 2.32/A or later, you can
  937. disable server command wait timeouts without turning the timer off, by using
  938. the command SET SERVER TIMEOUT 0.]
  939.  
  940. ------------------------------
  941.  
  942. Date: 6-Jul-90 22:03:17-GMT
  943. >From: Randy Schirmer@cuvmb.cc.columbia.edu
  944. Subject: IBM Mainframe Kermit and Trailing Blanks
  945. Keywords: IBM 370 Kermit, VM/CMS Kermit, Blanks
  946.  
  947. We're running Kermit CMS and MS Kermit to transfer files with a record length
  948. of 1500 bytes. What's happening is that the records have trailing blanks.
  949. Kermit strips these blanks off. We need them to remain intact for input into a
  950. commercial package we are running under DOS. Is there any way to keep the
  951. blanks from being stripped? I know that you can specify LRECL and RECFM at the
  952. CMS level but that is for files being received in CMS.  We need the same
  953. capability for MS-DOS Kermit.
  954.  
  955. Randy Schirmer
  956. West Chester University, West Chester, PA
  957. 215-436-1094, RSCHIRME@WCU
  958.  
  959. [From John Chandler - When the file type is TEXT and the recfm is F,
  960. Kermit-370 strips off trailing blanks from each record before sending a file.
  961. In order to keep trailing blanks in a downloaded file, you can convert it to
  962. RECFM V in CMS before sending it, e.g., by using COPYFILE with the RECFM V
  963. option.  This is true of versions from 4.1 onward.]
  964.  
  965. ------------------------------
  966.  
  967. Date: Wed, 20 Jun 90 08:33:25 -0500
  968. >From: C Lance Moxley <clmoxley@ux1.cso.uiuc.edu>
  969. Subject: IBM Kermit 4.2 and tn3270
  970. Keywords: IBM 370 Kermit, tn3270
  971.  
  972. I just assembled Kermit/370 on an IBM 3081 running VM/SP 5.  The program
  973. assembled just fine.  But, I have no idea how to set things to work through
  974. tn3270. I've used MS-DOS Kermit for quite a while now and also am very
  975. comfortable with C-Kermit, but this seems to be alot different. I would
  976. appreciate any help you can give on this.  I also have the ability to use 3708
  977. protocol converters, but I don't think there is any way I can put them in
  978.  
  979. INFO-KERMIT DIGEST V12 #2                                               Page 17
  980.  
  981. line-by-line mode. I would really prefer to use tn3270. If you are the wrong
  982. person to ask this to, please direct me to the right person. Also if there are
  983. any examples anywhere, let me know. I couldn't find anything on watsun.
  984.  
  985. Thanks for a very fine product. The other versions that I use are the best of
  986. their kind. Especially the new C version.
  987.  
  988.                                 Thanks,
  989.                                 Lance
  990.  
  991. [From John Chandler - Actually, no, I haven't used tn3270 myself, but Bob
  992. Babcock has -- and from a Unix system only.  It should have been essentially
  993. the same as your attempt, but it worked.  The special considerations are only
  994. two: you must SET CONTROLLER SERIES1 by hand, and you cannot (apparently)
  995. upload using long packets.  The system in question was a CONVEX, as opposed to
  996. a SUN, but I think the tn3270 would have been the same.]
  997.  
  998. ------------------------------
  999.  
  1000. End of Info-Kermit Digest
  1001.  
  1002. Page 18                                               INFO-KERMIT DIGEST V12 #3
  1003.  
  1004. Info-Kermit Digest         Tue, 28 Aug 1990        Volume 12 : Number 3
  1005.  
  1006. Today's Topics:
  1007.  
  1008.          Duplicate Copies of Info-Kermit Digest V12 #1 and #2
  1009.          Proposed Kermit Extension for SET FILE TYPE
  1010.           And ANOTHER New Test Release of MS-DOS Kermit 3.02
  1011.         A New Version of Kermit for OS/2 Presentation Manager
  1012.                New Release UCPECAN Kermit V1.1
  1013.              MS-Kermit 3.0x and German telephones
  1014.                  Looking for Stories
  1015.  
  1016.  
  1017. Digest submissions may be sent to Info-Kermit@WATSUN.CC.COLUMBIA.EDU,
  1018. requests for addition to or deletion from the Info-Kermit subscriber list to
  1019. Info-Kermit-Request@WATSUN.CC.COLUMBIA.EDU or to KERMIT@CUVMA.BITNET.
  1020.  
  1021. Kermit files may be obtained over networks and by mail order.  On the
  1022. Internetwork, use FTP to log in to host WATSUN.CC.COLUMBIA.EDU, a SUN-4/280
  1023. running UNIX (SUNOS 4.1), IP host number 128.59.39.2.  Login as user anonymous
  1024. (note, lower case), any password, and GET or MGET (MULTIPLE GET) the desired
  1025. files.  The Kermit files are in directories kermit/a, kermit/b, kermit/c,
  1026. kermit/d, and kermit/e.  Test versions are in kermit/test.  Binaries are in
  1027. kermit/bin (use ftp in binary mode).  You can also get Kermit files over the
  1028. BITNET/EARN network; to get started send a message with text HELP to KERMSRV,
  1029. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  1030. file kermit/a/aanetw.hlp (AANETW.HLP on KERMSRV).  To order by mail, request a
  1031. complete list of Kermit versions and an order form from Kermit Distribution,
  1032. Columbia University Center for Computing Activities, 612 West 115th Street,
  1033. New York, NY 10025 USA.
  1034.  
  1035. ----------------------------------------------------------------------
  1036.  
  1037. Date: Tue Aug 28 17:44:54 1990-EDT
  1038. >From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
  1039. Subject: Duplicate Copies of Info-Kermit Digest V12 #1 and #2
  1040.  
  1041. Please disregard any duplicate copies of recent Info-Kermit
  1042. Digests.
  1043.  
  1044. ------------------------------
  1045.  
  1046. Date: Thu Aug 16 17:44:54 1990-EDT
  1047. >From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
  1048. Subject: Proposed Kermit Extension for SET FILE TYPE
  1049. Keywords: Kermit Protocol, Labeled File Type
  1050.  
  1051. Kermit programs presently differentiate between text and binary files via
  1052. the SET FILE TYPE command.  When the file type is TEXT, the Kermit program
  1053. that is sending the file converts the file's record format and character set
  1054. from its local form to a standard form, and the receiving Kermit program
  1055. converts from the standard form into its own local record format and character
  1056. set.  When the file type is BINARY, the file is sent as-is, byte for byte.
  1057.  
  1058. These two ways of handling files are sufficient in many cases, but fall short
  1059. when one or both computers has a complex file system, such as the Macintosh or
  1060.  
  1061. INFO-KERMIT DIGEST V12 #3                                               Page 19
  1062.  
  1063. DEC operating systems like VAX/VMS.  The Kermit protocol includes a mechanism
  1064. for transmitting a file's attributes in a generic way, but these attributes
  1065. are of little use when the file is being received by a computer with a simple
  1066. file system, such as UNIX or MS-DOS, that can't do anything with them.  And
  1067. they are not sufficient to convey every conceivable bit of information about
  1068. about every kind of file on every kind of computer.
  1069.  
  1070. To allow a complicated file to be transferred, a new file type is proposed:
  1071.  
  1072.   SET FILE TYPE LABELED
  1073.  
  1074. Notice the spelling of "LABELED" (only one L between the E's).  When this
  1075. setting is in effect, the sending Kermit sends not only the file's data, but
  1076. also system-dependent structural information about the file.  Since this
  1077. information can be quite lengthy, it is sent as if it were file data, in data
  1078. packets preceding the actual file data, rather than in attribute packets.  All
  1079. of this information is sent in binary mode (no conversions).
  1080.  
  1081. The receiver of the file can elect whether to act upon the label information,
  1082. or simply store it.  Kermit programs for simple file systems need no changes,
  1083. and can still act as repositories for labeled files, to be sent back later to
  1084. the system of origin.  On the other hand, labeled file transfers between like
  1085. systems (e.g. two VAX/VMS computers) will result in the reception of a file
  1086. with all the correct characteristics.
  1087.  
  1088. There are two possible modes of operation.  The first (and safer) mode
  1089. requires human intervention on a per-file basis, but an automatic mode is also
  1090. possible.
  1091.  
  1092. DEFINITIONS
  1093.  
  1094. A "bare" file is a file stored on disk that does NOT contain, as part of
  1095. its contents, Kermit-constructed label information.
  1096.  
  1097. A "labeled" file is a file stored on disk whose contents include
  1098. Kermit-constructed label information.
  1099.  
  1100. FORMAT OF A LABELED FILE
  1101.  
  1102. This description applies to a labeled file as stored on disk and to a bare
  1103. file that has label information added to it by a sending Kermit.
  1104.  
  1105.   BANNER  The literal text string "KERMIT LABELED FILE:", just the twenty
  1106.           characters within the quotes including the trailing colon, in 7-bit
  1107.           ASCII.  ASCII is used even on non-ASCII systems (like IBM mainframes
  1108.           with EBCDIC).
  1109.  
  1110.   IDLEN   A 2-byte ASCII decimal numeric length field, with leading zeros if
  1111.           necessary, for example "02".
  1112.  
  1113.   SYSID   The system ID, one of the codes listed on pp.275-278 of the Kermit
  1114.       book.  This is a string of 1 to 99 ASCII characters (but typically
  1115.           1-3 characters), for example "D7" for VAX/VMS.  The length of the
  1116.           SYSID field is given by the preceding field (IDLEN).
  1117.  
  1118.   LBLEN   A 2-byte ASCII numeric length field, with leading zeros if
  1119.  
  1120. Page 20                                               INFO-KERMIT DIGEST V12 #3
  1121.  
  1122.           necessary, for example "00", "02", "09", "99".
  1123.  
  1124.   LABEL   A label field, 1 to 99 ASCII characters.  The length is given by
  1125.           the preceding length field (LBLEN).  The LABEL field contains the
  1126.           name of the file descriptor data that follows.
  1127.  
  1128.   VALEN   An 8-byte ASCII numeric length field, with leading zeros if
  1129.           necessary, for example "00000512".  Zero ("00000000") is a
  1130.           permissible length for a value.
  1131.  
  1132.   VALUE   The value associated with the label, 0 to 99999999 bytes of
  1133.           information about the file in system-dependent format.  The labels
  1134.           and the format and layout of the associated values for each system
  1135.           type should be clearly defined and documented.
  1136.  
  1137.      The sequence (LBLEN, LABEL, VALEN, VALUE) may be repeated for as many
  1138.      labeled values are needed (if VALEN is zero, then VALUE is omitted).
  1139.  
  1140.   CONTENT The file's contents, encoded in whatever form that, in combination
  1141.           with the labeled values, allows for its eventual reconstruction in
  1142.           its original form.
  1143.  
  1144. All fields in a labeled file are subject to whatever encoding, prefixing, or
  1145. compression options have been negotiated by the two Kermit programs.
  1146.  
  1147. REQUIRED LABELS
  1148.  
  1149. The following labels (composed of ASCII uppercase letters) are reserved, and
  1150. are required in every labeled file:
  1151.  
  1152.   VERS    The operating system version, e.g. "5.3-1".  If there is no version
  1153.           to be specified, this label must still be present, but with a
  1154.           zero-length value.
  1155.  
  1156.   DATA    This is the final label.  It always has a zero-length value, which
  1157.           is followed immediately by the file's contents, as defined above.
  1158.  
  1159. EXAMPLE
  1160.  
  1161. A hypothetical example for VMS (ignore the indentation and line breaks;
  1162. the length fields are marked with L's):
  1163.  
  1164.                       LL  LL    LLLLLLLL     LL   LLLLLLLL
  1165.   KERMIT LABELED FILE:02D704VERS000000055.3-103FAB00000512<512 bytes here>
  1166.   03ACL00000723<723 bytes here>02DATA00000000<file's contents here>
  1167.   LL   LLLLLLLL                LL    LLLLLLLL
  1168.  
  1169. Note: FAB and ACL are not necessarily real labels.
  1170.  
  1171. EFFECTS ON THE KERMIT PROTOCOL
  1172.  
  1173. When SET FILE TYPE LABELED is in effect, the sending Kermit should include
  1174. the file type (") attribute, specifying a type of binary (B), if attribute
  1175. packets have been negotiated.  Otherwise, there is no effect on the protocol
  1176. at all.  In particular, a label-wise Kermit program can interoperate perfectly
  1177. well with another Kermit program that is completely ignorant of labeling, in
  1178.  
  1179. INFO-KERMIT DIGEST V12 #3                                               Page 21
  1180.  
  1181. which case the latter program simply "archives" or "unarchives" the file,
  1182. labels and all (provided the user remembered to place it in binary mode).
  1183.  
  1184. This proposal does not rule out the system-independent type of file archiving,
  1185. based on file attribute packets, that is described in the Kermit book.
  1186. However, if the "system-dependent" attributes suggested by this proposal can
  1187. be worked out in a sufficiently generic fashion, this may lead to a more
  1188. effective type of transmission of complex files between unlike systems that
  1189. share similar types of file characteristics (block size, record format,
  1190. character set, carriage control, etc).
  1191.  
  1192. On the other hand, for the first pass at an implementation (for VMS), it might
  1193. make more sense to simply have a single system-dependent label, like "FAB",
  1194. whose value is simply the File Access Block, and perhaps another one like
  1195. "ACL" for its Access Control List.  The two approaches do not rule each other
  1196. out.
  1197.  
  1198. USER INTERFACE
  1199.  
  1200. When SENDING a file, the user should first determine whether it is a bare file
  1201. or a labeled file.  This can be done by visual inspection (looking at the
  1202. first 20 bytes), or from memory (e.g. because all labeled files have a
  1203. particular filetype, or are kept together in a certain directory).  Then:
  1204.  
  1205.   (a) If it is a bare file, use SET FILE TYPE LABELED if you want to send
  1206.       label information too.
  1207.  
  1208.   (b) If it is a labeled file, use SET FILE TYPE BINARY.
  1209.  
  1210. When receiving a file, use SET FILE TYPE LABELED if you want labels in the
  1211. incoming file to be interpreted and applied, and:
  1212.  
  1213.   (a) The other Kermit is sending a bare file from a like system using SET
  1214.       FILE TYPE LABELED, or:
  1215.  
  1216.   (b) The other Kermit is sending a labeled file, of the receiving system's
  1217.       system type, from any kind of system, without SET FILE TYPE LABELED.
  1218.  
  1219. The receiving Kermit program that has been given a SET FILE TYPE LABELED
  1220. command must inspect the incoming data.  The first 30-36 bytes of the first
  1221. packet contain the BANNER, IDLEN, SYSID, and the first LBLEN, and LABEL (which
  1222. should be "04VERS").  If these five items are in correct format, and the SYSID
  1223. matches the program's own, then the program will accept the file in labeled
  1224. mode.  Otherwise, it will treat it as a binary file and store all the data,
  1225. including BANNER, SYSID, LABELs, VALUEs, and all length fields.
  1226.  
  1227. If a receiving Kermit program has accepted the file in labeled mode, but then
  1228. encounters an unknown label or other inconsistency in the labeled file format,
  1229. it must interrupt the file transfer (by putting an X in the data field of the
  1230. first available ACK) and discard the file.  This is because it cannot be
  1231. expected to back up and undo whatever label interpretation it has already done.
  1232.  
  1233. It is also conceivable that "similar" systems -- such as the DEC operating
  1234. systems VMS and RSX-11 -- might be able to honor each other's labels.  This
  1235. eventuality should be allowed for, possibly with SET commands.
  1236.  
  1237.  
  1238. Page 22                                               INFO-KERMIT DIGEST V12 #3
  1239.  
  1240. At the discretion of the programmer, a command such as SET FILE
  1241. LABEL-DETECTION { AUTOMATIC, MANUAL } could be installed.  If so, the default
  1242. must be MANUAL, that is, processing of labeled files occurs only if the user
  1243. asks for it explicitly.  For an arriving file, AUTOMATIC mode would detect the
  1244. BANNER, SYSID, and first LABEL and go into action as if the user had SET FILE
  1245. TYPE LABELED.  When sending a file, the program would avoid adding surrounding
  1246. label information if the user had already SET FILE TYPE LABELED.  This sort of
  1247. thing is obviously dangerous, and users should be cautioned about it.
  1248.  
  1249. POSTPROCESSING
  1250.  
  1251. Separate programs should be provided to translate a labeled file into a real
  1252. file, to be used in case the user forgot to SET FILE TYPE LABELED, or the file
  1253. was deposited by means other than Kermit.  Such a program might have two
  1254. options (or come in two forms): interpret the labels, and simply strip the
  1255. labels.
  1256.  
  1257. ACKS
  1258.  
  1259. Thanks to Terry Kennedy and Frank da Cruz for suggesting this idea, and to Joe
  1260. Doupnik, John Chandler, and Paul Placeway for many valuable suggestions.
  1261. Comments welcome!
  1262.  
  1263. ------------------------------
  1264.  
  1265. Date: Mon, 27 Aug 90 10:13:45 EDT
  1266. >From: Christine M. Gianone <cmg@watsun.cc.columbia.edu>
  1267. Subject: And ANOTHER New Test Release of MS-DOS Kermit 3.02
  1268. Keywords: MS-DOS Kermit 3.02
  1269.  
  1270. >From JRD's update notes since V12 #2:
  1271.  
  1272. 48. Add recognition of IBM PC video boards STB VGA/EM Plus (Tseng 4000)
  1273.     and Everex Viewpoint EV-678, from Terry Kennedy. The list of boards known
  1274.     to Kermit for switching between 80 and 132 columns is now:
  1275.     ATI EGA Wonder
  1276.     AT&T
  1277.     Everex Viewpoint EV-659, FVGA-673, EV-678, Micro Enhancer Deluxe
  1278.     Paradise AutoSwitch EGA Mono
  1279.     STB VGA/EM (Tseng TVGA)
  1280.     STB VGA/EM Plus (Tseng 4000), VGA/EM-16, VGA/EM-16 Plus
  1281.     Tseng Labs EVA board w/132-col kit installed
  1282.     Tseng Labs UltraPAK mono/Hercules w/132 column modes
  1283.     Video 7 Vega Deluxe w/ 132X25.COM driver installed and Video 7 VGA
  1284.  
  1285. 49. Add a facility to the keyboard translator.  Let key definition strings
  1286.     of the form {\Kverb other material} activate an existing Macro if and
  1287.     only if the verb is not in the list of those predefined for the keyboard
  1288.     translator.  If "verb" matches the name of a regular Kermit Macro execute
  1289.     the Macro and stay on the Kermit command line; thus a CONNECT command is
  1290.     needed to return to Connect mode. If verb does not match a Macro then do
  1291.     nothing and stay in Connect mode. Be aware that searching tables for the
  1292.     name of the verb uses many cpu cycles and will reduce performance.
  1293.  
  1294.     Examples:
  1295.         DEFINE Accumlate Receive, Connect
  1296.  
  1297. INFO-KERMIT DIGEST V12 #3                                               Page 23
  1298.  
  1299.         SET KEY \315 {\Kaccumulate}
  1300.  
  1301.     Pressing the IBM PC "F1" key (code \315 above) makes the keyboard
  1302.     translator see string {\Kaccumlate}, verb "accumlate" is not in the
  1303.     predefined tables, and hence Kermit invokes the Macro named ACCMULATE.
  1304.     That Macro puts Kermit into file receive mode and when the file, or
  1305.     file group, transfer has been completed it returns to Connect mode.
  1306.  
  1307.         DEFINE WP Take wp30.ini, Connect
  1308.         SET KEY \316 {\Kwp}
  1309.  
  1310.     This makes IBM PC key "F2" invoke macro WP which in turn reads in
  1311.     and executes the contents of file WP30.INI. Kermit then returns to
  1312.     Connect mode. Nothing happens if the user has not defined Macro WP.
  1313.  
  1314.     Braces around the whole key definition are required so that the search
  1315.     for the verb name is done while executing Connect mode rather than when
  1316.     defining the key. If the braces are omitted then only the predefined
  1317.     keyboard verbs are available; an error will be reported if the verb is
  1318.     not one of these.
  1319.  
  1320.     Text sent to the host may preceed and follow the \Kverb part. To avoid
  1321.     confusion "Kverb" itself, but not the "\", may be placed in braces too,
  1322.     such as:
  1323.  
  1324.     DEFINE Accumulate cd c:\junkmail,Receive,Connect
  1325.     SET KEY \315 {kermit\13\send foo.bar\13\{KAccumulate}exit\13}
  1326.     the pieces:   |_______________________||............||_____|
  1327.                   string to host          do Macro    string to host
  1328.  
  1329.     This sends the string "Kermit<carriage return>Send foo.bar<carriage ret>"
  1330.     to the host to invoke its Kermit and asks it to send file foo.bar. It
  1331.     then invokes local Macro Accumulate to receive the file, and finally it
  1332.     sends the string "exit<carriage return>" to the host. Braces around
  1333.     KAccumlate deliniate it from the final string text.
  1334.  
  1335.     There is no host command which can invoke these Macros; this protects the
  1336.     PC against stray garbage text and unauthorized commands from the host.
  1337.     Files changed are MSUIBM.asm (procedure keysv) and MSYIBM.asm.
  1338.  
  1339. 50. Reinstate Dump screen: filename on Status display. msssho, msyibm
  1340.  
  1341. 51. Correct a misprint in the National Replacment Character sets for Spanish.
  1342.     Apparently a DEC typist inserted one accent mark too many in the table.
  1343.         char value    was        is now
  1344.     60h \96        accent        accent        just for reference
  1345.     7bh \123    accent        degree
  1346.     7ch \124    degree        n tilde
  1347.     7dh \125    n tilde        c cedilla
  1348.     7eh \126    c cedilla    tilde
  1349.     Corrections are in file mszibm.asm
  1350.  
  1351. 52. Embellish the VT300 character set abilities by permitting National
  1352.     Replacement Character (NRC) sets to be "designated" by the host to a
  1353.     particular Gn set table, viz:
  1354.  
  1355.  
  1356. Page 24                                               INFO-KERMIT DIGEST V12 #3
  1357.  
  1358.     Designator         Mnemonic       Description of Action
  1359.     ESC ( <ident>    SCS    Designates 94 byte character set <ident> to G0
  1360.     ESC ) <ident>    SCS    Designates 94 byte character set <ident> to G1
  1361.     ESC * <ident>    SCS    Designates 94 byte character set <ident> to G2
  1362.     ESC + <ident>    SCS    Designates 94 byte character set <ident> to G3
  1363.  
  1364.     where <ident> is:
  1365.     NRC country  <ident>        NRC country  <ident>
  1366.     British        A        Italian        Y
  1367.     Dutch        4        Norwegian/Danish ' (hex 60) or E or 6
  1368.     Finnish        5 or C        Portuguese    %6
  1369.     French        R        Spanish        Z
  1370.     French Canadian    9 or Q        Swedish        7 or H
  1371.     German        K        Swiss        =
  1372.  
  1373.    Example: ESC ) A  loads the British NRC set (A) into G1 (the ")" part).
  1374.    To load Latin1 into G1 use  ESC - A  because Latin1 is a 96 character set.
  1375.    NRCs are all 94 character sets.
  1376.  
  1377.    This embellishment does NOT turn on NRC operation in the Digital Equipment
  1378.    Corporation sense. To follow DEC usage employ SET TERM CHARACTER-SET to
  1379.    choose the country and have the host enable or disable NRCs by sending
  1380.    CSI ? 42 h  or CSI ? 42 l, respectively.
  1381.    Changes are in file mszibm.asm.
  1382.  
  1383. 53. Minor code corrections to parsing eight bit controls introduced by item 52.
  1384.     File mszibm.
  1385.  
  1386. 54. Incorporate support for Novell's TELAPI TCP/IP Telnet program. This
  1387.     is selected by command
  1388.  
  1389.             SET PORT TELAPI nnn.nnn.nnn.nnn
  1390.  
  1391.     where nnn.nnn.nnn.nnn is the Internet address of the remote host, in
  1392.     decimal.
  1393.  
  1394.         Example: SET PORT TELAPI 129.123.1.11   (machine netlab.usu.edu)
  1395.  
  1396.     TELAPI is a component of the Novell package LAN WorkPlace for DOS. At
  1397.     Utah State University it was run with the Novell (ex-Excelan) EXOS 205T
  1398.     Ethernet board (a smart mutiprotocol board supporting TCP with NetWare
  1399.     IPX 802.3 style packets). Novell Utility TSU.EXE is not required.
  1400.  
  1401.     SET PORT TELAPI must not be confused with SET PORT NOVELL(NASI); they
  1402.     use very different communications methods.
  1403.  
  1404.     Notes on this new faciity:
  1405.  
  1406.      a) Hosts not on the local network will likely require their
  1407.         Internet address and the local gateway to be specified with utility
  1408.         ROUTE, such as
  1409.             ROUTE ADD 128.59.39.2 129.123.1.254
  1410.     The first address is watsun.cc.columbia.edu, in New York City,
  1411.     and the second is that of a local gateway box (a cisco, Inc router
  1412.     at Utah State University in Logan, Utah).
  1413.  
  1414.  
  1415. INFO-KERMIT DIGEST V12 #3                                               Page 25
  1416.  
  1417.      b) The communications channel is opened as an 8-bit wide BINARY
  1418.         character at a time mode, thus Unix hosts may require Line Feed
  1419.     (Control J) as a line terminator. See also item g) below regarding
  1420.     character \255.
  1421.  
  1422.      c) Included in this work is sending a BREAK signal as a Telnet
  1423.         "Interrupt Process" command.
  1424.  
  1425.      d) A status message is shown for 3 seconds if the connection cannot
  1426.         be established.
  1427.  
  1428.      e) Once an Internet number has been specified with SET PORT TELAPI it
  1429.         will be retained for subsequent SET PORT TELAPI commands. Thus it
  1430.     need be stated only once. SHOW COMMUNICATIONS will display the number.
  1431.  
  1432.      f) The presence of the host can be checked by sending the two bytes
  1433.          \255\246  which is Telnet signal "Are You There", and to which the
  1434.     host is supposed to send back a bell or visible message. This can
  1435.     be arranged easily by command SET KEY <key's ident here> \255\246
  1436.     if SET TERMINAL CHARACTER-SET TRANSPARENT is selected.
  1437.         See TCP/IP documents RFC-854 and RFC-855 for Telnet specifications.
  1438.  
  1439.      g) Character \255 is special in Telnet; it introduces Options negotiation.
  1440.         To send \255 as data it must be sent twice, as \255\255 or use the
  1441.     new command SET SEND DOUBLE \255 to do this for Kermit packets.
  1442.         Please be aware of this character when sending Binary files.
  1443.  
  1444.      h) TELAPI version 3.5 sometimes has trouble hanging up a connection. It
  1445.         will attempt creating a new one for each new session and can exhaust
  1446.     its space (defaults to two of them) when an old one persists. The
  1447.     only cure I know is to unload the TELAPI TSR (DOS command TELAPI -u)
  1448.     or reboot the machine.
  1449.  
  1450.      i) TELAPI version 3.5 is also notably slow in delivering characters to
  1451.         terminal emulators. Expect an effective 2400 baud rate.
  1452.  
  1453.      j) TELAPI uses Interrupt 14h with high numbered function calls. Beware
  1454.         of using other Int 14h trapping programs simultaneously.
  1455.  
  1456.      k) Kermit uses only the Internet number form of addressing for TELAPI
  1457.         because the TCP "name resolver" (conversion of a name to a number)
  1458.     is not available directly with TELAPI. Thus, full compliance with
  1459.     TCP/IP specs is not available to Kermit or other external terminal
  1460.     emulators using the TELAPI interface. The full specs recommend using
  1461.     names and letting a separate group of machines, called Name Servers,
  1462.     provide the Internet number via TCP/IP methods.
  1463.  
  1464.     All the changes are confined to file msxibm.asm
  1465.  
  1466. 55. Correct a real bug: if SET LOCAL ON and while in Connect mode Control-
  1467.     PrintScreen (copy screen to printer) are active then Kermit crashes the
  1468.     system. The reason is a stack overflow when the same function ends up
  1469.     calling itself again by roundabout means. Adequate correction required
  1470.     small revisions about what is or is not echoed locally. Now almost every
  1471.     thing except DEC status reports is echoed (my real VT320 seems to do the
  1472.     same). Affected files are msyibm.asm and mszibm.asm, done on 14 August.
  1473.  
  1474. Page 26                                               INFO-KERMIT DIGEST V12 #3
  1475.  
  1476.  
  1477. 56. Add Enviroment and command line SET COM1 and COM2 to the list as per
  1478.     item 4 above.
  1479.  
  1480. 57. Significant change to \%n variables defined as macro arguments. For each
  1481.     Macro invocation first save variables \%0 through \%9, clear \%0..\%9,
  1482.     define \%0 to be the name of the macro being invoked and \%1..\%9 as
  1483.     the arguments to the macro. When the macro exits restore the previous
  1484.     \%0..\%9 variables. This process nests naturally, and ARGC still
  1485.     represents the one plus the number of macro arguments. C Kermit 5A and
  1486.     MS-DOS Kermit 3.02 are now in agreement on these items.
  1487.     Previously \%0 was not affected by macro invokation and \%1..\%9 were
  1488.     not saved and restored around the macro. The new behavior may affect
  1489.     scripts written in the past. File mssset.asm
  1490.  
  1491. 58. Add alternative command syntax:
  1492.     SET FILE COLLISION means the same as older SET FILE WARNING
  1493.     and SET FILE COLLISION DISCARD is the same as SET FILE COLL NO-SUPERSEDE.
  1494.     Changes are in mssset.asm and msssho.asm
  1495.  
  1496. 59. Correct problems with REM LOGIN. Is the last of problems in this area?
  1497.  
  1498. 60. Correct problems with CTS/RTS handshake timing and half duplex. Tnx to
  1499.     Brian Holley for spotting them. msxibm.asm
  1500.  
  1501. 61. Add DESQview support call to release the cpu timeslice when in Connect
  1502.     mode and no character is available from the serial port. msyibm.asm
  1503.  
  1504. 62. Modify the retry procedure to send a packet character to wait 60ms rather
  1505.     than 10ms between each of five attempts. This may help situations where
  1506.     a network device may be unresponsive for about 200ms between characters.
  1507.     msscom.asm
  1508.  
  1509.   More changes are expected.
  1510.  
  1511. The mstibm.boo file and the source files are in kermit/test on watsun and in
  1512. the T: area of KERMSRV.  The binary .EXE file is available on watsun only in
  1513. kermit/bin/mstibm.exe.  There is also a test 3.02 version for the DEC Rainbow,
  1514. contributed by Robert Weiner of Cooper Union.  It is in kermit/test/mstrb1.boo
  1515. and kermit/bin/mstrb1.exe.  The latest Rainbow test version tracks the non-IBM
  1516. specific portions of MS-DOS Kermit, and adds some new character translations
  1517. for screen dump and print screen.  Thanks to Joe and Robert for their
  1518. continuing work on MS-DOS Kermit!
  1519.  
  1520. ------------------------------
  1521.  
  1522. Date: Mon, 27 Aug 90 14:13:12 EDT
  1523. >From: Christine M. Gianone <cmg@watsun.cc.columbia.edu>
  1524. Subject: A New Version of Kermit for OS/2 Presentation Manager
  1525. Keywords: OS/2 Presentation Manager Kermit
  1526.  
  1527. Contributed by Brian R. Anderson of Burnaby, BC, Canada.  Here is the note
  1528. that accompanied Brian's submission:
  1529.  
  1530. "The version of Kermit contained herein is for OS/2-PM.  It was written from
  1531. scratch by the author (Brian R. Anderson) in Stony Brook Professional
  1532.  
  1533. INFO-KERMIT DIGEST V12 #3                                               Page 27
  1534.  
  1535. Modula-2.  A two part article in the September/October 1990 issues of Dr.
  1536. Dobb's Journal describe this implementation (which is a port of an earlier
  1537. program that I wrote for DOS in Logitech Modula-2).
  1538.  
  1539. This program is public domain, and may be copied freely, modified/enhanced,
  1540. and distributed freely (in original or enhanced form).
  1541.  
  1542. Included on the disk are an executable version of the program
  1543. (PCKERMIT.EXE); 17 Modula-2 source code files (8 .DEF and 9 .MOD); icon and
  1544. resource files; make and linker response files, and the EXE definition file.
  1545. To recompile the program: M2MAKE PCKERMIT <cr>.  The program includes:
  1546.  
  1547.       Kermit single and multiple file transfer
  1548.       TVI950 Terminal Emulation (optimized for IBM7171/PROFS)
  1549.       Automatic renaming of files in case of name clash
  1550.       Continuous display of communications settings"
  1551.  
  1552. Brian indicates that this program is the starting point for a program that
  1553. will be sold commercially.  Further details are in the file O2AAAA.HLP.
  1554.  
  1555. The files are in the Kermit test area for now, pending reactions from the
  1556. Kermit user community:
  1557.  
  1558. watsun.cc.columbia.edu    BITNET KERMSRV@CUVMA    Comments
  1559.   kermit/test/o2aaaa.hlp    T:O2AAAA.HLP            "boo" encoding of program
  1560.   kermit/bin/o2aker.exe                             Executable program
  1561.   kermit/test/o2aker.boo    T:O2AKER.BOO            "boo" encoding of program
  1562.   kermit/bin/o2aker.ico                             Program icon
  1563.   kermit/test/o2aico.boo    T:O2AICO.BOO            "boo" encoding of icon
  1564.   kermit/test/o2asrc.jar    T:O2ASRC.JAR            Source code
  1565.  
  1566. The boo files can be decoded into the original binary files using any of the
  1567. msbpct programs available in kermit/a.  The "jar" file is a text archive of
  1568. the source files that can be picked apart with a text editor or a simple
  1569. program.  Brian's submission did not include a user manual.
  1570.  
  1571. Many thanks to Brian for contributing this program.
  1572.  
  1573. ------------------------------
  1574.  
  1575. Date: Sat, 28 Jul 90 12:10:27 PDT
  1576. >From: rochester!ames!claris!portal!cup.portal.com!R_Tim_Coslet@columbia.edu
  1577. Subject: New Release UCPECAN Kermit V1.1
  1578. Keywords: Pecan, USCD Pascal Pecan Kermit
  1579.  
  1580. I am submiting a minor upgrade to UCPECAN my Kermit-UCSD (Version 1.1).
  1581.  
  1582. You should recieve 2 files: 1) a large (about 1400 lines) one containing
  1583. the source (UCPUPD.PAS) and a very short one containing "installation
  1584. instructions" and a brief list of changes made to V1.1 by this upgrade
  1585. (UCPUPD.UPD).  The changes are
  1586.  
  1587. 1)  Added a new Unit (DIR.FIXES) to work around problems
  1588.     encountered with SFS. This update removes Limitation #4,
  1589.     making BINARY file transfers exact in both SFS and AFS
  1590.     implementations.
  1591.  
  1592. Page 28                                               INFO-KERMIT DIGEST V12 #3
  1593.  
  1594.  
  1595. 2)  Fixed a bug that caused an incorrect time to be sent in
  1596.     the attribute packet if no time was specified with the
  1597.     creation date (this bug affected both AFS and SFS).
  1598.  
  1599.                       R. Tim Coslet
  1600.  
  1601. Usenet: R_Tim_Coslet@cup.portal.com
  1602. BIX:    r.tim_coslet
  1603.  
  1604. [Ed. - Many thanks, R. Tim!  The new files have been installed in the Kermit
  1605. Distribution C area.  Keep up the good work!]
  1606.  
  1607. ------------------------------
  1608.  
  1609. Date: Tue, 07 Aug 90 02:00:41 MEZ
  1610. >From: "Gisbert W.Selke" <UPP101%DBNRHRZ1@cuvmb.cc.columbia.edu>
  1611. Subject: MS-Kermit 3.0x and German telephones
  1612. Keywords: MS-DOS Kermit 3.0, German
  1613.  
  1614. Here's a (partial) solution to a problem no-one besides me seems to have had:
  1615. I'm running MS-DOS Kermit over a 1200 bps line in what is yet West Germany.
  1616. For some obscure reason, the German PTT transmits the signal to my phone that
  1617. can be used to record the fee due. (No, I never had such a gadget. Maybe some
  1618. former owner of this flat had.) Unfortunately, the tone used for this is
  1619. right within the range used for 1200 bps transmissions. So, if I'm lucky,
  1620. I get a corrupted packet every 8 to 12 minutes (for local calls - much more
  1621. often for trunk calls); if I'm in terminal mode, however, I get some 10
  1622. scrambled characters on the screen. So far, so annoying. But you ain't seen
  1623. nothing yet: some of the garbage obviously contained shift-in/-out sequences,
  1624. so that I was left with funny graphics characters all over the screen
  1625. henceforward, where ordinary 7-bit ASCII should have been. The solution came
  1626. with one of the 3.02 revisions of MS-Kermit: 'set term char ascii g1' takes
  1627. care of the odd shift!
  1628.  
  1629. Still, some of the garbage even throws me out of that; apparently, it loads
  1630. a different character set into gx, where x = 1..3. So, if anyone has a way
  1631. of doing away with the disturbances altogether, any help/hints/pointers will
  1632. be appreciated, with software solutions muchly preferred.
  1633.  
  1634. \Gisbert   c/o  <UPP101.DBNRHRZ1.BITNET>
  1635.  
  1636. ------------------------------
  1637.  
  1638. Date: Mon Aug 27 10:19:29 1990 EDT
  1639. >From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
  1640. Subject: Looking for Stories
  1641.  
  1642.  ... of how Kermit has been used in imaging and visualization applications,
  1643. including modelling, CAD, medical imaging, and so forth, for possible
  1644. publication.  MS-DOS Kermit graphics, Kermit protocol imbedded in imaging or
  1645. CAD systems, whatever you know about ...
  1646.  
  1647. ------------------------------
  1648.  
  1649. End of Info-Kermit Digest
  1650.  
  1651. INFO-KERMIT DIGEST V12 #3                                               Page 29
  1652.  
  1653.  
  1654. Page 30                                               INFO-KERMIT DIGEST V12 #4
  1655.  
  1656. Info-Kermit Digest         Tue, 11 Sep 1990        Volume 12 : Number 4
  1657.  
  1658.   MS-DOS KERMIT:
  1659.  
  1660.         Next Test Release of MS-DOS Kermit 3.02
  1661.         Kermit on Novell Networks
  1662.         Feedback on MS KERMIT 3.02 of July 20
  1663.         Kermit Problems - V3.01
  1664.         MS-Kermit and VAX ALLin1
  1665.         Question about Kermit 3.01 and the 128-byte history buffer
  1666.         Kermit & WordPerfect under PCSA
  1667.         Re: MSKERMIT <-- TCP/IP --> Unix-Host
  1668.  
  1669. Digest submissions may be sent to Info-Kermit@WATSUN.CC.COLUMBIA.EDU,
  1670. requests for addition to or deletion from the Info-Kermit subscriber list to
  1671. Info-Kermit-Request@WATSUN.CC.COLUMBIA.EDU or to KERMIT@CUVMA.BITNET.
  1672.  
  1673. Kermit files may be obtained over networks and by mail order.  On the
  1674. Internetwork, use FTP to log in to host WATSUN.CC.COLUMBIA.EDU, a SUN-4/280
  1675. running UNIX (SUNOS 4.1), IP host number 128.59.39.2.  Login as user anonymous
  1676. (note, lower case), any password, and GET or MGET (MULTIPLE GET) the desired
  1677. files.  The Kermit files are in directories kermit/a, kermit/b, kermit/c,
  1678. kermit/d, and kermit/e.  Test versions are in kermit/test.  Binaries are in
  1679. kermit/bin (use ftp in binary mode).  You can also get Kermit files over the
  1680. BITNET/EARN network; to get started send a message with text HELP to KERMSRV,
  1681. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  1682. file kermit/a/aanetw.hlp (AANETW.HLP on KERMSRV).  To order by mail, request a
  1683. complete list of Kermit versions and an order form from Kermit Distribution,
  1684. Columbia University Center for Computing Activities, 612 West 115th Street,
  1685. New York, NY 10025 USA.
  1686.  
  1687. ----------------------------------------------------------------------
  1688.  
  1689. Date: Mon, 10 Sep 90 16:03:23 EDT
  1690. >From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
  1691. Subject: Next Test Release of MS-DOS Kermit 3.02
  1692. Keywords: MS-DOS Kermit 3.02
  1693.  
  1694. 63. Correct a small problem leading to confusion in terminal emulation if an
  1695.     arriving control sequence is terminated prematurely by another. msscmd.asm
  1696.  
  1697. 64. Add command REMOTE SET FILE COLLISION UPDATE, meaning the file receiver
  1698.     on the other side should process files with the same name as "updates"
  1699.     (whose meaning is system dependent) to existing files. A MS-DOS Kermit
  1700.     server does not recognize the command. mssser.asm
  1701.  
  1702. 65. Another small adjustment to item 62. Move the long interval retrying of
  1703.     sending characters to be at the final launching point for BIOS comms
  1704.     channels. The intent is to compensate for a particular X.25 channel which
  1705.     rejects new characters with a failure indication; that channel is known
  1706.     to be very sluggish. This change permits character retries for Connect
  1707.     mode and scripts as well as packets. msscom.asm, msxibm.asm
  1708.  
  1709. 66. Ensure that the graphics terminal emulation writes characters as opaque,
  1710.     when so selected, for black and white displays. msgibm.asm
  1711.  
  1712.  
  1713. INFO-KERMIT DIGEST V12 #4                                               Page 31
  1714.  
  1715. 67. Accomodate LATIN5/Cryillic character set of Code Page 866 as much as
  1716.     possible. The command SET TRANSFER CHARACTER-SET has the new option
  1717.     LATIN5, and if selected will force FILE CHARACTER-SET to be CP866.
  1718.     Similarly, if the FILE CHARACTER-SET is selected to be the new option
  1719.     CP866 then it will force TRANSFER CHARACTER-SET to be LATIN5.
  1720.     Warnings are issued for both cases.
  1721.  
  1722.     When sending a text file with TRANSFER CHARACTER-SET LATIN5 the file
  1723.     attributes packet will send "I6/144" for the file character-set
  1724.     identification (the "C" attribute) and no translations will be done.
  1725.  
  1726.     When receiving such a file no translation will be done, regardless of
  1727.     the currently active FILE CHARACTER-SET. Actually the file should be
  1728.     rejected unless the user has stated SET UNKNOWN-CHARACTER-SET to KEEP,
  1729.     but the code gets very messy if the rule is enforced at this time.
  1730.  
  1731.     The reason is for the coupling of commands is CP866 has many Cryillic
  1732.     characters in the right, high bit set, section which have no equivalents
  1733.     in other Kermit character sets and regular Code Pages. Hence, when LATIN5
  1734.     or CP866 is selected there can be no translation of characters between
  1735.     CP866 and either the other (Western European) Code Pages or the LATIN1
  1736.     character set. Thus LATIN5 or CP866 selects a Transparent file transfer
  1737.     but with a LATIN5 "I6/144" character set identification.
  1738.  
  1739.     For terminal emulation SET TERMINAL CHARACTER-SET TRANSPARENT will prevent
  1740.     unwanted translations and also convert the C1 controls area into printable
  1741.     characters found in CP866.
  1742.     Files affected are mssfil.asm, mssset.asm, msyibm.asm.
  1743.  
  1744. 68. Add communications enhancement to using the Int 14h serial port
  1745.     interceptor TES from Interconnections Inc (distributed with Novell
  1746.     NetWare for VMS). The new command
  1747.  
  1748.         SET PORT TES <optional host name>
  1749.  
  1750.     invokes TES and make a connection to the indicated host. This competes
  1751.     with SET PORT BIOSn because they use some common operations.
  1752.  
  1753.     <optional host name>        Action
  1754.         host-name       Connect to that host, and disconnect from a previous
  1755.                         TES host if such a connection were active.
  1756.            *            Show all available TES hosts but do not connect.
  1757.         nothing         Use the current TES host name to reestablish an
  1758.                         existing connection. Most useful when switching
  1759.                         between TES and a real serial port. If no host name
  1760.                         exists then a list of available hosts is shown.
  1761.  
  1762.     Keyboard verb \Knethold is used to invoke the TES command interpreter
  1763.     while in Connect mode (to revise connections usually), and HANGUP breaks
  1764.     an existing connection. Sending a BREAK (ALT-B or \Kbreak) performs the
  1765.     same function for TES as \Knethold. Use this function when manual control
  1766.     of sessions is desired, such as creating multiple sessions managed by
  1767.     TES rather than by Kermit.
  1768.  
  1769.     These new TES operations couple closely with version 2.x of TES and will
  1770.     not work with earlier ones. Instead use SET PORT BIOS1 and the TES hot
  1771.  
  1772. Page 32                                               INFO-KERMIT DIGEST V12 #4
  1773.  
  1774.     key with older TES's. If the TES hot key is used then \Knethold may not
  1775.     be recognized; use the hot key again to control TES. If the hot key has
  1776.     not been used then TES material is shown on the Connect mode screen.
  1777.     SHOW COMMUNICATIONS will show the current host name, as read back from
  1778.     TES itself. The apparent serial port identifier on the Connect status
  1779.     line is the letter "I".
  1780.  
  1781.     Note that placing a connection on hold and later returning to it may
  1782.     require pressing the \knethold or \kbreak keys and typing RESUME to make
  1783.     the TES command interpreter return control to the data stream. It's
  1784.     apparently a problem in the interpreter.
  1785.  
  1786.     If Kermit is exited completly then the active session is terminated, but
  1787.     held sessions are retained. Thus, to preserve sessions after Kermit has
  1788.     exited use \knethold or \kbreak to get the TES command interpreter and
  1789.     place the connection on hold manually. The session may be resumed as above.
  1790.  
  1791.     I am testing with TES version 2.1 beta so both sides have growing pains.
  1792.     Modifications are all in file msxibm.asm
  1793.  
  1794. 69. Make an attempt to pass macro arguments \%0..\%9 through (around) a
  1795.     TAKE command to keep some consistency with MS-DOS Kermit 3.00 and 3.01.
  1796.     For example,
  1797.         DEFINE DIAL Take Hayes.tak      which has dialing commands
  1798.         DIAL 555-1212                   we want the number passed down to the
  1799.                                         interior of Hayes.tak as \%1.
  1800.  
  1801.     where file Hayes.tak (from the diskette with Chris' book) starts off as
  1802.         COMMENT - Initial setup
  1803.           def errstop echo Error: \%1\13,def \%1,hang,stop
  1804.           set speed 2400                ; *** Change to 1200 ...
  1805.           <more stuff in here>
  1806.           echo \13Dialing \%1.  Wait...\13\10  ; Dialing for the first time,
  1807.           goto dial                     ; so skip the "Redialing" message
  1808.  
  1809.         COMMENT - Dialing loop
  1810.         :REDIAL
  1811.           echo \13Redialing...\13\10    ; Give this message when redialing
  1812.         :DIAL
  1813.           output ATDT\%1\13             ; Dial the number
  1814.           <more stuff follows>
  1815.  
  1816.     Another example:
  1817.         DEFINE xxx Set term VT302,take x.tak,connect
  1818.         xxx A B C
  1819.     where file x.tak holds
  1820.         echo \%0 \%1 \%2 \%3
  1821.  
  1822.     This displays  XXX A B C and then enters Connect mode as a VT320 terminal.
  1823.  
  1824. 70. Add a new element to command line variables, taken from C Kermit 5A.
  1825.     Numerical variables ARGC, COUNT, ERRORLEVEL, and VERSION can be used as
  1826.     the ascii representation of their values by the new text substitution
  1827.     operator
  1828.  
  1829.         \v(numerical variable)
  1830.  
  1831. INFO-KERMIT DIGEST V12 #4                                               Page 33
  1832.  
  1833.  
  1834.     Soon this will be expanded to include other "named variables" such as
  1835.     TIME, DATE and so on which are also internal quantities in Kermit.
  1836.  
  1837.     Examples:
  1838.         ECHO The version number of this program is \v(version) today.
  1839.     which displays
  1840.         The version number of this program is 302 today.
  1841.     And DEFINE See ECHO \%0 - the num var is \v(\%1) here.
  1842.         See argc
  1843.     which displays
  1844.         SEE - the num var is 2 here.
  1845.         See version
  1846.     which displays
  1847.         SEE - the num var is 302 here.
  1848.  
  1849.     The expression \v(numerical variable) is replaced by it's ascii value.
  1850.     This may be used in any command. The current syntax of IF = ARGC 2 ...
  1851.     is retained for compatibility purposes but in new scripts it should be
  1852.     replaced by IF = \v(argc) 2  etc.
  1853.  
  1854.     If the item within the parentheses is not one of the above numerical
  1855.     variables, or if a right parenthesis does not end the variable name
  1856.     then the entire expression from \v( to the current point is discarded.
  1857.     Examples of bad syntax:
  1858.         ECHO testing \v(versions) stuff
  1859.     or
  1860.         ECHO testing \v(version stuff
  1861.     both of which display
  1862.         testing stuff
  1863.  
  1864.     All the changes for this item are in file msscmd.asm
  1865.  
  1866.  More changes are expected.
  1867.  
  1868. [Ed. - The Cyrillic and \v() items are not final.  An actual translation
  1869. between CP866 (Alternative Cyrillic) and ISO 8859-5 will probably be
  1870. installed before 3.02 is released.  \v(name) will probably be changed to
  1871. require the name of a builtin variable (which need not necessarily be
  1872. numeric) within the parentheses.]
  1873.  
  1874. ------------------------------
  1875.  
  1876. Date: Thu, 09 Aug 90 10:22:53 EDT
  1877. >From: Doug Salane <SALJJ%CUNYVM@cuvmb.cc.columbia.edu>
  1878. Subject: Kermit on Novell Networks
  1879. Keywords: MS-DOS Kermit, Novell Network
  1880.  
  1881. I am trying to use Kermit for station to station communication on my Novell
  1882. Network.  This is an ELS Level II Network which uses Arcnet cards in a bus
  1883. topology.  I would like to use Kermit to transfer files between machines
  1884. without relying on the server.  Is this possible?
  1885.  
  1886. I have tried using the SET PORT NETBIOS and SET PORT NOVELL after loading the
  1887. Novell IPX workstation shell.  I get an error message that says the Network is
  1888. not available.
  1889.  
  1890. Page 34                                               INFO-KERMIT DIGEST V12 #4
  1891.  
  1892.  
  1893. I'd appreciate your help. Thanks.
  1894.  
  1895. [From jrd - Doug, what's needed is NetBios running on top of IPX and NET3/4.
  1896. It's in the Novell NetWare collection. After that just follow the instructions
  1897. in the Kermit documentation about SET PORT NET <name>. I use Kermit that way
  1898. here, and it is also the official AT&T method of talking to an AT&T Unix
  1899. machine from a DOS PC. SET PORT NOVELL(NASI) uses the NASI/NACS async server,
  1900. not what you want at all.
  1901.         The short form of the documentation says start the Server Kermit
  1902. first (SET PORT NET, Server). No destination address is needed by the server.
  1903. Then start a client Kermit with SET PORT NET <server's NetBios name> and
  1904. issue file transfer or REMOTE commands.]
  1905.  
  1906. ------------------------------
  1907.  
  1908. Date: Sat, 21 Jul 90 14:23:19 PDT
  1909. >From: leland@scdt.intel.com (Leland Bruns )
  1910. Subject: Feedback on MS KERMIT 3.02 of July 20
  1911. Keywords: MS-DOS Kermit 3.02
  1912.  
  1913. I am forwarding some feed back on the test version of MS KERMIT version 3.02
  1914. and some comments on version 3 in general.  Please forward to the appropriate
  1915. persons.  Thank you.
  1916.  
  1917. ENVIRONMENT
  1918.  
  1919. First, my environment is a Wedge Technology 386-based PC-AT clone, with a
  1920. Video-7 VGA board.  I am compiling using MicroSoft's MASM version 5.01.  My
  1921. hosts are VAX VMS and Ultrix machines.  The PC's operating system in PC-DOS
  1922. version 3.30.
  1923.  
  1924. [From jrd - similar to my own environment: DELL 310, Video-7 VGA, VMS host,
  1925. PC-DOS 3.30]
  1926.  
  1927. I have recently copied MS KERMIT version 3.02 from the /kermit/test directory
  1928. at watsun.cc.columbia.edu.  My comments below are based on the versions dated
  1929. July 15 and July 20.  (I see there is a new version this morning, July 21, but
  1930. have not had the chance to port or test this version.)  I am comparing the
  1931. test version 3.02 to the original 3.00 distribution version.
  1932.  
  1933. TEKTRONIX EMULATION PROBLEMS
  1934.  
  1935. I see that there are significant changes to the MSGIBM.ASM module for
  1936. Tektronix emulation, and have concentrated on looking for differences there.
  1937. The one difference I would like to report is in the way GIN input apparently
  1938. works in the new version.  In the original 3.00 version, when a key was
  1939. entered at the PC, the host would receive the ASCII code for the character
  1940. entered, the 4 character xy coordinate, and one final character (I can't tell
  1941. easily from the host software I have to work with what that final character
  1942. is, as the software discards it).
  1943.  
  1944. The new version 3.02 test version apparently does not send the GIN report to
  1945. the host until the user has entered the desired key (at which time the
  1946. cross-hair cursors disappear) AND an additional carriage return.  Not too
  1947. surprising, the host receives an additional character which must be ignored to
  1948.  
  1949. INFO-KERMIT DIGEST V12 #4                                               Page 35
  1950.  
  1951. continue properly.
  1952.  
  1953. I believe that the behavior of the original version 3.00 of MS KERMIT is
  1954. correct, while that the new test 3.02 version is incorrect.  The user should
  1955. not have to enter both the desired key plus a carriage return for each GIN
  1956. input point.  I would very much like to see the original behavior restored.
  1957.  
  1958. [From jrd - advice is most welcomed because my Tek docs seem to differ
  1959. about what should happen (they are not consistent either).]
  1960.  
  1961. FEEDBACK & ENHANCEMENT REQUEST
  1962.  
  1963. Since I'm taking the time to report the problem above, I'd like at the same
  1964. time to pass on some positive feedback on MS KERMIT version 3 and an
  1965. enhancement request.  I very much like the work that has gone into MS KERMIT
  1966. between versions 2 and 3.  KERMIT is unique in its universal availability, its
  1967. faithful terminal and graphics emulation, and public availability of source
  1968. code.
  1969.  
  1970. The latter feature, availability of source code, is important to me as I have
  1971. one minor hardware incompatibility for which I must modify each KERMIT release
  1972. as it becomes available.  Specifically, I have dedicated hardware on all four
  1973. COM ports, and on IRQ2 thru 5 (see the comments below).  In order to make use
  1974. of my 9600 Baud Telebit Trailblazer modem on COM3, IRQ5, I must create a copy
  1975. of MS KERMIT modified to use IRQ5 rather than IRQ4.
  1976.  
  1977. I don't think my situation is too uncommon these days, as more and more
  1978. powerful PC's become available and are equipped with more and more hardware,
  1979. requiring use of COM ports 3 and 4.  I frequently see queries posted asking
  1980. how to get around the restrictions imposed by the current MS KERMIT.  I think
  1981. many users would benefit the ability to specify the interrupt number and
  1982. address on a per port basis...just as the newer MS KERMIT versions now allow
  1983. specification of the COM port address.  Failing that, the next best thing
  1984. would be to allow the user to define the interrupt numbers and addresses in a
  1985. straightforward manner in the MSXIBM.ASM source (vs. the convoluted encoding
  1986. required at present...see my difference file below).
  1987.  
  1988. [From jrd - But.... If one is not using all the COM ports simultaneously then
  1989. one can use two at once if they are on different IRQ lines. Just ask Kermit to
  1990. touch the competing ports so their interrupt driver chips are taken off the
  1991. bus (as decent software should have done anyway). I am reluctant to move far
  1992. from IRQ 3 and 4 because of the interference with other peripherals (IRQ 5 is
  1993. the hard disk controller on XTs). I can't easily change things in v3.02
  1994. anwyway because of the necessary revisions to the data structures. Btw, I too
  1995. have a Trailblazer (T2500), and I avoid most serial port conflicts via a $30
  1996. RS232 A-B-C-D box; COM1 is comms, COM2 is the mouse.]
  1997.  
  1998. Once again, let me say how much I appreciate the work going into MS KERMIT
  1999. version 3.  Thanx in advance for your consideration of the suggestions above.
  2000.  
  2001. [From jrd - Thank you for the suggestions and the compliment.]
  2002.  
  2003.                                         Leland M. Bruns
  2004.                                         Intel Corporation
  2005.                                         (408) 765-4219
  2006.                                         leland@scdt.intel.com
  2007.  
  2008. Page 36                                               INFO-KERMIT DIGEST V12 #4
  2009.  
  2010.  
  2011. ------------------------------
  2012.  
  2013. Date: Wednesday, 25 Jul 1990 13:06:12 EDT
  2014. >From: "Larry F Armbruster" <U22C5%WVNVM@cuvmb.cc.columbia.edu>
  2015. Subject: Kermit Problems - V3.01
  2016. Keywords: MS-DOS Kermit 3.01
  2017.  
  2018. I have experienced several problems using Kermit.  The configuration
  2019. of my system is:
  2020.  
  2021.   PS/2 Model 50Z with an AT&T Voice Data Multiplexer dialing into a
  2022.   DECserver 200 V3.0 (BL33) - LAT V5.1 and accessing VM/CMS through
  2023.   a DECnet/SNA Gateway into a SIM3278 session manager.  Also at times
  2024.   I access a VAX cluster.
  2025.  
  2026. I have redefined the PS/2 function keys to emulate the VT100/200/300 series
  2027. of terminals.
  2028.  
  2029. When I emulate a VT320 - the session messages (ie SYS AVL, etc) writes to
  2030. the top line of the screen thus wiping out the top line of the screen.
  2031. When I emulate a VT102 - the session messages are at the bottom of the screen.
  2032. This is only an annoyance and I can live with emulating the VT102.
  2033.  
  2034. The serious problem is the inability to use Kermit file transfers on either
  2035. CMS or VAX.  My computer center is aware of the CMS problem and attribute it
  2036. to the session manager.  However we cannot explain the VAX problem.  The VAX
  2037. Kermit Host (or maybe Mskermit) issues a parity error.  This is the case no
  2038. matter what parity, duplex, or flow control is used.  Any suggestionor help
  2039. is greatly apperciated.
  2040.  
  2041. [From jrd - since the pathway between the PC and either IBM/CMS or the VAX
  2042. goes through two external communications boxes one supposes that one or the
  2043. other box is adding its own parity, or more likely, is unable to handle a
  2044. stream of characters as long as a packet. Otherwise MS-DOS Kermit does run
  2045. fine to VAXen through common LAT boxes. Logging packets on both ends and
  2046. viewing the log files will show which machine is experiencing the troubles.]
  2047.  
  2048. ------------------------------
  2049.  
  2050. Date: Wed, 15 Aug 90 08:08 CDT
  2051. >From: DEC-WOLF <CC_BRYSON%SWTEXAS@cuvmb.cc.columbia.edu>
  2052. Subject: MS-Kermit and VAX ALLin1
  2053. Keywords: MS-DOS Kermit 3.01, VAX/VMS, ALLin1
  2054.  
  2055. Hello all,
  2056.  
  2057. I was wondering if there were other Ms-Kermit (3.01) users who are using PCs to
  2058. communicate with a VAX and who use ALLin1, who are also experiencing problems
  2059. with printing over the printer port.  We have determined that everything works
  2060. fine when the Vax thinks the terminal type is a VT200.  If the terminal type is
  2061. NOT VT200 series than the printing is erratic or non-existent.  After
  2062. researching the problem I dicovered that Allin1 is sending the following escape
  2063. sequences for the two terminal types:
  2064.  
  2065. VT20O   Esc [ 5 i
  2066.  
  2067. INFO-KERMIT DIGEST V12 #4                                               Page 37
  2068.  
  2069.  
  2070. VT100   Esc [ 5 i Esc [ 6 i
  2071. VT300
  2072.  
  2073. The Esc [ 5 i is the proper sequence to open the printer port for a VT100,
  2074. VT200, or VT300 terminal.  After calling Digital, I was told that (Esc [ 6 i)
  2075. was the escape sequence to open the printer port for the VT125 terminal.
  2076. Evidently the sloppy programmers at DEC check for VT200 ONLY and send the VT100
  2077. and VT125 sequences for other terminals.  It seems that either Kermit or the
  2078. printers are confused by the (Esc [ 6 i) sequence.  I know that Dec should just
  2079. straighten their act out but I was wondering if anyone knew how to work around
  2080. this problem from the Kermit end.  Could the (Esc [ 6 i) sequence be ignored or
  2081. stripped out by Kermit?  We plan on just changing the terminal type to VT200 on
  2082. the Vax end but it seems a bit kluging to me (normally the Vax and terminal
  2083. communicate the terminal type between each other at login).
  2084.  
  2085. Bill Bryson
  2086. User Services Analyst
  2087. CC_BRYSON@SWTEXAS
  2088.  
  2089. P.S.  Our systems personel could also patch allin1 to change the printer
  2090. scripts (i.e. not send ESC [ 6 i) but feel it is unlikely that they will (it is
  2091. considered a no-no to "mess" with Digital software).
  2092.  
  2093. [From jrd - MS-DOS Kermit v3.02 consumes the ESC [ 6 i and does not send it
  2094. to the printer. Yes, it's a carry over and should not be there. Here is the
  2095. entry in the file msr302.upd describing changes to version 3.02 -
  2096. 15. Correct bug in transparent printing within text emulator which corrupted
  2097.     back to back escape sequences sent to the printer.]
  2098.  
  2099. ------------------------------
  2100.  
  2101. Date: Fri, 27 Jul 90 17:47 EDT
  2102. >From: John W Manly <JWMANLY%AMHERST.BITNET@CUNYVM.CUNY.EDU>
  2103. Subject: Question about Kermit 3.01 and the 128-byte history buffer
  2104. Keywords: MS-DOS Kermit 3.01, CLEAR Command
  2105.  
  2106. Hi there.  Another question about MS-Kermit and the use of the history buffer
  2107. in Take or Script files.
  2108.  
  2109. Could anyone fill me in on the relationship between the INPUT statment, the
  2110. REINPUT statement, the CLEAR command, and the 128-byte history buffer?
  2111.  
  2112. In particular, the manual states that REINPUT only accepts additional input
  2113. from the serial port until the history buffer is full.  How does one then
  2114. clear it out?  A CLEAR command presumably does so.  Does an INPUT operation
  2115. clear it as well, or clear it only if it's full, or anything like that?
  2116.  
  2117. Any help would be greatly appreciated.  Thanks.
  2118.  
  2119.                                  - John W. Manly
  2120. BITNET:  JWMANLY@AMHERST           Systems Manager
  2121. PHONE:   (413)-542-2526            Amherst College
  2122.  
  2123. [From jrd - The CLEAR command does indeed clear things: the history buffer
  2124. and the serial port receive circular buffer (preceeds the history buffer).
  2125.  
  2126. Page 38                                               INFO-KERMIT DIGEST V12 #4
  2127.  
  2128. I agree that a separate script history buffer clearing command would be
  2129. beneficial. In the meanwhile, the history buffer can be made either larger
  2130. or smaller via the DOS command
  2131.    SET KERMIT=INPUT-BUFFER-LENGTH value
  2132. value is number of bytes (default 128)]
  2133.  
  2134. ------------------------------
  2135.  
  2136. Date: Tue, 31 Jul 90 14:44 EDT
  2137. >From: Pete Kanaitis <X979PK0P@VB.CC.CMU.EDU>
  2138. Subject: Kermit & WordPerfect under PCSA
  2139.  
  2140. Recently, we have been looking at WordPerfect 5.0 under VMS and have tried
  2141. the MS-Kermit macros and DCL command procedure (ftped from watsun) to set
  2142. the IBM PC keyboard to do WP bindings.  This seems to work fine under any
  2143. asynch connection, but does not seem to work with MS-Kermit running under
  2144. PCSA.   Under PCSA, when the terminalS and terminalR macros are envoked,
  2145. the key bindings are set, the connect command gets issued, but the original
  2146. terminal session gets disconnected!  (Since we use virtual terminals, we
  2147. can re-connect back to the WordPerfect session, and resume, but only after
  2148. we re-login in)
  2149.  
  2150. Here's an example of what is happening:
  2151.  
  2152. MSKERMIT.INI has the following:
  2153.  
  2154. set port decnet singer
  2155. set terminal wrap on
  2156. define terminalS take C:\wp30.ini, connect
  2157. set key \2344 \KTerminalS
  2158.  
  2159. Then from the PC:
  2160.  
  2161. C>kermit
  2162. IBM-PC Kermit ...
  2163.  
  2164. MS-Kermit>connect
  2165.  
  2166. Welcome to Singer...
  2167. Username: FOOBAR
  2168. Password:
  2169.  
  2170. $ @PUBLIC:PCWP
  2171. What type of Graphics adapter are you using (ega, cga, vga)?  vga
  2172.  
  2173. $ WP FILE.EXT
  2174. Defining Function Keys for WordPerfect Products....
  2175. (After a few seconds Kermit reconnects..)
  2176. (Nothing happens. When I hit any key...)
  2177.  
  2178. Welcome to Singer...
  2179. Username: FOOBAR
  2180. Password:
  2181.  
  2182.    You have the following disconnected process:
  2183.  
  2184.  
  2185. INFO-KERMIT DIGEST V12 #4                                               Page 39
  2186.  
  2187.  ...
  2188.    Connect to above listed process [YES]:
  2189.  
  2190. (Now you are back in WordPerfect)
  2191. (When you exit..)
  2192. Save Document? (Y/N) Yes
  2193. Exit WP? (Y/N) Yes
  2194.  
  2195. WP key definitions cleared...
  2196.  
  2197. (screen clears, back to normal)
  2198.  
  2199. Does anyone have any ideas?
  2200.  
  2201. Thank you...
  2202.  
  2203. Pete Kanaitis
  2204. Manager, Computational Resources
  2205. Allegheny-Singer Research Institute
  2206. X979PK0P@VB.CC.CMU.EDU (Internet)
  2207. X979PK0P@CMCCVB (Bitnet)
  2208.  
  2209. [From jrd - When WordPerfect starts up it requests a long Color Palette
  2210. report from the VT300 terminal. Kermit sends the report fine, but some
  2211. communications channels can't handle 200 some odd byte bursts. Also, the
  2212. VAX must have SET TERM/HOSTSYNC so that the VAX can XON/XOFF the terminal.
  2213. There is really nothing I can do, because on my VAX some paths just can't
  2214. cope, even when I insert massive pauses between characters. Sometimes
  2215. it's the length of the DEC Color Palette report which is the root problem.
  2216. My test copy of WordPerfect/VMS has exceeded it's 6 month lifetime, but
  2217. I recall that I could use PCSA/DECnet-DOS fine this way. The only "but" is
  2218. the VMS command SET TERM/HOSTSYNC must be stated.]
  2219.  
  2220. ------------------------------
  2221.  
  2222. Date: Wed, 18 Jul 90 14:34:55 -0400
  2223. >From: mregeste@amo.Wichita.NCR.COM
  2224. Subject: Re: MSKERMIT <-- TCP/IP --> Unix-Host
  2225. Keywords: MS-DOS Kermit, TCP/IP, Unix
  2226.  
  2227. In article <1318@kullmar.se> you write:
  2228. >In <1990Jul10.134437.18854@duc220.uni-duisburg.de>
  2229.  hz231gr@duc220.uni-duisburg.de (Gressel) writes:
  2230. >
  2231. >>Hello,
  2232. >
  2233. >>we tried it with an Interlan-NI5210-Card, no success!
  2234. >
  2235. >In Kermit News number 4, June 1990 Joe R. Doupnik and Christine M. Gianone
  2236. >writes:
  2237. >
  2238. >TCP/IP is a very widespread networking method linking machines locally and
  2239. >around the world.  ...
  2240.  
  2241. I would like to add that Wollongong's Pathway Client Plus also supports
  2242. the INT 14 service.  We use both Kermit and Procomm Plus Network Version
  2243.  
  2244. Page 40                                               INFO-KERMIT DIGEST V12 #4
  2245.  
  2246. and they work great!
  2247.  
  2248. Mark Regester Information Systems & Services, NCR Peripheral Products Division
  2249.  NCR:654-8340 <M.Regester@Wichita.NCR.COM>
  2250. (316)636-8340 <uunet!ncrlnk!ncrwic!m.regester>
  2251.  FAX:636-8889
  2252.  
  2253. [From jrd - Yeah! A WINning way.]
  2254.  
  2255. ------------------------------
  2256.  
  2257. End of Info-Kermit Digest
  2258.  
  2259. INFO-KERMIT DIGEST V12 #5                                               Page 41
  2260.  
  2261. Info-Kermit Digest         Mon, 17 Sep 1990        Volume 12 : Number 5
  2262.  
  2263. Today's Topics:
  2264.  
  2265.        Announcing IBM Mainframe VM/CMS Kermit-370 Version 4.2.1
  2266.            Selecting CONTROLLER type in Kermit-370
  2267.               Announcing  KERMIT-12 Version 10g
  2268.             Kermit Proposal SET FILE TYPE
  2269.                Prime8 Help for Dividing Source
  2270.       Re: A New Version of Kermit for OS/2 Presentation Manager
  2271.             Re: Info-Kermit Digest V12 #2
  2272.             Kermit incapatability with LSE
  2273.             Kermit 3.01 Arrow Key Problem
  2274.             132 column mode in MS-Kermit?
  2275.                   Kermit & Telebits
  2276.  
  2277. Digest submissions may be sent to Info-Kermit@WATSUN.CC.COLUMBIA.EDU,
  2278. requests for addition to or deletion from the Info-Kermit subscriber list to
  2279. Info-Kermit-Request@WATSUN.CC.COLUMBIA.EDU or to KERMIT@CUVMA.BITNET.
  2280.  
  2281. Kermit files may be obtained over networks and by mail order.  On the
  2282. Internetwork, use FTP to log in to host WATSUN.CC.COLUMBIA.EDU, a SUN-4/280
  2283. running UNIX (SUNOS 4.1), IP host number 128.59.39.2.  Login as user anonymous
  2284. (note, lower case), any password, and GET or MGET (MULTIPLE GET) the desired
  2285. files.  The Kermit files are in directories kermit/a, kermit/b, kermit/c,
  2286. kermit/d, and kermit/e.  Test versions are in kermit/test.  Binaries are in
  2287. kermit/bin (use ftp in binary mode).  You can also get Kermit files over the
  2288. BITNET/EARN network; to get started send a message with text HELP to KERMSRV,
  2289. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  2290. file kermit/a/aanetw.hlp (AANETW.HLP on KERMSRV).  To order by mail, request a
  2291. complete list of Kermit versions and an order form from Kermit Distribution,
  2292. Columbia University Center for Computing Activities, 612 West 115th Street,
  2293. New York, NY 10025 USA.
  2294.  
  2295. ----------------------------------------------------------------------
  2296.  
  2297. Date: Wed, 1990 Sep 12   18:30 EDT
  2298. >From: "John F. Chandler" <PEPMNT@CFAAMP.HARVARD.EDU>
  2299. Subject: Announcing IBM Mainframe VM/CMS Kermit-370 Version 4.2.1
  2300. Keywords: IBM 370 Kermit, VM/CMS Kermit
  2301. Xref: CMS Kermit, See VM/CMS Kermit, IBM 370 Kermit
  2302.  
  2303. This is to announce the release of Kermit-370 version 4.2.1 for CMS.  As
  2304. usual, the new version comes in VM/SP and VM/XA/SP flavors, but the
  2305. changes are the same for both.
  2306.  
  2307. Version 4.2.1 has several improvements over 4.2.0, the most important
  2308. being:
  2309.  
  2310.  1. Spurious flow-control "packets" from MS-DOS Kermit are now ignored.
  2311.  
  2312.  2. Overflow of the fullscreen buffer is now avoided when the receiving
  2313.     Kermit asks for 2K packets.
  2314.  
  2315.  3. Kermit-370 now supports transfers in LATIN2 and LATIN3 and file
  2316.     storage in CP870, CP880, and CP905.  In addition, L1, L2, and L3 are
  2317.  
  2318. Page 42                                               INFO-KERMIT DIGEST V12 #5
  2319.  
  2320.     recognized as aliases for the three LATIN sets, and two-character
  2321.     abbreviations are accepted for the other transfer sets as well.  The
  2322.     new sets add support for Afrikaans, Albanian, Catalan, Croatian,
  2323.     Czech, Esperanto, Galician, Hungarian, Maltese, Polish, Romanian,
  2324.     Slovak, Slovene, and Turkish.
  2325.  
  2326.  4. Kermit-370 supports file transfers through the IBM 3174 AEA with B2
  2327.     microcode.  The support is restricted to terminals with the ASCII
  2328.     Graphics capability in three ways:
  2329.  
  2330.     a) The terminal type must be defined in the 3174 to support graphics
  2331.        (only the built-in VT241 and Tektronix 4205 types plus suitable
  2332.        user-defined terminal types).
  2333.  
  2334.     b) The line must be defined without an associated Host Addressable
  2335.        Printer.
  2336.  
  2337.     c) If the 3174 is owned by VTAM, the connection must be made with a
  2338.        logmode that allows the Read Partition Query (such as M2SDLCNQ).
  2339.  
  2340.     Kermit-370 automatically detects the B2 AEA and sets CONTROLLER
  2341.     accordingly (to AEA if graphics is allowed, to NONE if not, or to
  2342.     GRAPHICS if Query is denied).  Since the 3174 supports full 8-bit
  2343.     communication, it may be useful to configure the ports for 8-bit
  2344.     data and to set both SEND and RECEIVE PARITY to NONE in Kermit-370.
  2345.  
  2346.  5. Kermit-370 now uses the FILE COLLISION settings for all files in a
  2347.     group rather than just the first.
  2348.  
  2349.  6. Kermit-370 has three new subcommands: REMOTE MAIL, REMOTE PRINT, and
  2350.     REMOTE SUBMIT.  They transmit a file (or group of files via wild
  2351.     cards) tagged for mailing, printing, and submitting as job,
  2352.     respectively.
  2353.  
  2354. The new release is in the form of updates to be applied to the 4.2.0
  2355. source.  The new files are IKCKER.UPD, IKCKER.BWR, IK0AAA.HLP, and
  2356. IK0KER.UPD (the latter is only a catalog of all the updates, not the
  2357. updates themselves).  The new code has been tested on most known types
  2358. of protocol converter (many thanks to the beta testers!) to make sure
  2359. the 3174 support does not harm the existing support for Kermit file
  2360. transfer, but problems may still turn up.  Bug reports are welcome, as
  2361. usual.
  2362.  
  2363. A similar release 4.2.1 will soon be available for TSO and MUSIC.
  2364.  
  2365.                                          John
  2366.  
  2367. [Ed. - Thanks, John!  The new files are installed in kermit/b on watsun
  2368. and are also available from KERMSRV@CUVMA on BITNET.  This program is truly
  2369. amazing in its adaptability to the infinitely varied 3270 emulation
  2370. communications environment, and it is a groundbreaker in character set
  2371. conversion.  Let's hope that the other Kermit programs catch up in the latter
  2372. department soon.  For that to happen, we need examples and listings of the
  2373. PC, UNIX, Macintosh, etc, character sets used for all the language supported
  2374. by Latin Alphabets 1 through 5, Latin/Cyrillic, and others.  If you have them,
  2375. please send them in and we'll do our best to support them.]
  2376.  
  2377. INFO-KERMIT DIGEST V12 #5                                               Page 43
  2378.  
  2379.  
  2380. ------------------------------
  2381.  
  2382. Date: Thu, 1990 Sep 13   11:50 EDT
  2383. >From: "John F. Chandler"   <PEPMNT@CFAAMP.HARVARD.EDU>
  2384. Subject: Selecting CONTROLLER type in Kermit-370
  2385. Keywords: IBM 370 Kermit
  2386.  
  2387. One of the improvements in release 4.2.1 is a more thorough DEBUG output
  2388. that proved very useful in bringing the support for the IBM 3174 to
  2389. completion.  This enhancement makes no difference to routine file transfer,
  2390. but it raises the possibility of determining easily the exact response of
  2391. any kind of protocol converter to the efforts of Kermit to decide which
  2392. CONTROLLER type to use.  To this end, I ask all installers of release 4.2.1
  2393. to take a few minutes and create a debug log for each (fullscreen)
  2394. environment where Kermit-370 might be used, provided, that is, that you do
  2395. not apply the optional update SC89058.  Even if you have applied SC89058 in
  2396. the past, it might be worthwhile to omit it as an experiment.  The desired
  2397. debug log is created by starting Kermit-370 and entering the subcommands SET
  2398. DEBUG I/O LONG, SET LINE, and QUIT.  I would be grateful if you would send
  2399. the resulting KER LOG along with the following information: level of VTAM
  2400. (if any) driving the session, model of front-end processor (if any) and
  2401. software, model of protocol converter (if any) and software.  These results
  2402. would be interesting even for real IBM 3270-type terminals or protocol
  2403. converters unable to support Kermit file transfers.  It is possible that the
  2404. information collected will enable Kermit to distinguish certain environments
  2405. that currently fail to trigger the correct default CONTROLLER type.
  2406.  
  2407.                                     John
  2408.  
  2409. P.S. My e-mail address is:
  2410.  
  2411.  Internet:    PEPMNT@CFAAMP.HARVARD.EDU
  2412.  BITNET:      PEPMNT@CFAAMP
  2413.  
  2414. ------------------------------
  2415.  
  2416. Date: Thu Sep 6 1990 11:00:00 EDT
  2417. >From: Charles Lasner <lasner@watsun.cc.columbia.edu>
  2418. Subject: Announcing  KERMIT-12 Version 10g
  2419. Keywords: PDP-8, PDP-12, VT-78, DECmate, OS/8
  2420. Xref: DEC PDP, See PDP
  2421.  
  2422. This is a maintenance release of KERMIT-12.  A minor problem relating to
  2423. incorrect CPU identification messages has been fixed.  The problem only
  2424. appeared when the CPU was a KK-8A single-board CPU; this configuration was
  2425. previously untested.  Thanks to Johnny Billquist of Sweden for his
  2426. assistance in pinning down the problem.
  2427.  
  2428. KERMIT-12 operation was not affected in any other way, as only the
  2429. DECmate-specific identification is crucial; earlier PDP-8 family members
  2430. are treated in a generic fashion except for the "frill" of model
  2431. identification (all PDP-8, PDP-12, VT-78 models use software-compatible
  2432. port hardware;  all DECmates are incompatible and must be handled
  2433. individually).  We are still looking for volunteers to test the various
  2434. DECmate III and DECmate III+ configurations.
  2435.  
  2436. Page 44                                               INFO-KERMIT DIGEST V12 #5
  2437.  
  2438.  
  2439. The rest of the release concerns the encoding of files into the
  2440. "ASCII-fied" format.  The format has been modified to be more robust, since
  2441. the original method has proven itself to be problematic in certain
  2442. practical circumstances (as reported in K12MIT.BWR).
  2443.  
  2444. The new ENCODing format is based on five-bit encoding with repeat
  2445. compression.  As much as 256 repeated 12-bit words will be expressed in a
  2446. five character field.  Any repeated 12-bit value can be compressed, as
  2447. opposed to simple zero compression, as in other common encoding schemes.
  2448. (PDP-8 files often have repeated strings of the value 7402 octal, which is
  2449. the HLT instruction.)
  2450.  
  2451. The only printing characters required to pass through any distribution
  2452. "path" are 0-9, A-V, X, and Z. The alphabetic characters can also be
  2453. lower-case.  All command lines are framed by ( and );  all data lines are
  2454. framed by < and >.  These characters can be changed if required, as they
  2455. are not part of the data; they could be replaced by W (w) and Y (y) if
  2456. necessary.  (Changing the framing characters requires slight modification
  2457. of the ENCODing and DECODing programs.)
  2458.  
  2459. The new format supports a 60-bit file checksum to ensure proper decoding at
  2460. the other end.  The former 12-bit checksum could be compromised on long
  2461. files.
  2462.  
  2463. The new ENCODing programs creates internal (REMARK commands stating the
  2464. ENCODed file's creation date, and the original file's creation date.  This
  2465. will aid in distribution of PDP-8 files where the user wishes to maintain
  2466. proper file dates.  The date algoritm used is the one proscribed by the
  2467. OS/8 DIRECT program. (OS/8 systems only OPTIONALLY support file dates, and
  2468. there is an eight-year "anomaly" associated with identifying the year; the
  2469. user must determine the credibility of the year portion of the date.  The
  2470. value provided by the ENCODE program for the original file creation date is
  2471. always today's year or the previous seven years as necessary; this field
  2472. will not be provided if the system doesn't support the required AIW
  2473. feature.)
  2474.  
  2475. Overall file size is theoretically as much as 6/5 of the original encoding
  2476. format (as the earlier format was based on six-bit encoding), but actual
  2477. size varies downward due to slightly less file overhead (wider lines mean
  2478. less CR LF; there is now less automatically generated verbiage), and the
  2479. random improvement afforded by simple repeat compression.
  2480.  
  2481. Virtually all K12MIT-related files are re-released at this time.  There are
  2482. several new files.  Due to the "fragile" nature of TECO macro files, the
  2483. file K12GLB.TEC is no longer being distributed directly; the file K12GLB.ENC
  2484. is the same file in the new ENCODE format.
  2485.  
  2486. The new files have been installed in the regular places:
  2487.  
  2488. BITNET/EARN       Internet
  2489. KERMSRV@CUVMA     watsun.cc.columbia.edu     Description
  2490.  
  2491.  K12MIT   ENC      kermit/d/k12mit.enc        Encoded binary of KERMIT-12
  2492.  K12MIT   DOC      kermit/d/k12mit.doc        Documentation file
  2493.  K12MIT   BWR      kermit/d/k12mit.bwr        Updated "beware" file
  2494.  
  2495. INFO-KERMIT DIGEST V12 #5                                               Page 45
  2496.  
  2497.  K12MIT   DSK      kermit/d/k12mit.dsk        Description of RX02 diskettes
  2498.  K12MIT   ANN      kermit/d/k12mit.ann        Announcement of KERMIT-12
  2499.  K12MIT   UPD      kermit/d/k12mit.upd        Release update file
  2500.  K12DEC   PAL      kermit/d/k12dec.pal        Decoding program
  2501.  K12ENC   PAL      kermit/d/k12enc.pal        Encoding program
  2502.  K12PL8   ENC      kermit/d/k12pl8.enc        Encoded binary of PAL8 Ver B0
  2503.  K12CRF   ENC      kermit/d/k12crf.enc        Encoded binary of CREF Ver B0
  2504.  K12MIT   PAL      kermit/d/k12mit.pal        Main source file of KERMIT-12
  2505.  K12PCH   PAL      kermit/d/k12pch.pal        KERMIT-12 source patch file
  2506.  K12CLR   PAL      kermit/d/k12clr.pal        Memory clearing file
  2507.  K12MIT   LST      kermit/d/k12mit.lst        Symbols-only listing file
  2508.  K12PRM   PAL      kermit/d/k12prm.pal        Sample VT-78 config file
  2509.  K12GLB   ENC      kermit/d/k12glb.enc        Encoded TECO file macro
  2510.  K12ENC   DOC      kermit/d/k12enc.doc        Encoding format description
  2511.  
  2512. [Ed. - Many thanks, Charles.  Believe it or not, there are still quite a
  2513. few PDP-8 based systems out there, and even some PDP-12s.  You won't find
  2514. very many other new software packages that support them!]
  2515.  
  2516. ------------------------------
  2517.  
  2518. Date: Thu, 6 Sep 90 08:47 CST
  2519. >From: <PK6811S@DRAKE.BITNET>
  2520. Subject: Kermit Proposal SET FILE TYPE
  2521. Keywords: Kermit Protocol
  2522.  
  2523. Please accept these remarks regarding the proposed SET FILE TYPE extension
  2524. to Kermit.
  2525.  
  2526. 1.  The labels are transmitted in normal Data packets.  Therefore an
  2527. ignorant Kermit receiver will write them into the file as data.  This is a
  2528. MAJOR change to the protocol, which has always provided transparent file
  2529. exchange.  When reading such a file, how is a program to know that labels
  2530. have already been included?  Will these files cause problems for other
  2531. file-transfer protocols such as Xmodem?
  2532.  
  2533. [Ed. - No, it won't cause problems with Xmodem.  Xmodem or any other file
  2534. transfer protocol, including a Kermit program that does not know about labeled
  2535. files, will store or send a labeled file just like any other file, without
  2536. interpreting the label information.]
  2537.  
  2538. 2.  Stored-in-File labels are incompatible with the MacBinary protocol in
  2539.     standard use on Macintosh computers.
  2540.  
  2541. [Ed. - That's true.  But MacBinary format cannot be adapted to other kinds
  2542. of file systems such as VAX/VMS.  Kermit can still be used to transfer
  2543. MacBinary files, and a MacBinary option will be added to Macintosh Kermit
  2544. before its next release.]
  2545.  
  2546. 3.  The sending operator must now decide whether to send with labels or not.
  2547.     Previously, the sender had no choices to make.  We should not be adding
  2548.     complexity to the process if we can help it.
  2549.  
  2550. [Ed. - This is true.  However, without labels it is impossible to transfer a
  2551. complex file and retain all of its features.  Therefore this proposal
  2552. addresses only computers with complex file systems, like VAX/VMS, IBM
  2553.  
  2554. Page 46                                               INFO-KERMIT DIGEST V12 #5
  2555.  
  2556. mainframes, etc, and does not affect other computers.  Users can continue to
  2557. use Kermit on these systems in the normal way, but now they will also have
  2558. an additional tool to let them successfully transfer files that they could
  2559. not transfer before.]
  2560.  
  2561. 4.  Perhaps a better approach would be to use:
  2562.  
  2563.         SET FILE TYPE system type
  2564.  
  2565.     Where 'system' identifies the receiving file management system,
  2566.     (RMS, VSAM, etc.) and 'type' identifies the appropriate attributes.
  2567.     Example:
  2568.  
  2569.         SET FILE TYPE RMS INDEXED
  2570.  
  2571.     where RMS represents Vax RMS file management, and INDEXED identifies
  2572.     the incoming file as an indexed file.
  2573.  
  2574.     This approach requires that the sending and receiving routines be
  2575.     similarly intelligent about the information required to successfully
  2576.     transmit and write the file, but allows unintelligent intermediary
  2577.     programs (repositories, bulletin boards, etc.) to receive and send
  2578.     the files without any special settings (other than SET FILE TYPE BINARY).
  2579.  
  2580. [Ed. - It is a cardinal principle of Kermit or any other well-designed file
  2581. transfer protocol that any particular computer must not be required to have
  2582. knowledge about the formats and conventions of any other kind of computer.
  2583. Rather, each Kermit implementation knows only the data formats of its own
  2584. computer and those defined for the "standard intermediate representation" on
  2585. the wire.  Otherwise, hundreds of different programs would require
  2586. modifications to know about hundreds of different computers.  Not only is
  2587. this a waste of computing and human resources, but it's a moving target.]
  2588.  
  2589.     It also allows the following sequence to take place:
  2590.  
  2591.       machine 1 Kermit send  -> machine 2 Kermit receive (FILE TYPE BINARY)
  2592.       machine 2 BrandX send  -> machine 3 BrandX receive
  2593.       machine 3 Kermit send  -> machine 4 Kermit receive (FILE TYPE xxx ttt)
  2594.  
  2595.     Machine 1 sends the file BELIEVING that the receiver knows how to handle
  2596.     it, but only the machine 4 program needs to know.  However, machine 3
  2597.     can also use the file, if it recognizes, or is told to recognize, the
  2598.     file type.
  2599.  
  2600. [Ed. - The Labeled File proposal works the same way.]
  2601.  
  2602.     A simplification of this approach would be to use SET FILE TYPE AUTO
  2603.     which would cause the receiver to perform some recognition process
  2604.     before writing the file.
  2605.  
  2606. [Ed. - The labeled file proposal includes this possibility too.]
  2607.  
  2608. We have recently developed a Macintosh VT100 terminal emulator which
  2609. incorporates Kermit file exchange, and automatically distinguishes TEXT
  2610. documents from applications and formatted documents.  The users need only
  2611. tell the Vax to SET FILE TYPE ASCII/BINARY depending on whether they intend
  2612.  
  2613. INFO-KERMIT DIGEST V12 #5                                               Page 47
  2614.  
  2615. to edit it on the Vax, or transfer it to another Mac.  The program pre-pends
  2616. the Macbinary header for non-text documents when sending, and recognizes it
  2617. when receiving.  I think you'll agree that the LABELED proposal only adds
  2618. complexity to our situation, and requires users to make unnecessary
  2619. decisions about the process.
  2620.  
  2621. [Ed. - Your application seems to be Macintosh-centered, in the sense that
  2622. you regard the VAX as a repository for Macintosh applications, which you
  2623. encode in MacBinary, but you make no provision for storing VAX applications
  2624. or complex binary VAX files on the Mac.  For your purposes, MacBinary will
  2625. do just fine.]
  2626.  
  2627. ------------------------------
  2628.  
  2629. Date: Tue, 28 Aug 90 17:27:31 EDT
  2630. >From: John M. Crawford <CRAW4D@prime.cob.ohio-state.edu>
  2631. Subject: Prime8 Help for Dividing Source
  2632. Keywords: Prime Kermit
  2633.  
  2634. Enclosed is a short file which provides the necessary Primos editor (ED)
  2635. tricks to divide (and then build) Kermit 8.12 for Primos. You might consider
  2636. appending it to the PRIME8.ANN file (or another) for general distribution to
  2637. Prime kermit users.
  2638.  
  2639. John M. Crawford      (614) 292-1741                Computing Services Center
  2640.                                                     College of Business
  2641. craw4d+@osu.edu                                     1775 College Road
  2642. craw4d@prime.cob.ohio-state.edu                     The Ohio State University
  2643. crawford-j@osu-20.ircc.ohio-state.edu               Columbus, Ohio 43210
  2644.  
  2645. [Ed. - Many thanks!  Your ED file is now enshrined in PRIME8.ANN.]
  2646.  
  2647. ------------------------------
  2648.  
  2649. Date: Wed, 29 Aug 90 10:23:11 PLT
  2650. >From: Wim Bonner <27313853%WSUVM1@cuvmb.cc.columbia.edu>
  2651. Subject: Re: A New Version of Kermit for OS/2 Presentation Manager
  2652. Keywords: OS/2 Kermit
  2653.  
  2654. The message said that the Kermit files were in the OS/2 test area pending
  2655. comments from the user community.  Is this where I make comments?
  2656.  
  2657. I tried the Kermit program just now.  It has an annoying habit of going full
  2658. screen.  I have a 1024x768 screen, so an 80,25 screen will fit in less than
  2659. a quarter of the screen, and leave room for much more on the screen.
  2660.  
  2661. I am running OS/2 1.2, and was not able to get the lights on my modem to
  2662. blink when I typed characters after Connecting.  That is normally a good
  2663. indication something is wrong.  All of the other communication programs that
  2664. I've tried work fine.
  2665.  
  2666. Wim Bonner - 27313853@WSUVM1.CSC.WSU.EDU - V(509)335-4436
  2667.  
  2668. [Ed. - Thanks.  We'll collect all such comments and put them in an envelope
  2669. and send them back to the contributor.  Meanwhile, has anyone else gotten
  2670. the program to work?]
  2671.  
  2672. Page 48                                               INFO-KERMIT DIGEST V12 #5
  2673.  
  2674.  
  2675. ------------------------------
  2676.  
  2677. Date: Sat 25 Aug 90 11:51:37-CDT
  2678. >From: Rob Pettengill <SW.PETTENGILL@MCC.COM>
  2679. Subject: Re: Info-Kermit Digest V12 #2
  2680. Keywords: MS-DOS Kermit 3.02, Macros, Variables, Scripts
  2681.  
  2682. The behavior of the macro positional parameters has changed in the new test
  2683. release of MSKermit 3.02.  Previously when a take script was taken in a
  2684. macro the \%n positional arguments were defined in the take script - with
  2685. this version they are undefined.
  2686.  
  2687. The previous behavior was reasonable for a script.  Are you tring to make
  2688. take files behave more like macros?  If so then it should be possible to
  2689. pass positional arguments to the take files explicitly.  In any case the
  2690. behavior in the current 3.02 does not seem desirable.
  2691.  
  2692. ;rob
  2693.  
  2694. [Ed. - That was a mistake, which has been corrected in the latest 3.02
  2695. release.  Macro parameters are now on a call stack so if macro A calls macro
  2696. B, A's parameters are still intact when B returns.  The mistake was indeed
  2697. that a TAKE file was being treated like a macro.  In the latest edit, it is
  2698. not: if macro A TAKEs a command file, A's parameters are available to the
  2699. TAKE file, as before.]
  2700.  
  2701. ------------------------------
  2702.  
  2703. Date: 29 Jul 90 23:18:00 CDT
  2704. >From: "COLLINS, STERRETT" <z3col@ttacs1.ttu.edu>
  2705. Subject: Kermit incapatability with LSE
  2706. Keywords: MS-DOS Kermit 3.0, LSE, VAX/VMS, Terminal Emulation
  2707.  
  2708. I am using MS-Kermit v3.0 on a Kaypro 286 operating MS-DOS v3.21, connecting
  2709. to a VaxCluster operating VMS v5.3 .  I have tried using the language
  2710. sensitive editor (LSE v3.0), but Kermit fails to emulate the terminal that
  2711. LSE expects.  I have tried setting terminal/nodec_crt3, which partially
  2712. corrects the problem, but still not satisfactorily.  I am not sure that LSE
  2713. does not think that every member of the VT300_Family of terminals is a Regis
  2714. terminal.  I have access to a Regis_Emulating communications software
  2715. package, which I do not normally use for licensing considerations, but
  2716. which, if I switch to that after having entered LSE using Kermit, is able to
  2717. emulate the terminal properly.
  2718.  
  2719. The possible answers would seem to be:
  2720.  
  2721. "The operating system does not successfully detect the correct terminal
  2722. characteristics."
  2723.  
  2724. "The Operating system does successfully detect the correct terminal
  2725. characteristics, but this information is not propperly transferred to LSE."
  2726. I only know that SHOW TERMINAL reveals that it does not have the REGIS
  2727. characteristic, but does have the SIXEL charactersitic, and the DECCRT_3
  2728. characteristic.
  2729.  
  2730.  
  2731. INFO-KERMIT DIGEST V12 #5                                               Page 49
  2732.  
  2733.                                         sterrett collins
  2734.                                         physics department
  2735.                                         texas tech university
  2736.  
  2737. [From jrd - Another case where Kermit command SET DISPLAY 8 needs be done
  2738. to avoid clobbering the 8-bit control sequences sent by the host.]
  2739.  
  2740. ------------------------------
  2741.  
  2742. Date: Mon, 23 Jul 90 19:44:42 GMT
  2743. >From: seminara@penelope.oswego.edu (Greg Seminara)
  2744. Subject: Kermit 3.01 Arrow Key Problem
  2745. Keywords: MS-DOS Kermit, Arrow Keys
  2746.  
  2747. Arrow keys work only sporadically when connected to a UNIX host from a DOS
  2748. PC running Kermit 3.01 if you are using a full-screen "curses" application.
  2749. Seems to be sending either the wrong ansi sequence or is sending the
  2750. sequence too fast.  If you press arrow keys with a significant pause between
  2751. each press, keys usually work OK.  Emulation is vt102 for kermit and TERM is
  2752. set to "kermit" or "vt100".  Kermit 2.32a and before work fine under
  2753. identical conditions.  - HELP!
  2754.  
  2755. [From jrd - your Unix host has a problem with communications.  If characters
  2756. in a control sequence are NOT sent in rapid succession then EMACS reacts
  2757. differently than expected.  In addition, some host machines apparently have
  2758. intrinsic difficulties running in true full duplex.  Unix loves to echo
  2759. arriving characters willy nilly.  So when Unix is sending a control sequence
  2760. while one is arriving from Kermit and Unix is also echoing it then Kermit
  2761. receives both sets of information with characters intermingled.  If the
  2762. sequences arrived separately then Kermit could cope.  This is pretty silly
  2763. behavior by Unix; the application should attempt supressing echos.  There is
  2764. nothing that Kermit (nor a real terminal) can do about it.]
  2765.  
  2766. ------------------------------
  2767.  
  2768. Date: Thu, 30 Aug 90 14:28:57 CDT
  2769. >From: moore@ncsc.navy.mil (Moore)
  2770. Subject: 132 column mode in MS-Kermit?
  2771. Keywords: MS-DOS Kermit, 132 Columns, Screen Settings
  2772.  
  2773. Greetings.
  2774.  
  2775. I have a very persistent user here who wants the word directly from the
  2776. horse's mouth regarding a feature of MS-Kermit:
  2777.  
  2778. I've always thought that EGAs cannot do 132-column mode (except by simulating
  2779. it in graphics mode).  MS-Kermit doesn't change that, right?  Isn't it the
  2780. "responsibility" of the hardware to switch modes, and then Kermit just detects
  2781. and uses that mode?
  2782.  
  2783. [Ed. - True.  Except that in version 3.0 and later it also supports automatic
  2784. switching between 80 and 132 column mode via the COLS80.BAT and COLS132.BAT
  2785. mechanism.  That is, if Kermit gets the escape sequence telling it to switch
  2786. modes, it will try to run the appropriate BAT file that does the PC- or
  2787. adapter-specific things required to switch modes.]
  2788.  
  2789.  
  2790. Page 50                                               INFO-KERMIT DIGEST V12 #5
  2791.  
  2792. Can Kermit simulate 132-column mode by letting the user pan an 80-column
  2793. window left and right?
  2794.  
  2795. [Ed. - No.]
  2796.  
  2797. Thanks for any help.
  2798.  
  2799. Jim
  2800. moore@NCSC.navy.mil
  2801.  
  2802. ------------------------------
  2803.  
  2804. Date: 27 Aug 90 11:42:05 GMT
  2805. >From: mrsvr!saturn.shaw@uwm.edu (Tom Shaw ct58 Ex 5084)
  2806. Subject: Kermit & Telebits
  2807. Keywords: Modems
  2808.  
  2809. I'm looking for anyone who has/is doing something similar to this: I am
  2810. transferring binary and ASCII files across dial-up phone lines using Telebit
  2811. Trailblazer Plus modems and kermit.  The transfers are between 2 Suns, one
  2812. running Sun OS 3.5 and the other running Sun OS 4.01.  The size of the files
  2813. range from 2500 bytes to 800kbytes.  Does anyone have any tips, traps to
  2814. avoid, benchmarks of what kind of rate I should be expecting for ASCII and
  2815. binary files or any hands on advice.  I am using kermit version 4E(72).
  2816.  
  2817. In the past few weeks, I've heard stories about using Telebits in Germany,
  2818. what kind of problem is there and are there any other countries I might have
  2819. a problem connecting and transfering to?
  2820.  
  2821. Any help would be appreciated.  Thanks
  2822.  
  2823. Tom
  2824.  
  2825. [Ed. - As you may know, Telebits have Kermit built inside them.  If you
  2826. activate this feature, Computer A actually talks Kermit protocol to Modem A,
  2827. Modem A talks high-speed error-correcting PEP protocol to Modem B and Modem B
  2828. talks Kermit protocol to Computer B.  This is all done transparently to the
  2829. Kermit programs, but you have to put the originating Telebit into "Kermit
  2830. spoof" mode.  See your Telebit manual.]
  2831.  
  2832. ------------------------------
  2833.  
  2834. End of Info-Kermit Digest
  2835.  
  2836. INFO-KERMIT DIGEST V12 #6                                               Page 51
  2837.  
  2838. Info-Kermit Digest         Wed, 10 Oct 1990        Volume 12 : Number 6
  2839.  
  2840. Today's Topics:
  2841.  
  2842.                New Honeywell DPS-6 Superkermit
  2843.              Announcing Kermit for the Tandy 100
  2844.                CMS Kermit 4.2.1 erratum
  2845.                OS/2 PM Kermit Feedback
  2846.                OS/2 PM Kermit Feedback
  2847.               Easy Exit from Terminal Emulation
  2848.           Running MS-DOS Kermit 3.0x under DESQview
  2849.            USING MS-DOS Kermit 3.0 with MNP Modems
  2850.               MS-Kermit 4010 GIN mode by Mouse?
  2851.               Suppresing Kermit's Error Checking
  2852.               MacKermit Settings Trouble
  2853.              Larger Font for Mac Kermit 0.98(63)?
  2854.               Prime Kermit Version 8.12
  2855.               Kermit REMOTE SET Command
  2856.  
  2857. Digest submissions may be sent to Info-Kermit@WATSUN.CC.COLUMBIA.EDU,
  2858. requests for addition to or deletion from the Info-Kermit subscriber list to
  2859. Info-Kermit-Request@WATSUN.CC.COLUMBIA.EDU or to KERMIT@CUVMA.BITNET.
  2860.  
  2861. Kermit files may be obtained over networks and by mail order.  On the
  2862. Internetwork, use FTP to log in to host WATSUN.CC.COLUMBIA.EDU, a SUN-4/280
  2863. running UNIX (SUNOS 4.1), IP host number 128.59.39.2.  Login as user anonymous
  2864. (note, lower case), any password, and GET or MGET (MULTIPLE GET) the desired
  2865. files.  The Kermit files are in directories kermit/a, kermit/b, kermit/c,
  2866. kermit/d, and kermit/e.  Test versions are in kermit/test.  Binaries are in
  2867. kermit/bin (use ftp in binary mode).  You can also get Kermit files over the
  2868. BITNET/EARN network; to get started send a message with text HELP to KERMSRV,
  2869. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  2870. file kermit/a/aanetw.hlp (AANETW.HLP on KERMSRV).  To order by mail, request a
  2871. complete list of Kermit versions and an order form from Kermit Distribution,
  2872. Columbia University Center for Computing Activities, 612 West 115th Street,
  2873. New York, NY 10025 USA.
  2874.  
  2875. ----------------------------------------------------------------------
  2876.  
  2877. Date: Tue, 2 Oct 90 11:42:02 PDT
  2878. >From: dreano@trout.nosc.mil (Frank Dreano)
  2879. Subject: New Honeywell DPS-6 Superkermit
  2880. Keywords: DPS-6 Kermit, Honeywell Kermit
  2881.  
  2882. I am sending you a new version of the Honeywell DPS-6 Superkermit.  This
  2883. version basically 'blows out' the server mode to provide much more
  2884. functionality, (After Joe R. Doupnik's excellent effort to support the
  2885. Honeywell DPS-6 in MS-DOS Kermit 3.02 dev. 28, I felt like I had to respond
  2886. in kind):
  2887.  
  2888. 1) Superkermit server 'gets' and 'sends' now support wildcard filespecs.
  2889.  
  2890. 2) MS-DOS Kermit 'remote cwd' commands are now supported.
  2891.  
  2892. 3) MS-DOS Kermit 'remote set file type { text, binary }' commands are now
  2893.    supported.
  2894.  
  2895. Page 52                                               INFO-KERMIT DIGEST V12 #6
  2896.  
  2897.  
  2898. 4) A new Superkermit file type, called 'foreign', supports MSDOS *.EXE,
  2899.    *.COM, *.WKS, *.ZIP, etc. file transfers WITHOUT converting to ASCII.
  2900.    MS-DOS Kermit 'remote kermit foreign' sets Superkermit for these
  2901.    transfers.  UNIX executables, etc, are likewise supported by this new
  2902.    file type.  The 'foreign' type also works with the VIP-3 terminal
  2903.    emulator.
  2904.  
  2905. 5) MS-DOS Kermit 'remote type [filename]' commands are now supported.
  2906.  
  2907. 6) MS-DOS Kermit 'remote host [host command]' commands are now supported.
  2908.  
  2909. 7) MS-DOS Kermit 'remote dir [filespec]' commands are now supported.
  2910.  
  2911. 8) MS-DOS Kermit 'remote delete [filespec]' commands are now supported.
  2912.  
  2913. 9) MS-DOS Kermit 'remote mesage [text]' commands are now supported.
  2914.  
  2915. 10) MS-DOS Kermit 'remote who' commands are now supported.
  2916.  
  2917. NOTE:  There ARE some warnings about using these commands...  See the
  2918.        HD6KER.BWR 'beware' file for details.
  2919.  
  2920. I received a request to allow the Superkermit server to select either
  2921. 8th-bit or image mode binary transfers.  If I didn't implement this ability
  2922. the default would always be 8th-bit (not satisfactory in all cases).  So I
  2923. have implemented two additional commands:
  2924.  
  2925.                 'remote kermit image'
  2926.                 'remote kermit prefix'
  2927.  
  2928.    Performed a much-needed clean up of code placing all compiler pre-
  2929.    processor code along with global/external variables and function
  2930.    prototypes in a separate file.  In addition, I have adopted your
  2931.    naming conventions for the source files:
  2932.  
  2933.             HD6KER.DOC -  Version 2.00 documentation.
  2934.             HD6KER.H   -  New include file described above.
  2935.             HD6KER.C   -  Kermit state switchers and heuristics.
  2936.             HD6PRI.C   -  Kermit protocol primitives.
  2937.             HD6COMP.EC -  New Kermit batch compilation JCL.
  2938.  
  2939. I believe the above improvements have resulted in a much more useful
  2940. Superkermit 'server' mode as well as more manageable source code.  My thanks
  2941. to Tim Ewing of Honeywell Federal Systems Incorporated (HFSI) in McLean,
  2942. Virginia for suggesting and testing some of the above changes.  I feel that
  2943. the code reorganization and additional server functionality have made this a
  2944. major release and I am calling it Version 2.00.
  2945.  
  2946. Frank Dreano Jr.
  2947.  
  2948. [Ed. - Many thanks, Frank!  The new files have been installed in the "D"
  2949. area of Kermit distribution under the names you sent them with
  2950. (kermit/d/hd6*.* on watsun).  Special thanks for adjusting the filenames.]
  2951.  
  2952. ------------------------------
  2953.  
  2954. INFO-KERMIT DIGEST V12 #6                                               Page 53
  2955.  
  2956.  
  2957. Date: Tue Oct  9 21:29:27 1990
  2958. >From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
  2959. Subject: Announcing Kermit for the Tandy 100
  2960. Keywords: Tandy 100
  2961.  
  2962. Apparently floating around since 1984, discovered recently and sent in by
  2963. Jay Rouman: a Kermit program, written in BASIC by Dave Straayer, for the
  2964. Tandy 100.  No documentation.  In kermit/c/ta100.bas on watsun, TA100 BAS
  2965. on CUVMA.  Thanks, Jay!
  2966.  
  2967. ------------------------------
  2968.  
  2969. Date: Mon, 1990 Sep 17   20:04 EDT
  2970. >From: "John F. Chandler"   <PEPMNT@CFAAMP.HARVARD.EDU>
  2971. Subject: CMS Kermit 4.2.1 erratum
  2972. Keywords: Kermit-370, CMS Kermit, IBM Mainframe
  2973.  
  2974. The announcement mentioned restrictions on VTAM logmodes for successful file
  2975. transfer through the IBM 3174, but the example cited was one that does *not*
  2976. work.  The logmode that is known to work is M2SDLCQ.
  2977.  
  2978. [Ed. - A corrected copy of IKCKER.ANN (the announcement text for CMS Kermit
  2979. 4.2.1) has been installed.]
  2980.  
  2981. ------------------------------
  2982.  
  2983. Date: Tue, 18 Sep 90 11:20:12 -0700
  2984. >From: agb@cs.washington.edu (Alan Bishop)
  2985. Subject: OS/2 PM Kermit Feedback
  2986. Keywords: OS/2 Kermit, Modems
  2987.  
  2988. In article <CMM.0.88.653601846.cmg@watsun.cc.columbia.edu> Wim Bonner writes:
  2989. >
  2990. > I am running OS/2 1.2, and was not able to get the lights on my modem to
  2991. > blink when I typed characters after Connecting.  That is normally a good
  2992. > indication something is wrong.  All of the other communication programs
  2993. > that I've tried work fine.
  2994. >
  2995. You need to play with the octs & so forth settings.  Look under the
  2996. communcations setup bit & toggle all the settings on the line that has OCTS
  2997. listed.  It works fine for me.
  2998.  
  2999. ------------------------------
  3000.  
  3001. Date: Tue, 18 Sep 90 14:23:25 EDT
  3002. >From: "Robert E. Zaret" <ZARET%MITVMA@cuvmb.cc.columbia.edu>
  3003. Subject: OS/2 PM Kermit Feedback
  3004. Keywords: OS/2 Kermit
  3005.  
  3006. I did get the Presentation Manager version of Kermit working, but haven't
  3007. spent much time with it.  I already have the full screen OS/2 Kermit and my
  3008. monitor is strict VGA, so I have no urge for a PM version.  Thus, I wasn't
  3009. going to comment.  However, the digest I received today asked for comments,
  3010. so:
  3011.  
  3012.  
  3013. Page 54                                               INFO-KERMIT DIGEST V12 #6
  3014.  
  3015. My major dissappointment is lack of VT100 support.  I assume its absence is
  3016. temporary.  I've seen enough mangled emulators to believe VT100 emulation is
  3017. not trivial, so I'ld support a decision to leave it out until other parts
  3018. are settled.  However, I have little use for a program without it.
  3019.  
  3020. I was able to communicate with my internal modem; the modem responded to a
  3021. query (ATS1?) and to dial commands.
  3022.  
  3023. I found no way to save settings.
  3024. The speed setting is labelled baud, rather than bps.
  3025. I would prefer one pull down menu for all communications settings
  3026.     (com port, bps, parity, etc.)
  3027. I didn't find a documentation file, and the program itself mentioned
  3028.     no way to get out of communication mode except by exiting the
  3029.     program completely.  I tried ctl-[, and it worked.  This sequence
  3030.     seems inconsistent with the DOS version, but I'm not sure
  3031.     consistency matters, or is even possible.
  3032. I, too, found the program insisting on using a full screen.  The
  3033.     behavior of the maximimize box (upper right corner) seems odd,
  3034.     but I'm not sure what it should do given the other problem.
  3035. I believe PM programs can find and use the user's preferences for colors,
  3036.     as set in the control box.  Definitely not a high priority.
  3037.  
  3038. Technical note: I'm using IBM OS/2 EE 1.2 on an IBM Model 80/311 with
  3039. an IBM 8513 monitor and an internal BusinessLand modem (not quite
  3040. all blue :-).
  3041.  
  3042. ------------------------------
  3043.  
  3044. Date: Wed, 18 Apr 90 11:49:40 -0700
  3045. >From: Tshapin@orion.oac.uci.edu
  3046. Subject: Easy Exit from Terminal Emulation
  3047. Keywords: Terminal Emulation, MS-DOS Kermit, Macros
  3048.  
  3049. Here is an easy way to leave terminal emulation, hangup the phone, and drop
  3050. to DOS (I put it on alt-Z):
  3051.  
  3052.         def terminalr hangup,exit
  3053.         set key \2348 \kterminalr
  3054.  
  3055. (Thanks to Joe D. for pointing this out.)
  3056.  
  3057. Here is why I couldn't display IBM graphic characters even though I had my
  3058. display set for 8-bit and terminal character set transparent.
  3059.  
  3060. Ted.
  3061.  
  3062. [From jrd - Alas, VT100's are really 7-bit devices and this time I have
  3063. enforced the rule, to make character set stuff behave as DEC intended. Try
  3064. SET TERMINAL VT320.]
  3065.  
  3066. ------------------------------
  3067.  
  3068. Date: Sun, 16 Sep 90 16:19 MDT
  3069. >From: Joe Doupnik <JRD@cc.usu.edu>
  3070. Subject: Running MS-DOS Kermit 3.0x under DESQview
  3071.  
  3072. INFO-KERMIT DIGEST V12 #6                                               Page 55
  3073.  
  3074. Keywords: MS-DOS Kermit, DESQview
  3075.  
  3076. Assistance in creating a DESQview .DVP configuration file (DVSETUP.TXT)
  3077. for MS-DOS Kermit.
  3078.  
  3079. My copy of DESQview is version 2.26 running on a 386 machine with 2MB, with
  3080. no expanded memory, but with QEMM/386. The display adapter is a Video-7
  3081. "VGA".  The tables below are similar to the DESQview screens.
  3082.  
  3083. Please, when configuring QEMM to load-high drivers etc always exclude video
  3084. memory.
  3085.  
  3086. I have been able to run two copies of MS-DOS Kermit v3.02 together, to
  3087. different destinations with different communications ports. In addition,
  3088. full graphics capability was retained.
  3089.                                                 Joe D.
  3090.  ============================================================================
  3091.                         Change a Program
  3092.  Program Name............: MS-DOS Kermit v3.02
  3093.  Keys to Use on Open Menu: KE                   Memory Size (in K): 200
  3094.                                                 (say 150 minimum)
  3095.  ----------------------------------------------------------------------------
  3096.  Program...: Kermit.exe
  3097.  Parameters:            (Kermit command line items, if any, go here.)
  3098.  Directory.: C:\yourdir
  3099.  ----------------------------------------------------------------------------
  3100.  Options:
  3101.                 Writes text directly to screen.......: [N]      *
  3102.                 Displays graphics information........: [N]      **
  3103.                 Virtualize text/graphics (Y,N,T).....: [T]      ***
  3104.                 Uses serial ports (Y,N,1,2)..........: [Y]
  3105.                 Requires floppy diskette.............: [N]
  3106.  
  3107.    *   Not true, but MS-DOS Kermit is DESQview-aware and covers up nicely.
  3108.    **  Not true, naturally, but causes DESQview to put Kermit into full
  3109.        screen mode while graphics are executed and then to return to windowed
  3110.        mode after switching to text terminal emulation.
  3111.    *** Can be [Y] with the same effect as above.
  3112.  ----------------------------------------------------------------------------
  3113.                         Change a Program Advanced Options
  3114.  
  3115.  System Memory (in K).....:     0  Maximum Program Memory Size (in K)..:
  3116.  Script Buffer Size.......:     0  Maximum Expanded Memory Size (in K) :
  3117.  Text Pages: 1  Graphics pages: 0  Initial Mode:  3    Interrupts: 00 to FF
  3118.  
  3119.  Window Position:
  3120.         Maximum Height:  25     Starting Height:  20    Starting Row...:  5
  3121.         Maximum Width.:  80     Starting Width.:  40    Starting Column:  5
  3122.   (Choose your own window sizes, naturally. 132 column mode causes DESQview
  3123.    to switch to full screen.)
  3124.  ----------------------------------------------------------------------------
  3125.                                 Shared Program
  3126.  Pathname..:
  3127.  Data......:
  3128.  ----------------------------------------------------------------------------
  3129.  Close on exit (Y,N,blank)......: [ ]   Uses its own colors..............: [N]
  3130.  
  3131. Page 56                                               INFO-KERMIT DIGEST V12 #6
  3132.  
  3133.  Allow Close Window command.....: [Y]   Runs in background (Y,N,blank)...: [Y]
  3134.  Uses math coprocessor..........: [N]   Keyboard conflict (0-F)..........: [0]
  3135.  Share CPU when foreground......: [Y]   Share EGA when foreground/zoomed.: [Y]
  3136.  Can be swapped out (Y,N,blank).: [ ]   Protection level (0-3)...........: [3]
  3137.  ============================================================================
  3138.  
  3139. ------------------------------
  3140.  
  3141. Date: 9 Sep 90 06:27:49 GMT
  3142. >From: ts@uwasa.fi (Timo Salmi LASK)
  3143. Subject: Using MS-DOS Kermit 3.0 with MNP Modems
  3144. Keywords: MS-DOS Kermit, Scripts, MNP Modems
  3145.  
  3146. In article <1990Sep09.012634.22430@ddsw1.MCS.COM> andyross@ddsw1.MCS.COM
  3147. (Andrew Rossmann) writes:
  3148.  
  3149. > I just bought a 2400 baud MNP5 modem (ATI 2400etc/e.) I have quickly
  3150. > found out that some modems do not like the attempt to set up an MNP
  3151. > protocol. I am currently using Procomm Plus. The problem is that there is
  3152. > no way for me to have a particular entry in the dialer set the modem's MNP
  3153. > on or off (a linked script file is only called AFTER connection.) Are there
  3154. > any other communications programs out there that let you send special
  3155. > initializations to the modem BEFORE dialing?  Is there a sneaky way to do
  3156. > this in Procomm Plus??
  3157. >
  3158. Yes, I've thought about the same problem, and MS-DOS Kermit has this
  3159. feature.  With proper script programming you can have MS-DOS Kermit do
  3160. whatever initializations do wish before dialing, and then on top of that
  3161. doing the dialing from a multi-entry dialing directory.  This MS-DOS Kermit
  3162. dialing directory facility is available in the /pc/ts/tskerm24.arc MS-DOS
  3163. Kermit utilities collection downloadable by anonymous ftp from
  3164. chyde.uwasa.fi, Vaasa, Finland.  The flexibility and potential of MS-DOS
  3165. Kermit as a fully fledged communication program rather than just an
  3166. excellent terminal emulator are often underrated and little known.
  3167.  
  3168. ....................................................................
  3169. Prof. Timo Salmi        (Moderating at anon. ftp site 128.214.12.3)
  3170. School of Business Studies, University of Vaasa, SF-65101, Finland
  3171. Internet: ts@chyde.uwasa.fi Funet: gado::salmi Bitnet: salmi@finfun
  3172.  
  3173. ------------------------------
  3174.  
  3175. Date: Thu, 20 Sep 90 17:41 MDT
  3176. >From: Pete Klammer 303/556-3915 <PKLAMMER@cudnvr.denver.colorado.edu>
  3177. Subject: MS-Kermit 4010 GIN mode by Mouse?
  3178. Keywords: MS-DOS Kermit, Terminal Emulation, GIN mode, Graphics, Mouse
  3179.  
  3180. We find that the 4010 emulation in MS-KERMIT supports GIN mode, that is,
  3181. graphical cursor position input, which is great!  For instance, the ZOOM
  3182. command in ANSYS puts up a crosshair target on our screen, which we can
  3183. position with the cursor keys.  Pretty neat.
  3184.  
  3185. Is it possible to get our PC mouse connected to this cursor?  The GIN mode
  3186. steps with the cursor keys are pretty small, hence slow.  It would be even
  3187. neater than the already neat that it is, if MS-KERMIT Tektronix emulation
  3188. would support a PC mouse for GIN mode.  Thanks!
  3189.  
  3190. INFO-KERMIT DIGEST V12 #6                                               Page 57
  3191.  
  3192.  
  3193. /** --poko ________________________________* "cut on dashes for mailing label"
  3194. Pete Klammer (303)556-3915 FAX(303)556-4822|pklammer@cudnvr.Denver.Colorado.EDU
  3195. CU-Denver Computing Services Campus Box 169|bitnet: PKLAMMER@CUDENVER
  3196. 1200 Larimer St NC2506 Denver CO 80217-3364|uucp:...!boulder!pikes!pklammer **/
  3197.  
  3198. [Ed. - MS-DOS Kermit does indeed support mouse movement of the graphics
  3199. cursor.  Make sure you have given the necessary DOS commands to turn the
  3200. mouse on, such as MOUSE ON and/or DEFAULT.]
  3201.  
  3202. ------------------------------
  3203.  
  3204. Date: Mon, 24 Sep 90 12:01:37 CDT
  3205. >From: david@wubios.wustl.edu (David J. Camp)
  3206. Subject: Suppresing Kermit's Error Checking
  3207. Keywords: MS-DOS Kermit, Error Checking, Kermit Protocol, Performance
  3208. Keywords: MNP Modems
  3209.  
  3210. I just read the December inquiry about suppressing Kermit's error checking
  3211. to save time when using an MNP reliable connection.  You could do a raw
  3212. download.  For instance, if you are on a host that will let you simply
  3213. 'type', 'cat', or 'fstype' a file, you can tell Ms-Kermit to capture the
  3214. screen and let it flow.  The MS-Kermit command is 'log session filename'.
  3215. Make sure that you have 'set flow on' and that the host will recognize the
  3216. flow control to gate the transfer.  This will not work for binary files.
  3217.  
  3218. I do not recommend this.  The reliability of the link goes beyond the
  3219. mode-to-modem portion of the communication.  There are several things you
  3220. can do to improve the speed of Kermit transfers.  The first is to increase
  3221. your packet size.  Recent versions of Kermit allow packets near 1000 bytes
  3222. in length.
  3223.  
  3224. [Ed. - Kermit's theoretical maximum packet size is 9024.  MS-DOS Kermit 3.0
  3225. supports packets up to 2000 bytes.  IBM mainframe Kermit supports packets up
  3226. to about 1700.  The forthcoming release of C-Kermit 5A (watch Info-Kermit for
  3227. an announcement) can be configured to support the full 9K.]
  3228.  
  3229. I usually limit myself to 500, because MNP modems doing baud rate
  3230. translation fail with larger packets in non-MNP mode.  The speed of transfer
  3231. with 500-byte packets is very nearly that with 1000-byte packets.
  3232.  
  3233. I did some benchmarks to compare the speed in MNP stream mode vs. MNP block
  3234. mode.  My conclusion is that the difference is small, and you are better off
  3235. using stream mode to make interaction less jumpy.  A much bigger payoff is
  3236. gained by using long packets.
  3237.  
  3238. Next, use sliding windows if they are available.  I have done some
  3239. benchmarks to determine the optimum number of windows, and it is 2.
  3240.  
  3241. [Ed. - Actually, the optimum window size varies according to the round-trip
  3242. delay of the connection and its noise characteristics.  See Kermit News #4.
  3243. For a clean and relatively direct connection with MNP modems, 2 appears to
  3244. be a good size.  Also bear in mind that, in general, the large the window
  3245. size, the smaller the maximum packet size, because most Kermit programs have
  3246. a finite total packet buffer size.]
  3247.  
  3248.  
  3249. Page 58                                               INFO-KERMIT DIGEST V12 #6
  3250.  
  3251. The last thing I can recommend is to use 8-bit communications if possible.
  3252. It may be necessary to do more than just 'set parity none', but I am not
  3253. sure of this.  If you are downloading from an IBM mainframe, this will
  3254. probably not work.  -David-
  3255.  
  3256. [Ed. - SET PARITY NONE is usually the default, but only works if the
  3257. connection is truly 8 bits wide.  Most IBM mainframes do not give you an
  3258. 8-bit path.  Neither to telnet connections, nor most public data network
  3259. connections.]
  3260.  
  3261. david@wubios.wustl.edu                 ^        Mr. David J. Camp
  3262. david%wubios@wugate.wustl.edu        < * >      +1 314 382 0584
  3263. ...!uunet!wugate!wubios!david          v        "Be kind to sinners."
  3264. "Concentration impedes Inspiration."  ---  "Depend on God, who has dominion."
  3265.  
  3266. ------------------------------
  3267.  
  3268. Date: Thu, 19 Jul 90 10:34:05 MST
  3269. >From: Tom Aurand <tom@longs.LANCE.ColoState.Edu>
  3270. Subject: MacKermit Settings Trouble
  3271. Keywords: MacKermit 0.98
  3272.  
  3273. I'm having trouble getting XON/XOFF flow control to stick in the settings
  3274. file when using MacKermit 0.98 on different SE's.  Can I patch it?  Thanks
  3275. for your help.
  3276.  
  3277. [From Paul Placeway - Get a copy of 0.98(63) -- the difference between edit
  3278. 62 and 63 was a fix for that and saving of parity.  BTW: DON'T USE MACBINARY
  3279. MODE -- sending a file in MacBinary mode will delete the file on the Mac (yes,
  3280. that's a bug).  I suggest you disable the MacBinary button with ResEdit.]
  3281.  
  3282. [Ed. - 0.98(63) is available in kermit/test/ckmker.hqx.]
  3283.  
  3284. ------------------------------
  3285.  
  3286. Date:  Mon, 23 Jul 90 10:36:59 -0500
  3287. >From: harms@emx.utexas.edu (Robert T. Harms)
  3288. Subject: Larger Font for Mac Kermit 0.98(63)?
  3289. Keywords: MacKermit 0.98(63), Fonts
  3290.  
  3291. I would like to use a larger font with my Mac Kermit on a Mac IIci - related
  3292. to general eye problems of mine.  I have used ResEdit to replace the VT100
  3293. built in font with a Courier 12 font.  BUT this obviously isn't enough.  (1)
  3294. The WIND resource seems to permit increasing the width of the window, but
  3295. somewhere the dimensions of the window need to be coded as well, and (2) the
  3296. cursor position is erratic with respect to character width.
  3297.  
  3298. Is there some way I can customize the font size using ResEdit?  Or would
  3299. you advise me to leave well enough alone?
  3300.  
  3301. Bob Harms       harms@emx.utexas.edu
  3302.  
  3303. [From Paul Placeway - Unfortnatly, MacKermit 0.98 has the font size
  3304. parameters hard coded into it, so you can't change the font even if you find
  3305. all the right places to patch.
  3306.  
  3307.  
  3308. INFO-KERMIT DIGEST V12 #6                                               Page 59
  3309.  
  3310. This will be fixed in the next release (I have that part of the code now
  3311. working, but other parts don't yet).  (I'm getting older and blinder too...
  3312. I like to run with a 10pt font (usually Courier 10)).]
  3313.  
  3314. ------------------------------
  3315.  
  3316. Date: 04 Sep 90 18:02:39 O
  3317. >From: WAGNER@atlas.smc.univie.ac.at
  3318. Subject: Prime Kermit Version 8.12
  3319.  
  3320. Yesterday I transferred a copy of PRIME Kermit Version 8.12, and new
  3321. fresh copies of Kermit Version 3.0.1 for IBM PCs. Both sources compiled
  3322. and run flawless except:
  3323.  
  3324. 1. Generic commands issued from PC to PRIME Kermit in server mode usually
  3325. respond with unimplemented command (e.g. remote who, remote dir, remote
  3326. delete, etc.) But they are implemented in Prime KERMIT!  There seems to be a
  3327. definition problem of generic commands either on the sender (PC) or upon
  3328. reception (PRIME Kermit).
  3329.  
  3330. [Ed. - This is because vanilla 3.01 sends improperly formed generic
  3331. commands.  The patch file MSR301.PCH fixes this problem and, of course, it
  3332. will be fixed in 3.02.]
  3333.  
  3334. 2. The only generic command which led to an action was remote type file
  3335. (file being a text file).  However, PRIME Kermit sends garbage between every
  3336. character from the text file !
  3337.  
  3338. [Ed. - Hmmm...     Could this be because PRIME uses "negative ascii"?]
  3339.  
  3340. 3. We run PRIMOS 22.1.2 on a PRIME CPU.  At Rev 22 major changes affected
  3341. AMLC I/O (more assignable lines then ever, especially lines over Ethernet
  3342. cables, with line numbers larger than 128! Same applies to user numbers, we
  3343. have numbers up to 996 users! That means that several old interfaces to
  3344. PRIMOS, e.g. calls to DUPLX$, ASNLN$ etc have new replacements, which MUST
  3345. be used to change communication properties when either a user number or a
  3346. line number is larger than 256!  This seems to be a serious problem to all
  3347. users of PRIME Kermit 8.12 using Rev 22 and up, on larger machines, since
  3348. some of the new interfaces did not officially exist on older Prime
  3349. Revisions, and I expect that they are users which are forced to use either a
  3350. newer revision (e.g.  all TCP/IP users) or an older revision (for example
  3351. older hardware, running into performance problems when changing to rev 22).
  3352. Coding of those routines must be revision dependent, to give all Prime users
  3353. a perfect program.
  3354.  
  3355. Marcus C. Wagner
  3356. Department of Statistics and Computer Science
  3357. Universitaetsstrasse 5/9
  3358. A-1010 Vienna, Austria, Europe
  3359. University of Vienna
  3360.  
  3361. Email: since our TCP/IP connection is in experimental operation (since
  3362. two months, with some severe TCP/IP implementation errors of PRIME),
  3363. it will happen that my email address via Internet fails (host seems to be
  3364. down for the outer world) Alternatively, I am reachable via EARN/BITNET.
  3365. wagner@atlas.smc.univie.ac.at (IP)
  3366.  
  3367. Page 60                                               INFO-KERMIT DIGEST V12 #6
  3368.  
  3369. A4421DAC@AWIUNI11  (EARN/BITNET)
  3370.  
  3371. Thanks to all programmers, who invested so much time to redesign the
  3372. old version of Prime Kermit, which we implemented five years ago, several
  3373. known bugs of that old version during normal file transfers vanished in
  3374. the new program, not to mention the performance gain.
  3375.  
  3376. [Ed. - Thanks for the comments.  They have been sent to the PRIME Kermit
  3377. developers.]
  3378.  
  3379. ------------------------------
  3380.  
  3381. Date: Fri, 28 Sep 90 20:00:00 PDT
  3382. >From: Bruce_A._Cowan@cc.sfu.ca
  3383. Subject: Kermit REMOTE SET Command
  3384. Keywords: Kermit Protocol, Remote Set, Character Sets
  3385.  
  3386. I'd like to suggest a small extension to the Kermit remote set command.  As
  3387. you have it defined, the REMOTE SET TRANSFER CHARACTER-SET allows setting any
  3388. transfer alphabet that has a designator.  However, there is also the option
  3389. called TRANSPARENT or NONE, i.e. suppress translation completely.  Since that
  3390. option does not have a designator, one cannot set it.  I would like to have
  3391. some kind of extension to allow setting TRANSPARENT.  Perhaps we could invent
  3392. a designator, say, 0 (zero), or perhaps some other method to guarantee that
  3393. the TRANSPARENT setting does not conflict with any possible future designator.
  3394. We could perhaps invent a new registration, kermit (K), and have TRANSPARENT
  3395. be K0.
  3396.  
  3397. I came up with this problem while implementing translation in my kermit, but
  3398. for time reasons the remote set isn't going to make it into my next release,
  3399. so I'm not in a gigantic rush for this, but I'd like to resolve it.
  3400.  
  3401. [Ed. - This is actually a tricky situation.  The idea was that TRANSPARENT
  3402. should be the default, for compatibility with the hundreds of Kermit
  3403. programs that do not support the character-set extension.  Adding a new
  3404. designator to denote TRANSPARENT might confuse Kermits that don't understand
  3405. it, including Kermits that do support the character-set extension, if that
  3406. designator were to arrive in an Attribute packet.  Incoming files bearing
  3407. the new designator might be rejected because of "unknown character set" --
  3408. clearly undesirable behavior, avoidable perhaps by having the user SET
  3409. UNKNOWN CHARACTER-SET KEEP, but how to explain this to mere mortals?  So
  3410. far, it has been an implementation decision about whether character-set
  3411. changes based on incoming attribute packets are "sticky".  If they are, this
  3412. presents the problem that there is no way for a client to tell a server to
  3413. switch from translation to transparent.  So it would appear that a decent
  3414. compromise would be to invent a new designator, say K0, for TRANSPARENT, and
  3415. to support it in REMOTE SET TRANSFER CHARACTER-SET commands, but, at least
  3416. for a time, not use it in Attribute packets.  Opinions?]
  3417.  
  3418. By the way, the last info I have on remote set is dated 11 Dec 89.  Was there
  3419. anything later?
  3420.  
  3421. [Ed. - Three additional parameters were added in Vol 11 #4: REMOTE SET FILE
  3422. { BLOCKSIZE, RECORD-LENGTH, RECORD-FORMAT }.  The details for RECORD-FORMAT
  3423. have not yet been worked out.]
  3424.  
  3425.  
  3426. INFO-KERMIT DIGEST V12 #6                                               Page 61
  3427.  
  3428. The 11 Dec info has a minor problem in that file collision discard and ask
  3429. are both 4; I presume ask should be 5, to fit the pattern.
  3430.  
  3431. [Ed. - Right, ASK should be 5.  Thanks.]
  3432.  
  3433. Bruce Cowan
  3434. KEA Systems Ltd.
  3435.  
  3436. ------------------------------
  3437.  
  3438. End of Info-Kermit Digest
  3439.  
  3440. Page 62                                               INFO-KERMIT DIGEST V12 #7
  3441.  
  3442. Info-Kermit Digest         Tue, 20 Nov 1990        Volume 12 : Number 7
  3443.  
  3444. Today's Topics:
  3445.  
  3446.       Announcing IBM Mainframe MVS/TSO Kermit-370 Version 4.2.1
  3447.            Announcing CP/M Kermit Version 4.10 for Testing
  3448.          New Test Release of Microsoft Windows Kermit
  3449.          Another New Prerelease of MS-DOS Kermit 3.02
  3450.            Rainbow Kermit Updated Too (for testing)
  3451.               New Kermit for Gould/SEL 32/77 MPX
  3452.           MS-Kermit 3.02 Argument Handling, Keyboard Mapping
  3453.             Re: Kermit REMOTE SET Command
  3454.                Kermit for STRATUS/VOS?
  3455.                  Kermit vs Hypercard?
  3456.  
  3457. Digest submissions may be sent to Info-Kermit@WATSUN.CC.COLUMBIA.EDU,
  3458. requests for addition to or deletion from the Info-Kermit subscriber list to
  3459. Info-Kermit-Request@WATSUN.CC.COLUMBIA.EDU or to KERMIT@CUVMA.BITNET.
  3460.  
  3461. Kermit files may be obtained over networks and by mail order.  On the
  3462. Internetwork, use FTP to log in to host WATSUN.CC.COLUMBIA.EDU, a SUN-4/280
  3463. running UNIX (SUNOS 4.1), IP host number 128.59.39.2.  Login as user anonymous
  3464. (note, lower case), any password, and GET or MGET (MULTIPLE GET) the desired
  3465. files.  The Kermit files are in directories kermit/a, kermit/b, kermit/c,
  3466. kermit/d, and kermit/e.  Test versions are in kermit/test.  Binaries are in
  3467. kermit/bin (use ftp in binary mode).  You can also get Kermit files over the
  3468. BITNET/EARN network; to get started send a message with text HELP to KERMSRV,
  3469. the Kermit file server, at host CUVMA.  For detailed instructions, read the
  3470. file kermit/a/aanetw.hlp (AANETW.HLP on KERMSRV).  To order by mail, request a
  3471. complete list of Kermit versions and an order form from Kermit Distribution,
  3472. Columbia University Center for Computing Activities, 612 West 115th Street,
  3473. New York, NY 10025 USA.
  3474.  
  3475. ----------------------------------------------------------------------
  3476.  
  3477. Date: Thu Nov 15 11:45:38 1990
  3478. >From: "John F. Chandler"  <PEPMNT@CFAAMP.HARVARD.EDU>
  3479. Subject: Announcing IBM Mainframe MVS/TSO Kermit-370 Version 4.2.1
  3480. Keywords: IBM 370 Kermit, MVS/TSO Kermit
  3481. Xref: TSO Kermit, See MVS/TSO Kermit, IBM 370 Kermit
  3482.  
  3483. Kermit-370 version 4.2.1 for TSO has now been released.  For the most
  3484. part, this release matches the version 4.2.1 for CMS announced in
  3485. September, but there are some system-specific differences.
  3486.  
  3487. Version 4.2.1 has several improvements over 4.2.0, the most important
  3488. being:
  3489.  
  3490.  1. Kermit-370 supports file transfers through the IBM 3174 AEA with B2
  3491.     microcode (or higher).  The support is restricted to terminals with
  3492.     the ASCII Graphics capability in three ways:
  3493.  
  3494.     a) The terminal type must be defined in the 3174 to support graphics
  3495.        (only the built-in VT241 and Tektronix 4205 types plus suitable
  3496.        user-defined terminal types).
  3497.  
  3498.  
  3499. INFO-KERMIT DIGEST V12 #7                                               Page 63
  3500.  
  3501.     b) The line must be defined without an associated Host Addressable
  3502.        Printer.
  3503.  
  3504.     c) If the 3174 is owned by VTAM, the connection must be made with a
  3505.        logmode that allows the Read Partition Query (such as M2SDLCQ).
  3506.  
  3507.     Kermit-370 automatically detects the B2 AEA and sets CONTROLLER
  3508.     accordingly (to AEA if graphics is allowed, to NONE if not, or to
  3509.     GRAPHICS if Query is denied).  Since the 3174 supports full 8-bit
  3510.     communication, it may be useful to configure the ports for 8-bit
  3511.     data and to set both SEND and RECEIVE PARITY to NONE in Kermit-370.
  3512.  
  3513.  2. Overflow of the fullscreen buffer is now avoided when the receiving
  3514.     Kermit asks for 2K packets.
  3515.  
  3516.  3. Kermit-370 now supports transfers in LATIN2 and LATIN3 and file
  3517.     storage in CP870, CP880, and CP905.  In addition, L1, L2, and L3 are
  3518.     recognized as aliases for the three LATIN sets, and two-character
  3519.     abbreviations are accepted for the other transfer sets as well.  The
  3520.     new sets add support for Afrikaans, Albanian, Catalan, Croatian,
  3521.     Czech, Esperanto, Galician, Hungarian, Maltese, Polish, Romanian,
  3522.     Slovak, Slovene, and Turkish.
  3523.  
  3524.  4. Kermit-370 has three new subcommands: REMOTE MAIL, REMOTE PRINT, and
  3525.     REMOTE SUBMIT.  They transmit a file (or group of files via wild
  3526.     cards) tagged for mailing, printing, and submitting as job,
  3527.     respectively.
  3528.  
  3529.  5. Spurious flow-control "packets" from MS-Kermit are now ignored.
  3530.  
  3531.  6. When downloading ISPF PDS members, Kermit-TSO now uses the ISPF
  3532.     modification date as the time stamp (if attributes are enabled).
  3533.  
  3534.  7. Kermit-370 now uses the FILE COLLISION settings for all files in a
  3535.     group rather than just the first.
  3536.  
  3537.  8. The format of KER.LOG for SET DEBUG I/O has been extended to give
  3538.     hexadecimal buffer dumps like those of CMS Kermit.
  3539.  
  3540. The new release is in the form of updates to be applied to the 4.2.0
  3541. source.  The new files are IKTKER.UPD and IKTKER.BWR.  The new code has
  3542. been tested on both IBM 7171's (demonstrating downward compatibility)
  3543. and IBM 3174's (many thanks to the beta testers!), and the CMS version
  3544. had already been tested on other types of protocol converters, but
  3545. problems may still turn up.  Bug reports are welcome, as usual.
  3546.  
  3547. A similar release 4.2.1 will soon be available for MUSIC.  Also, watch
  3548. for release 4.2.1 for CICS.
  3549.  
  3550. [Ed. - Thanks, John!  The new files are in in the B area of Kermit
  3551. distribution: iktker.ann (this message), iktker.bwr (the beware file), and
  3552. iktker.upd (the updates to be applied to the program to produce the new
  3553. version; see iktker.ins for instructions.]
  3554.  
  3555. ------------------------------
  3556.  
  3557.  
  3558. Page 64                                               INFO-KERMIT DIGEST V12 #7
  3559.  
  3560. Date: Thu, 1 Nov 90 14:18:55 EST
  3561. >From: Mike Freeman <freeman@watsun.cc.columbia.edu>
  3562. Subject: Announcing CP/M Kermit Version 4.10 for Testing
  3563. Keywords: CP/M-80 Kermit
  3564.  
  3565. Just to prove that the CP/M world still exists, here is CP/M Kermit-80
  3566. Version 4.10.
  3567.  
  3568. New features include:
  3569.  
  3570. o SET COLLISION {BACKUP/DISCARD/RENAME/REPLACE}
  3571. o SET INCOMPLETE-FILES {DISCARD/KEEP}
  3572. o Many REMOTE commands, including some REMOTE SET commands
  3573. o RENAME command to rename CP/M files from within Kermit-80
  3574. o SET RECEIVE/SEND PACKET-LENGTH nn (nn <= 94)
  3575. o Many bug fixes
  3576. o Kermit-80 Version 4.10 now supports the Microbee family of
  3577.    computers (56K, 64K, 128K and 256K) manufactured by
  3578.    Microbee Systems, Ltd, of Australia.
  3579.  
  3580. Technical changes:
  3581.  
  3582. o An "I" packet is now tried before the request for files is sent
  3583.   in a GET command
  3584. o The overlay address is now 7000H
  3585. o The overlay (CPXLNK.ASM) has an added INCFLG variable to handle
  3586.   INCOMPLETE-FILES status
  3587.  
  3588. Syntax hints:
  3589.  
  3590. In REMOTE commands such as REMOTE COPY, REMOTE MESSAGE, REMOTE RENAME,
  3591. REMOTE WHO etc., where two arguments are required, the syntax is as for Vax
  3592. VMS Kermit, i.e.,
  3593.  
  3594.   REMOTE <command> <argument 1>
  3595.   Prompt:  <argument 2>
  3596.  
  3597. For example:
  3598.  
  3599.   REMOTE COPY <old-file>
  3600.   New file: <new-file>
  3601.  
  3602. The REMOTE LOGIN command has three arguments, of which the last two are
  3603. prompted for an entered on new lines.  Arguments are optional.
  3604.  
  3605. In commands such as REMOTE SET FILE BLOCK-SIZE which take a numeric
  3606. argument, Kermit-80 does not check the argument for validity.  Whatever is
  3607. typed is what the host sees.
  3608.  
  3609. The FCOPY command has been renamed COPY.
  3610.  
  3611. The syntax for the RENAME command is as for the COPY command, that is:
  3612. RENAME <old-filespec> <new-filespec>
  3613.  
  3614. More rigorous checking for wildcards is now done by both of these
  3615. commands.
  3616.  
  3617. INFO-KERMIT DIGEST V12 #7                                               Page 65
  3618.  
  3619.  
  3620. SET BAUD-RATE is now SET SPEED.
  3621.  
  3622. The CONNECT command no longer can be abbreviated to "C"; the minimum
  3623. acceptable abbreviation is now "CON".
  3624.  
  3625. The option SET FILE DEFAULT has ben re-enabled.  Use a TAKE-file to set the
  3626. initial file-mode to ASCII or BINARY.
  3627.  
  3628. Building Kermit-80 ver. 4.10:
  3629.  
  3630. Kermit-80 ver. 4.10 is built using the same procedure as for version 4.09.
  3631.  
  3632. Many thanks to those who have contributed bug fixes, and, in particular,
  3633. to Mr. Russell Lang of Monash University, Australia, for contributing
  3634. a bug-fix and the family file for the Microbee systems.
  3635.  
  3636. [Ed. - Many thanks, Mike!  We hauled out our original Kermit Superbrain from
  3637. nine years ago, downloaded the system-independent hex file and the
  3638. "Brain"-dependent hex file, combined them with MLOAD, and the new version
  3639. worked perfectly.  But because CP/M Kermit runs on so many different
  3640. computers, we are installing it in the Kermit test area for now.  If you
  3641. encounter any problems with it, send e-mail to Mike at the address above.
  3642. After a decent testing interval, it will replace version 4.09 in the main
  3643. distribution area.  The Kermit test area is kermit/test on watsun, and T: on
  3644. CUVMA.  All CP/M-80 Kermit filenames start with the letters CP.  There are
  3645. more than a hundred files, so first get the file CPAAAA.HLP, which lists and
  3646. describes the files, read it, and then request the files you need.  An
  3647. updated version of MLOAD is also included as MLOAD.*.  An updated manual
  3648. will be available shortly.  Thanks again to Mike for all the work that went
  3649. into this release.]
  3650.  
  3651. ------------------------------
  3652.  
  3653. Date: Fri, 19 Oct 90 18:10:57 PDT
  3654. >From: william@xlnvax.novell.com (William Hall)
  3655. Subject: New Test Release of Microsoft Windows Kermit
  3656. Keywords: Microsoft Windows Kermit
  3657.  
  3658. WIN100.EXE, a version of Kermit with VT100 and VT52 terminal emulation for
  3659. Microsoft Windows, is available on an experimental basis from the Kermit
  3660. distribution center.  Programs are available for both Windows 2.x and Windows
  3661. 3.0 and replace the previously released WN100K.EXE.  Except for bug fixes, the
  3662. 2.x version will not receive further enhancements.  The 3.0 version will
  3663. undergo several improvements over the next several months as time permits
  3664. including the release of all sources when legally possible.  Plans include
  3665. basic server support as well as extensions of the emulation to include VT220
  3666. and H19 as well as 8 bit character sets.  4010/4014 support and Telapi support
  3667. are also planned for the future.  Accompanying the programs is a font file
  3668. suitable for use on VGA displays which allow for proper display of bold
  3669. characters and DEC special graphics.
  3670.  
  3671. [Ed. - Many, thanks, Bill!  The new files are in ~kermit/test/win*.* on
  3672. watsun for Internet anonymous ftp access, and T:WIN*.* on CUVMA for BITNET
  3673. KERMSRV access.]
  3674.  
  3675.  
  3676. Page 66                                               INFO-KERMIT DIGEST V12 #7
  3677.  
  3678. ------------------------------
  3679.  
  3680. Date: Thu Nov 15 11:47:07 1990
  3681. >From: Christine M Gianone <cmg@watsun.cc.columbia.edu>
  3682. Subject: Another New Prerelease of MS-DOS Kermit 3.02
  3683. Keywords: MS-DOS Kermit 3.02
  3684.  
  3685. >From Joe Doupnik's change log:
  3686.  
  3687. 71. Expand the command line internal variable value substitution syntax
  3688.     \v(variable) to encompass internal variables:
  3689.         ARGC        PATH
  3690.         COUNT        TIME
  3691.         DATE        VERSION
  3692.         ERRORLEVEL
  3693.     Allow "?" help and <ESC> word completion on these items as well.
  3694.  
  3695.     Now scripts can include additional controls such as
  3696.         IF equal \v(date) 09-16-1990 echo it's \v(time) on Sunday
  3697.     Notice that \v(variable) yields a text string and cannot be combined
  3698.     with the strictly numerical IF test conditions of <, =, and >. For
  3699.     numerical tests employ the form
  3700.         IF > time 12:00 echo the time is past noon
  3701.  
  3702.     Some basic concepts are as follows.
  3703.     - Intrinsic (built-in) commands do not affect the \%0..\%9 variables,
  3704.       but user created commands (Macros) do create a private version of
  3705.       \%0..\%9.
  3706.  
  3707.     - To be a command the keyword, intrinsic or macro, must be the first word
  3708.       at the Kermit prompt. Commas in Macros start a fresh command line even
  3709.       though the Kermit prompt is usually not displayed on the screen while
  3710.       within a macro.
  3711.  
  3712.       Thus in the definition below of macro test
  3713.         DEFINE test echo \%0 \%1,mywords,echo test mywords \%1
  3714.       there are three commands (they start at the effective Kermit prompt):
  3715.         echo TEST <first argument><carriage return>
  3716.     mywords  becomes  whatever macro mywords is defined as<carriage ret>
  3717.     echo test mywords <first argument to test><carriage return>
  3718.             The argument to this last echo cmd is the simple text
  3719.             "test mywords" plus the first argument to test.
  3720.             In the text neither test nor mywords is replaced by
  3721.             its definition, and echo is an intrinsic command.
  3722.  
  3723.     - String substitution variables, \%<character> and \v(variable), are
  3724.       always immediately replaced by their definition string no matter where
  3725.       they are used outside of a DEFINE command.
  3726.  
  3727.     - ASSIGN is similar to DEFINE except that whereas DEFINE uses the
  3728.       definition text as a literal string ASSIGN first evaluates the text,
  3729.       i.e. it replaces \%<character> and \v(variable) items with their string,
  3730.       at the time the ASSIGN statement is executed.
  3731.  
  3732. 72. More work on Tektronix BYPASS mode. By popular demand, ensure that BYPASS
  3733.     mode stays on (discards incoming characters) until a control character
  3734.  
  3735. INFO-KERMIT DIGEST V12 #7                                               Page 67
  3736.  
  3737.     arrives from the host, or the screen is cleared, or the emulator is reset.
  3738.     Previous test editions of version 3.02 waited for a local control code to
  3739.     be generated. Files affected are msgibm.asm and msyibm.asm.
  3740.  
  3741. 73. Correct tiny bug introduced by item 70. File mssker.asm
  3742.     And plug a security hole of two Control-C's in a row from a remote client
  3743.     causing server mode to exit if FIN/BYE is disabled. Files msscom.asm and
  3744.     mssser.asm.
  3745.  
  3746. 74. Modify the action of the BYE/FINISH/LOGOUT command from an MS-DOS Kermit
  3747.     client. Previously these would complete successfully whether or not the
  3748.     server host sent an Error packet. It has been discovered that C Kermit
  3749.     version 5A sends an Error packet response if BYE has been Disabled. To
  3750.     keep things rational MS-DOS Kermit now stays active if an Error packet
  3751.     is received for these commands. Script writers may want to issue an
  3752.     extra command to EXIT Kermit regardless. An MS-DOS Kermit server returns
  3753.     an ACK with a message that the command is disabled, it logs out the user,
  3754.     it hangs up the connection, and it returns to being a server waiting for
  3755.     another customer, all if BYE is disabled.
  3756.  
  3757. 75. Adjust the FILE TRANSFER CHARACTER-SET tables for LATIN5/Cyrillic to
  3758.     use ISO 8859-5 (Latin/Cyrillic) on the wire and presume Code Page 866
  3759.     (Alternate Cyrillic) in the machine. Modify the character set name
  3760.     from LATIN5 to CYRILLIC. Cyrillic uses an ISO 2022 announcer of ESC L.
  3761.     File mssfil.asm is changed.
  3762.  
  3763. 76. Add to list of \v(variable) items the variables PLATFORM and SYSTEM.
  3764.     Platform returns the machine name, such as IBM-PC, and System returns
  3765.     the operating system name, MS-DOS, for MS-DOS Kermit. This is to match
  3766.     C Kermit 5A.
  3767.     In addition, add the objects PLATFORM and SYSTEM to the WRITE command;
  3768.     they may be followed by optional text:
  3769.         WRITE SESSION Platform and this is a good machine.
  3770.     FIles changed are msscmd.asm and msster.asm
  3771.  
  3772. 77. Correct a sneaky bug regarding testing for DESQview which led to over
  3773.     writing of a few Help messages in the main body of MS-DOS Kermit/IBM-PC.
  3774.     File msyibm.asm.
  3775.  
  3776. 78. Add numerical variable KEYBOARD to IF statements, \v(variable), and
  3777.     WRITE statements so that Take files can be tailored to fit either 88
  3778.     (so-called Old AT) or 101 (so-called Enhanced) keyboards. The value is
  3779.     either 88 (default) or 101 or 250. The determination of 88 or 101 is made
  3780.     at Kermit startup. The IBM PC version also permits the value 250 if the
  3781.     LK250 keyboard is selected (SET KEY LK250) and external driver is present
  3782.     for the Digital keyboard. Files affected are msuibm, msscmd, mssscp,
  3783.     and msster.
  3784.  
  3785. 79. Enhancements to SET PORT EBIOS for IBM LANACS.
  3786.     - Let the baud rate be set for BIOSn and EBIOS ports for IBM-PC version.
  3787.     Initially they will show as "unknown" until set by the user. In addition
  3788.     to the speed 8 data bits and no parity are set. For EBIOS, the port
  3789.     being adjusted is really that on the asynchronous server; the local PC
  3790.     uses NetBios and EBIOS to simulate that remote port across the network.
  3791.  
  3792.     - Add new optional final command argument of the EBIOS server port name
  3793.  
  3794. Page 68                                               INFO-KERMIT DIGEST V12 #7
  3795.  
  3796.         SET PORT EBIOS <local coms port, 1..4> <server port name>
  3797.     such as  SET PORT EBIOS 1 modem-2400
  3798.  
  3799.     If the server port name is omitted then the port name is taken from the
  3800.     EBIOS table on the client machine, such as resuming an old connection.
  3801.     This option removes the necessity of using the REDIRECT.EXE program, part
  3802.     of the LANACS package, because Kermit does a simple redirection of the
  3803.     selected local port to the named host port (and the name is persumed to
  3804.     be a unique one rather than a group name). If another type of connection
  3805.     is desired then (a) omit the sever port name in the SET PORT EBIOS cmd
  3806.     and (b) perform the redirection with REDIRECT.EXE. Each instance of SET
  3807.     PORT EBIOS <local comms port> uses its own port information of speed,
  3808.     parity, flow control, handshake, and local-echo.  msxibm.asm, msster.asm
  3809.  
  3810. 80. On the formatted file transfer screen add " of n" to the Window slots
  3811.     line, where n is the maximum number negotiated between Kermits. msssho.asm
  3812.  
  3813. 81. Correct bug in processing CSI ? 256 n leading to crashes. Thanks to
  3814.     Risto Laitinen. File msgibm.asm
  3815.  
  3816. 82. Modify the strategy for DEC Autoprint, CSI ? 5 (and 4) i  from printing
  3817.     each character as it arrives to printing display lines only as the cursor
  3818.     is moved off it by an autowrap or LF, FF, or VT (otherwise do not print
  3819.     the line). Thanks to Michael Wolter for pointing out the problem. mszibm
  3820.  
  3821. 83. Also, preserve the state of VTxxx terminal emulator printing around
  3822.     processing of DEC Set Conformance Level commands, e.g. CSI 63; 1 " p.
  3823.     Plus, correct a small bug resetting the cursor column to zero when writing
  3824.     to a host owned status line. Tnx to Michael Wolter and John E. Davis.
  3825.  
  3826. 84. 12 Oct, two small fixes - ensure REM FILE TYPE remembers the command while
  3827.     being an MS-DOS Kermit server (mssser.asm), and cure a small \v(variable)
  3828.     bug in file msscmd.asm.
  3829.  
  3830. 85. Correct a problem of not sending an XON following a failed RUN or PUSH
  3831.     command. An XOFF sent automatically as Kermit shells to DOS. The XOFF/XON
  3832.     pair is used to suspend host output while DOS is in charge. mssker.asm
  3833.  
  3834. 86. 25 Oct. Fix some parser problems (Control-C handling, ASK command line
  3835.     editing) introduced with internal changes made over the past three weeks.
  3836.     Files changed are msscmd.asm and mssset.asm.
  3837.  
  3838. 87. Modify the action of pressing the Enter key while a script INPUT command
  3839.     is in progress. Previously it meant "assume a match has occurred" so the
  3840.     script could progress. Now it means "assume a timeout has occurred" and
  3841.     report a failure status. People seem to prefer the latter approach. mssscp
  3842.  
  3843. 88. 2 Nov. Add internal support of Microsoft compatible mouse for Tektronix
  3844.     GIN crosshairs. Both the regular cursor keys and the mouse can function
  3845.     together. Pressing either mouse button is the same as pressing the Enter
  3846.     key. The mouse driver must be active for Kermit to use it (include line
  3847.     DEVICE=path\MOUSE.SYS in file CONFIG.SYS for permanent residency, or give
  3848.     the DOS command MOUSE ON for temporary residency of the MS mouse driver).
  3849.     This addition bypasses any active mouse menu programs such as DEFAULT.COM
  3850.     during Tektronix emulation but they continue to run otherwise. MS-DOS
  3851.     Kermit uses standard Microsoft Interrupt 33h mouse calls. msgibm.asm
  3852.  
  3853. INFO-KERMIT DIGEST V12 #7                                               Page 69
  3854.  
  3855.  
  3856. 89. Extend automatic detection of unexpected parity in packets to include the
  3857.     first four protocol characters rather than the first two. In addition,
  3858.     perform the test only on the first good received packet of a file transfer
  3859.     group. The test is effective only if the local parity setting is NONE,
  3860.     and if another setting is detected from the host then the local setting
  3861.     is changed permanently. Files msscom, mssrcv, msssen, msssser.asm.
  3862.  
  3863. 90. Correct some subtlities in command line parsing when a local prompt is
  3864.     offered, such as in GETting files with an overriding local name. Files
  3865.     msscmd.asm and mssker.asm.
  3866.  
  3867. 91. Add a text cursor symbol, an 8x8 dot hollow box, in Tektronix graphics
  3868.     terminal emulation mode so we can see where we are. The box is removed
  3869.     before generating a TIFF format screen dump (default keys Control-End) and
  3870.     it is restored afterward. The box is XOR'd with the picture so that no
  3871.     information is lost. File msgibm.asm.
  3872.  
  3873. 92. Add command SET TERMINAL GRAPHICS CURSOR {ON, OFF}, default is ON. This
  3874.     permits one to turn off the text cursor while in graphics (Tek) mode.
  3875.     The cursor is suppressed around almost all drawing and Sixel commands.
  3876.  
  3877.     Add command SET TERMINAL GRAPHICS COLOR <color values> which is analogous
  3878.     to the text mode command SET TERM COLOR and uses the same color values.
  3879.     Two technical notes: the background color cannot be given bold attributes,
  3880.     and the colors are stored in DEC VT340 color palette slots 0 (background)
  3881.     and 7 (foreground).
  3882.  
  3883.     Be wary of two side effects of colored graphics screens: most Print Screen
  3884.     programs will put a black dot on the printer paper if a dot is illuminated
  3885.     at all, and Sixel graphics normally add (OR) to the exiting dot coloring
  3886.     and hence the coloring may not be what was intended. Resetting the
  3887.     emulator restores the colors specified by these SET TERMINAL commands.
  3888.     Files msyibm.asm and msgibm.asm
  3889.  
  3890. 93. To the mouse support add setting of the span of cursor motions to be the
  3891.     current screen dimensions, for mouse drivers which may not do this
  3892.     automatically. An internal item, no user commands. Thanks to Leland Bruns
  3893.     of Intel.
  3894.  
  3895. More changes are expected.
  3896.  
  3897. ------------------------------
  3898.  
  3899. Date: Fri, 2 Nov 90 22:55:21 EST
  3900. >From: Robert Weiner <rweiner>
  3901. Subject: Rainbow Kermit Updated Too (for testing)
  3902. Keywords: Rainbow Kermit, DEC Rainbow, MS-DOS
  3903.  
  3904. The current version of DEC Rainbow 100 MS-DOS Kermit is ready for testing
  3905. and is available on watsun.cc.columbia.edu in the following files:
  3906.  
  3907.     ~kermit/test/mstrb1.boo        BOO Encoded MS-DOS Executable
  3908.     ~kermit/bin/mstrb1.exe        MS-DOS Executable
  3909.  
  3910. This version was built using the current generic MS-DOS Kermit modules which
  3911.  
  3912. Page 70                                               INFO-KERMIT DIGEST V12 #7
  3913.  
  3914. exist on watsun.cc.columbia.edu in the ~kermit/test directory, and therefore
  3915. contains many of the bugfixes and enhancements which
  3916. have been mentioned for the IBM-PC version.
  3917.  
  3918. [Ed. - A version of MS-DOS Kermit 3.02 for the Heath/Zenith 100 is expected
  3919. soon too.  Watch Info-Kermit for an announcement.]
  3920.  
  3921. ------------------------------
  3922.  
  3923. Date: Wed Nov  7 17:01:54 1990
  3924. >From: Christine M Gianone
  3925. Subject: New Kermit for Gould/SEL 32/77 MPX
  3926. Keywords: Gould/SEL Kermit, MPX Kermit
  3927.  
  3928. This is to announce a Kermit server program for the Gould/SEL 32/77 computer
  3929. with the MPX 1.5E operating system, from Barry M. Wilson, Queensland
  3930. Electricity Commision, Belmont State Control Centre, 311 Wecker Road,
  3931. Mansfield, Queensland 4122, Australia, phone (07) 343 0211.
  3932.  
  3933. There is no user interface, it is a server only.  It supports long packets,
  3934. up to 768 bytes in and 1000 bytes out.  It handles run-length compression
  3935. encoding in incoming packets, but does not compress when sending.  There is no
  3936. further documentation.
  3937.  
  3938. The source files and build procedure are collected together into a single
  3939. file, kermit/d/gm3ker.jar (a "join archive") on watsun for anonymous FTP
  3940. (Internet) and GM3KER.JAR from KERMSRV at CUVMA (BITNET).
  3941.  
  3942. Thanks to Barry for doing the work and sending it in!
  3943.  
  3944. ------------------------------
  3945.  
  3946. Date: Sun, 14 Oct 90 17:20:36 -0700
  3947. >From: mcc@WLV.IMSD.CONTEL.COM (Merton Campbell Crockett)
  3948. Subject: MS-Kermit 3.02 Argument Handling, Keyboard Mapping
  3949. Keywords: MS-DOS Kermit 3.02
  3950.  
  3951. When using version 3.02 of MS-Kermit, I get an error, "?word xxxxx not usable
  3952. here.", when entering the command
  3953.  
  3954.         dial xxxxx
  3955.  
  3956. however, although an error is reported, the HAYES.TAK command file is invoked
  3957. and attempts to dial "ATDTxxxxx"--not exactly the intended operation.
  3958.  
  3959. The DIAL macro is the one defined in Christine Gianone's MSKERMIT.INI file.
  3960. The LOOKUP macro was modified as shown below.  The macros worked under version
  3961. 3.00 and 3.01.
  3962.  
  3963. [Ed. - Right you are.  There were many complaints about the way Kermit's macro
  3964. arguments were handled -- they were not saved and restored across calls like
  3965. real function arguments.  Now they are.  Unfortunately this breaks the nice
  3966. LOOKUP and DIAL macros.  They must be slightly recoded to use a global
  3967. variable.  Hopefully the benefits are worth it.  Something like this,
  3968. maybe:
  3969.  
  3970.  
  3971. INFO-KERMIT DIGEST V12 #7                                               Page 71
  3972.  
  3973. def lookup -
  3974.   if eq \%1 compuserve  def \%9 7654321,-
  3975.   if eq \%1 telenet     def \%9 8765432,-
  3976.   if eq \%1 fishnet     def \%9 9876543,-
  3977.   if eq \%1 hairnet     def \%9 5551234,-
  3978.   if def \%9 assign \%n \%9
  3979.  
  3980. ;   DIAL macro to dial a number.  The argument can be either a telephone
  3981. ;   number or a string defined in a lookup(x) macro.  The macro requires
  3982. ;   HAYES.TAK to be in the current directory or in a defined DOS PATH.
  3983.  
  3984. def dial if = argc 2 assign \%n \%1,-
  3985.   if < argc 2 if not def \%n fatal {Dial what?},-
  3986.   if > argc 2 fatal {No spaces please.},-
  3987.   lookup \%n,-
  3988.   assign \%1 \%n,-
  3989.   take hayes.tak
  3990.  
  3991. I connect to several different office automation systems based on Unix
  3992. platforms which presume that the "terminal" is a DEC VT100 class terminal and
  3993. make heavy use of the VT100 keypad for functions and subfunctions.  I also use
  3994. EDT on IAS and VMS platforms.  The following excerpt from my MSKERMIT.INI
  3995. command file may be of interest to others with 101 key keyboards who find
  3996. MS-DOS Kermit standard mappings hard to remember or would just prefer to have
  3997. the keypad in a manner more similar to a VT100, VT200, or VT300 class
  3998. terminal.
  3999.  
  4000. [Ed. - Many thanks, your contribution has been added to MSIVT3.INI, our
  4001. collection VT300 keyboard mappings...]
  4002.  
  4003. ------------------------------
  4004.  
  4005. Date: Wed, 1990 Oct 10   22:35 EDT
  4006. >From: "John F. Chandler"   <PEPMNT@CFAAMP.HARVARD.EDU>
  4007. Subject: Re: Kermit REMOTE SET Command
  4008. Keywords: REMOTE SET, Kermit Commands, Remote Commands
  4009.  
  4010. > This is actually a tricky situation.  The idea was that TRANSPARENT should
  4011. > be the default, for compatibility with the hundreds of Kermit programs that
  4012. > do not support the character-set extension.
  4013. >
  4014. Let's not forget that it would be the default only for systems whose native
  4015. character set is ASCII.  Its meaning, in fact, is rather nebulous, since it
  4016. is very similar to FILE TYPE BINARY for ASCII systems.  What it means, in
  4017. short, is "use the FILE CHARACTER-SET for the TRANSFER (or vice versa)."
  4018. However, a TRANSPARENT sender cannot presume to tell the receiver what
  4019. character set to use, so the situation is, indeed, tricky.  The natural
  4020. Encoding attribute for TRANSPARENT transfers is "*!A" on all ASCII systems (or
  4021. perhaps no attribute at all), but there is *no* natural value on non-ASCII
  4022. (e.g., EBCDIC) systems.
  4023.  
  4024. > Adding a new designator to denote TRANSPARENT might confuse Kermits that
  4025. > don't understand it, including Kermits that support the character-set
  4026. > extension, if that designator were to arrive in an Attribute packet.
  4027. >
  4028. You bet.
  4029.  
  4030. Page 72                                               INFO-KERMIT DIGEST V12 #7
  4031.  
  4032.  
  4033. > So far, it has been an implementation decision about whether character-set
  4034. > changes based on incoming attribute packets are "sticky".  If they are, this
  4035. > presents the problem that there is no way for a client to tell a server to
  4036. > switch from translation to transparent.
  4037. >
  4038. .. except, of course, REMOTE KERMIT SET TRANSF CHAR TRANSP.
  4039.  
  4040. > So it would appear that a decent compromise would be to invent a new
  4041. > designator, say K0, for TRANSPARENT, and to support it in REMOTE SET
  4042. > TRANSFER CHARACTER-SET commands, but, at least for a time, not use it in
  4043. > Attribute packets.  Opinions?]
  4044. >
  4045. Yes.  I would say, though, that K0 should never be allowed in A-packets.
  4046.                                            John
  4047.  
  4048. ------------------------------
  4049.  
  4050. Date: 11 Oct 90 20:17:40 GMT
  4051. >From: dje@uswat.uswest.com
  4052. Subject: Kermit for STRATUS/VOS?
  4053. Keywords: Stratus, VOS
  4054.  
  4055. Looking for a version of KERMIT that runs on STRATUS and supports full set of
  4056. KERMIT commands.  Currently have a version of KERMIT that only knows how to
  4057. receive ASCII files.  Would like to have a version that supports both sending
  4058. and the receiving and can act as both the initiating and terminating end of a
  4059. KERMIT session.  Also, would like a version that supports the transportation
  4060. of binary files.
  4061.  
  4062. A portable version would be acceptable.
  4063.  
  4064. [Ed. - Good luck!  In fact, you seem to have more than we do -- our listings
  4065. don't show any entries for Stratus at all.  Anybody out there know anything
  4066. or working on anything for the Stratus?  We have a similar request from
  4067. Bill Oliver, MD, Department of Hospital Labs, UNC Hospitals,
  4068. oliver@durham.med.unc.edu.]
  4069.  
  4070. ------------------------------
  4071.  
  4072. Date: 30 Oct 90 18:15:13 GMT
  4073. >From: gf07+@andrew.cmu.edu (Gordon Maskew Fair, III)
  4074. Subject: Kermit vs Hypercard?
  4075. Keywords: Hypercard Kermit
  4076.  
  4077. Has there been a stack created that will allow one to use all the
  4078. kermit, and/or xmodem, and/or ymodem commands?  If there is anything
  4079. somewhat like this I would greatly appreciate it.
  4080.  
  4081. Thanks in advance,
  4082.  
  4083. Gordon
  4084.  
  4085. ------------------------------
  4086.  
  4087. End of Info-Kermit Digest
  4088.  
  4089. INFO-KERMIT DIGEST V12 #7                                               Page 73
  4090.  
  4091.  
  4092. Page 74                                               INFO-KERMIT DIGEST V12 #7
  4093.  
  4094.  
  4095. Index                                                                   Page 75
  4096.  
  4097. 132 Columns, 49
  4098. ABC-80/800, 4
  4099. ALLin1, 36
  4100. Arrow Keys, 49
  4101. Atari Portfolio Kermit, 8
  4102. Blanks, 16
  4103. Burroughs A-Series, 4
  4104. Character Sets, 60
  4105. CLEAR Command, 37
  4106. CMS Kermit, 53
  4107. CP/M-80 Kermit, 64
  4108. DEC Rainbow, 13, 69
  4109. DECmate, 43
  4110. DESQview, 55
  4111. DPS-6 Kermit, 51
  4112. Error Checking, 57
  4113. Fonts, 58
  4114. German, 28
  4115. GIN mode, 56
  4116. Gould/SEL Kermit, 70
  4117. Graphics, 56
  4118. Honeywell Kermit, 51
  4119. HP-3000 Kermit, 2
  4120. Hypercard Kermit, 72
  4121. IBM 3174 AEA, 5
  4122. IBM 370 Kermit, 5, 16, 41, 43, 62
  4123. IBM Mainframe, 53
  4124. Kermit Commands, 71
  4125. Kermit Protocol, 18, 45, 57, 60
  4126. Kermit-370, 53
  4127. Labeled File Type, 18
  4128. LSE, 48
  4129. Luxor, 4
  4130. MacKermit 0.98, 58
  4131. MacKermit 0.98(63), 58
  4132. Macros, 48, 54
  4133. Microsoft Windows Kermit, 7, 65
  4134. MNP Modems, 56-57
  4135. Modems, 50, 53
  4136. Mouse, 56
  4137. MPX Kermit, 70
  4138. MS-DOS, 69
  4139. MS-DOS Kermit, 5, 13, 33, 39, 49, 54-57
  4140. MS-DOS Kermit 3.0, 8, 28, 48
  4141. MS-DOS Kermit 3.0 Troubleshooting, 14
  4142. MS-DOS Kermit 3.01, 36-37
  4143. MS-DOS Kermit 3.02, 1, 10, 22, 30, 34, 48, 66, 70
  4144. MS-DOS Kermit Server Mode, 15
  4145. MVS/TSO Kermit, 62
  4146. NCR 9800 Kermit, 3
  4147. Novell Network, 33
  4148. OS/2 Kermit, 47, 53
  4149. OS/2 Presentation Manager Kermit, 26
  4150. OS/8, 43
  4151. PDP-12, 43
  4152. PDP-8, 43
  4153.  
  4154. Page 76                                                                   Index
  4155.  
  4156. Pecan, 27
  4157. Performance, 57
  4158. PRIME Kermit, 13, 47
  4159. Rainbow Kermit, 13, 69
  4160. Remote Commands, 71
  4161. REMOTE PRINT Command, 5
  4162. Remote Set, 60, 71
  4163. Screen Settings, 49
  4164. Scripts, 48, 56
  4165. Stratus, 72
  4166. Tandy 100, 53
  4167. TCP/IP, 39
  4168. Terminal Emulation, 48, 54, 56
  4169. tn3270, 16
  4170. UNISYS A-Series, 4
  4171. Unix, 39
  4172. USCD Pascal Pecan Kermit, 27
  4173. Variables, 48
  4174. VAX/VMS, 36, 48
  4175. VAX/VMS Kermit, 6, 13
  4176. VM/CMS Kermit, 16, 41
  4177. VOS, 72
  4178. VT-78, 43
  4179. X.25 Network, 6
  4180.