SY.CHRISTINE@CU20B.COLUMBIA.EDU (Christine M Gianone) (02/05/88)
Info-Kermit Digest Thu, 4 Feb 1988 Volume 7 : Number 5
Departments:
ANNOUNCEMENTS -
New Documentation for Old CIE-680 Kermit
MS-DOS KERMIT -
MS-Kermit Under OS/2
Thanks for rollback help!
The New Tek-Emulating Kermit and SAS
HP-150 Terminal Program
Kermit problems with Visual Commuter computer
Kermit-MS 2.30 & Macintosh II
MISCELLANY -
CMS Kermit 4.0
Problems with Apple Kermit.
GNU Emacs with C64 Kermit
First-Time Download of Kermit on Commodore-64 with Only Tape
Kermit Sources Wanted for iRMX-86
VME/10 Kermit?
VT-52 Emulator for Osborn Exec
One Character Checksum
----------------------------------------------------------------------
Date: Fri 29 Jan 88 17:26:53-EST
From: Frank da Cruz <SY.FDC@CU20B.COLUMBIA.EDU>
Subject: New Documentation for Old CIE-680 Kermit
Keywords: CIE Kermit
Sent in by the author, David S. Lawyer, of Irvine, CA, plus a termcap
entry for it. In KER:CIE680.HLP and .TRM.
------------------------------
Date: Thu, 28 Jan 88 13:53 MDT
From: Joe Doupnik <JRD@USU.BITNET>
Subject: MS-Kermit Under OS/2
Keywords: MS-DOS Kermit, OS/2
[Ed. - A question we're getting a lot these days is "Will (or can) MS-Kermit
run under OS/2?" As far as 2.30 (and earlier) are concerned, the answer is
"yes, but only in DOS compatibility mode", i.e. taking over the whole machine.
As to the possibility of turning it into a real OS/2 application, we have the
following from Joe Doupnik...]
A switch to OS/2 is possible, even in assembler, but a C version would make
this much easier. In simple terms, OS/2 is not an interrupt-driven system but
a more conventional "call" type. So all the Bios/Dos software interrupts get
repackaged to do calls to system functions, much like a C interface. The tough
part concerns hardware interrupts where either the machine interrupt enable bit
needs toggling (STI/CLI), not permitted under OS/2 full mode, and/or the
physical hardware needs attention from the program. Rumors and the Microsoft
Journal say that interrupt handler latency is very, very long on 286 machines
and communications programs should expect troubles. And, naturally, OS/2 runs
only on AT's and above.
So, the short forms are: Yes, it runs now in the DOS box, but a full OS/2
version needs a substantial development effort. The Presentation Manager specs
are still fluid but when stable will require a second pass at OS/2 Kermit to
adapt to window sizing things.
Regards,
Joe D.
[Ed. - Thanks, Joe. If anybody else has any insight into the issues involved
in bringing Kermit to OS/2, please send a message to Info-Kermit!]
------------------------------
Date: Mon, 01 Feb 88 22:04:50 -0800
From: Alastair Milne <milne@ICS.UCI.EDU>
Subject: Thanks for rollback help!
Keywords: MS-DOS Kermit 2.30
To James Coombs, Eric Boehm, Doug Brenner, RECK@DBNUAMA1, Gisbert Selke,
and all the others who were kind enough to respond about accidentally
hitting HOME:
Thanks very much for the quick and thoughtful replies. Though remapping
the HOME key had never occurred to me, it is obviously the quickest and
easiest thing to do. I don't mind having the PAGE UP and PAGE DOWN, since
those let you move back with some control; and I think the END is a good
idea, to get you back where you actually are as quickly as possible. But
the HOME is just grief.
It's marvellous what the net can do for you. The amusing thing is that I
have not yet seen the info-kermit digest in which my own message appeared,
and would never have know that the editor included a reply in it if one of
you hadn't forwarded the copy to me.
Thanks again to all.
Alastair Milne
[Ed. - Also, did you know that there's a kind of automatic END feature? Try
SET TERMINAL ROLL ON.]
------------------------------
>>Date: Mon, 11 Jan 88 11:12:15 est
>>From: snorthc@NSWC-OAS.ARPA
>>Subject: Screen Scroll in MS-Windows
>>Keywords: MS-DOS Kermit, MS-WIndows
>>
>>I have had problems making Kermit scroll up/down screens under MS-Windows
>>when in a window*.
>>
>>Version 2.29b and 2.29c (21 DEC) have been tested with version 1.01 - 1.03
>>and 2.03 of MS-Windows. You are only able scroll up one line.
>>
>>Both Version 2.29b and 2.29c will scroll properly in a window under
>>Windows/386.
>>
>>* In a window refers to setting up the PIF file so that Kermit does not
>>write directly to screen and does not "modify" a com port.
>>
>> Stephen Northcutt (snorthc@nswc-g.arpa)
>>
>>[Ed. - This is a restriction of the program, noted in the MSKERM.BWR
>>file. However, you should still be able to scroll up by using the mouse on
>>the scroll bar.]
[ From jrd - MS Windows 1.x, 2.x operates the screen in graphics mode but
has trouble reporting back text written by a program to the shadow text
screen. Thus, manually scrolling a Kermit screen within a shared MS Windows
window reports total gibberish or worse to Kermit as it tries to save text
lines scrolling off the screen. It's really an MS Windows concern. To
suppress this characteristic roll back is turned off when in MS Windows. MS
Windows/386 is not available locally and I have no idea of how to test for
versions of Windows. Those little white lies we tell in the .PIF file
(Kermit does not directly access the screen, does not use serial ports,
etc!) are to make Windows behave itself but not to control Kermit. What you
see Kermit do in a shared window is accomplished by magic.]
------------------------------
Date: Tue, 12 Jan 88 22:22 N
From: Jnet%"STREB@YORKVM1" 12-JAN-1988 19:15
Via: <IUS%DACTH51.BITNET@CUVMA.COLUMBIA.EDU> (Eberhard W. Lisse)
Subject: The New Tek-Emulating Kermit and SAS
Keywords: MS-DOS Kermit 2.30, Terminal Emulation, TEK
The line that I have used in my SAS routines to start Tek mode and then
turn it off is:
GOPTIONS DEVICE=TEK4010,GEPILOG='18'X,GPROLOG='1B0C'X,GPROTOCOL=GSAS7171
That seems to work just fine....
Jim Streb
Micro Support
York University
Downsview, Ontario
Canada
[Ed. - Presumably, this is when SAS is running on a mainframe, and the user
has a full-screen connection through an IBM 7171 protocol converter.]
------------------------------
>>Date: Tue, 26 Jan 88 10:55:52 PST
>>From: David Mendel <mendel@playfair.stanford.edu>
>>Subject: Kermit 2.30 graphics
>>Keywords: MS-DOS Kermit 2.30A
>>
>> This is a suggested improvement for msvibm.exe 2.30. Please forward
>>to the correct person.
>>
>> I am using msvibm.exe 2.30 on an ATT6300, and I have had the following
>>problem with the graphics mode. Normally I use the graphics mode to build
>>interactive plots using S which puts the command line near the bottom of
>>the screen. If I make one typo, it prints an error message, and puts the
>>cursor at the top of the screen. The problem is that this clears the
>>picture that I have created.
>>
>> What I would like is an option at the 'More >' prompt to move the
>>cursor to the top of the screen without clearing the screen in the
>>process. Perhaps pressing n would move the cursor to the top of the
>>screen without clearing the screen, while pressing any other key would
>>clear the screen as it does now.
>>
>> Several of us use kermit with S here at Stanford and we would find
>>this to be a big improvement.
>>
>>Thanks,
>>David Mendel
[From jrd - The temptation to convert a simple Tek terminal into a modern
one is to be resisted. My suggestion is to educate 'S' to behave differently
when it knows that it owns the bottom line of a screen (Tek or regular
text). How about using the top left corner for user text? 'S' appears to be
a Stanford program so sources are likely to be nearby.]
------------------------------
>>Date: Sun, 10 Jan 88 20:59:48 PST
>>From: Ya'akov_Miles%UBC.MAILNET@um.cc.umich.edu
>>Subject: MSKERMIT ver 2.29C tektronix 4010 won't overlay ALPHA and VECTOR
>>Keywords: MS-DOS Kermit 2.29C
>>
>>Possible bug with MSKERMIT version 2.29C:
>>
>>I have been using MSKERMIT vers 2.29C and have encountered a bug (feature?)
>>in the Tektroniks 4010 emulation mode. Specifically, alpha mode characters
>>will erase and NOT OVERLAY vector mode data. This frequently causes plot
>>titles and labels to obliterate the graphed data.
>>
>>ps: I have an IBM-PC/xt clone with the CGA adaptor.
>>
>>[Ed. - This seems to be the behavior on the CGA, but not the EGA...]
[From jrd - A complete 8 by 8 dot character cell is written for each
character, so that both foreground and background colors are maintained and
characters can be erased. Supression of the background is being considered
but may not be do-able and still maintain the above characteristics. If
annotation is done first then the problem does not occur. I'd like
overlaying too.]
------------------------------
Date: Thu, 17 Dec 87 15:19 PST
From: MEPMESA%UCBCMSA.BITNET@CUVMA.COLUMBIA.EDU (Adlai)
Subject: HP-150 Terminal Program
Keywords: HP-150 Kermit
Hi we have about 20 HP 150's (Hewlett packard Micro computers). We need to
use them as full screen terminals for CMS. Can You Help Us???
We have a copy of mskermit ver 2.27 for the HP-150, it transfer files great.
But it does not work as a full screen terminal, has someone writen a
terminal emulator for the HP-150 ? It would be a very big help to use to
get such a program. Any ideas, information or sujestions would be very much
appreciated.
Adlai Jordan
MESA
Mathematics Engineering Science Achievement
[Ed. - HP-150 Kermit doesn't do any terminal emulation at all, but simply
passes incoming characters to the screen. This goes through the HP-2621
terminal firmware, so you can use HP-150 Kermit for full screen applications
with hosts that know how to drive an HP-2621 terminal. This means you'd have
to modify your protocol converter (Series/1, 7171, or whatever it is) to know
about the HP-2621 control sequences. But don't attempt to use HP-150 as a
terminal emulator at speeds above 4800 baud. It just doesn't work. Finally,
you might want to pick up the 2.30 version of HP-150 Kermit from KERMSRV
at CUVMA. It's in MSTHP1 *.]
------------------------------
Date: Sat, 23 Jan 88 17:08:20 EST
From: Marshall D. Abrams <vugraph@mitre.arpa>
Subject: Kermit problems with Visual Commuter computer
Keywords: MS-DOS Kermit, Visual Commuter
I am having problems getting Kermit to work on a Visual Commuter portable
IBM compatible computer and would appreciate communicating with someone else
who has been successful with this particular brand. I am not on the info-
kermit list, so please respond directly.
I have tried both the Kermit in VTERM and stand-alone Kermit Vers 2.28. Both
work from my office computer, but neither from home on the Visual Commuter.
I don`t think it's the modem or phone line, because I also have an Atari at
home and the Kermit on the Atari works just fine when plugged into the
modem.
I did turn on debug mode. It took about 6-8 tries to get a file name
accepted for transfer, but no data packets ever got through. It just timed
out. I tried adjusting the parity; that didn't help either.
I'm open to suggestions and thank you for your consideration.
Sincerely,
- Marshall D. Abrams, phone: (703) 883-6938
The MITRE Corporation, 7525 Colshire Drive
Mail Stop Z670, Mc Lean, VA 22102
[From jrd - Marshall. Sorry, but I have no information on the Visual
Commuter. Is there anyone else who can help?]
------------------------------
Date: Wed, 03 Feb 88 22:00 EST
From: "Mark B. Johnson" <CDTAXW%IRISHMVS.BITNET@CUVMA.COLUMBIA.EDU>
Subject: Kermit-MS 2.30 & Macintosh II
Keywords: MS-DOS Kermit 2.30 on Mac II, MacKermit
Just as a note, Kermit-MS V2.30 runs standard VT-102 emulation and file
transfer (in background under MultiFinder) just great using the AST286 board
in a Macintosh II. The new Macintosh version is much faster of course, but
for those people who have to have MS-DOS software, it is quite useable. I
will give the TEK emulation a try next. Just thought someone out there
might be able to use this information...
Mark
------------------------------
Date: Thu, 1988 Jan 28 14:21 EST
From: (John F. Chandler) PEPMNT@CFAAMP.BITNET
Subject: CMS Kermit 4.0
Keywords: CMS Kermit
Comments on comments about CMS Kermit 4.0 in recent Info-Kermit Digests...
> 2) This may be a local problem, but I get the following message on startup:
>
> "Handshake is XON -- not needed"
This startup message is just a friendly reminder to heed paragraph 4 in
section 1.2 of the CMS Kermit User's Guide:
CMS is different from some other IBM mainframe systems in that allows
a program to take control of prompting and synchronization on TTY
lines. Kermit-CMS takes advantage of this option, and it is not, in
general, necessary to enable handshaking on the micro Kermit before
connecting to CMS. In other words, handshaking should be suppressed
for both TTY and SERIES1 devices (the micro Kermit should have
HANDSHAKE set OFF, and Kermit-CMS should have HANDSHAKE set to 0).
Since the generic Kermit-370 default handshake (XON) is retained in
Kermit-CMS, the subcommand SET HANDSHAKE 0 is a good candidate for
inclusion in SYSTEM KERMINI.
By the way, I noticed that you didn't "Ed." the comments from Brown in Digest
#3 about trying to set the SEND packet size -- surely, you don't feel that one
Kermit can even try to force another to accept Long Packets; that's what it
would mean if Kermit-CMS accepted the command SET SEND PACKET 1900. Perhaps
the documentation should explicitly say that one can't do that.
[Ed. - Right... that one slipped by.]
------------------------------
Date: Mon, 1 Feb 88 11:36 N
From: <EIFV%EICLUS.CNRTO%ITOPOLI.BITNET@CUVMA.COLUMBIA.EDU>
Subject: Problems with Apple Kermit.
Kewyords: Apple II Kermit
I recently got a copy of version 3.79 of Apple II KERMIT. I experienced
some problems under ProDOS to up/download text files since they are always
handled in 7 bit mode (setting unconditionally the high order bit when
receiving and stripping it off when sending) like under DOS 3.3. This is
incorrect since now the Apple IIgs uses this high order bit to extend the
ASCII character set in the same way as on the Macintosh. Unfortunately I
have no source code to try to patch the code in order to enable a full 8 bit
text file transfer under ProDOS. Can you help me? Thanks in advance & best
regards.
Fabio Viviani
C.S.E.L.T. - Turin, Italy
(eifv@eiclus.cnrto)
[Ed. - This is certainly a trend. Your message has been forwarded to the
current developer of Apple II Kermit, who's working on a new release.]
------------------------------
Date: Wed, 03 Feb 88 22:05:09 -0500
From: ray@j.cc.purdue.edu (Ray Moody)
Subject: GNU Emacs with C64 Kermit
Keywords: C64 Kermit
In Info-Kermit Digest Volume 7, Number 3 you write:
>While using GNU Emacs with C64 Kermit at 1200 baud, I noticed when scrolling
>UP ONLY, the screen would get garbled. (Probally overflowing the input
>buffer? I haven't notice this happening at 300 baud (300 baud?!? ARGH!))
Yes, you are quite right. Since the Commodore-64 has no built-in
80-column screen, I have to simulate 80-columns with bitmap graphics.
Scrolling a bitmap screen backwards is a non-trivial operation and takes a lot
of time. If Kermit receives scroll-reverse requests too fast, it will overflow
its input buffer.
Normally, when Kermit's input buffer is in danger of being overflowed,
Kermit will transmit a ^S to stop the remote host from sending. The only
problem is that GNUemacs chooses to ignore this stop request.
There are several ways to solve this problem:
1) The best solution I can think of is to tell GNUemacs that you wish to
use flow-control. You can do this by putting (set-input-mode nil t)
in your .emacs file.
2) Ask GNUemacs to provide a delay after scrolling the screen backwards.
You can do this by adding sr=<XXX>\EM to your terminal description,
where <XXX> is the number of milliseconds of delay that you want. I
experimented a little and found that 200 milliseconds is "about" right.
This isn't very elegant because it will slow GNUemacs down a lot.
3) Use a Commodore-128 or wait for Kermit 2.1 to be released and use a
Batteries-Included 80-column add on card. If you use one of these,
Kermit will not be forced to simulate an 80 column screen with graphics.
Ray Moody, Author of Commodore Kermit version 2.0
ray@j.cc.purdue.edu
ihnp4!pur-ee!j.cc.purdue.edu!ray
moody@purccvm.BITNET
Many thanks to Jay Vosburgh for providing the magic emacs incantations.
[Ed. - And thanks for passing along the hints. They've also been added to
the Commodore-64 Kermit "beware" file.]
------------------------------
Date: 12 Jan 88 15:12 +0100
From: Alf Christophersen <l_christophe%use.uio.uninett@TOR.NTA.NO>
Subject: First-Time Download of Kermit on Commodore-64 with Only Tape
Keywords: C64 Kermit
A friend of me want to get KERMIT on his Commodore 64 with CP/M, but he
has only a tape station. How do we upload first time, e.g. from a
Olivetti M24 with direct connection over RS232? I remember there was
a procedure when I loaded first time with our Altos 8000/7, but have
forgotten the procedure. Does anyone have some idea? Is there any tape?
Alf Christophersen
Engineer
Dep. of Nutrition Research
PO. Box 1046 BLindern
N-0316 Oslo 3
Norway
------------------------------
Date: Fri Jan 15 00:20:02 1988
From: peregrine!imt3b2!seila!don@uunet.UU.NET
Subject: Kermit Sources Wanted for iRMX-86
Keywords: iRMX Kermit
Anyone out there have working C or PLM sources for kermit under release 6+ of
iRMX-86 running on a 310 box? Please email replies. Thanks in advance!
According to info from Columbia U., someone at Grinnel College (?) has done
this...
Don Kossman, SEI Information Technology, Los Angeles
usenet: {ccicpg!imt3b2 | peregrine!imt3b2 | sun!tsunama!tsunami}!seila!don
[Ed. - There are two families of Kermit programs for (i)RMX(2)86. One
consists of variations on a version written in PL/M, and the other is an
adaptation of MS-DOS Kermit by Jack Bryans at Calstate. Does anyone know
of any reason why the former would still be necessary in the presence of
the latter? And if there is any reason for keeping a PL/M version, which
of the three that we have should be kept? Below is a list...]
IRM C Intel 86/380 iRMX-86 PL/M 2.41 87/03/04 Grinnell Col.
I86 C Intel 86/380 iRMX-86 PL/M 2.3 85/09/23 Grinnell Col.
RMX C Intel 86,286 RMX 1.0 PL/M 1.0 85/10/25 Cornell U
MS A Intel 300 Series iRMX-86 MASM/ASM86 2.29Z 88/01/07 Cal State
MS A Intel 300 Series iRMX-286 MASM/ASM86 2.29Z 88/01/07 Cal State
------------------------------
Date: Thu, 31 Dec 87 07:57:23 CST
From: rod@cnt.mn.org (rod merry)
Subject: VME/10 Kermit?
Keywords: VME Kermit
Does anyone know where I can get a binary copy of Kermit for the Motorola
VME/10 running VERSADOS. I will supply or pay for the diskettes.
Thanks,
Rod Merry rod@cnt.MN.ORG
Computer Network Technology {quest|meccts}!cnt!rod
9440 Science Center Drive
Minneapolis, MN 55428 (612)535-8111
[Ed. - Try Wm. Pierce, Motorola Semiconductor, 3102 N. 56th St, MS/56-122
Phoenix, AZ 85108. We have had reports that this person has, and distributes,
a VME/10 VERSADOS Kermit, but we have never received it. We'd like to carry
a VERSADOS Kermit in our distribution, and would like an "official" source
to refer people to who need it on diskette. If anybody finds out the status
of this purported Kermit version, please let us know.]
------------------------------
Date: 29 Jan 1988 06:18-CST
From: John A. Wright <SAC.HQSAC-DOCT@E.ISI.EDU>
Subject: VT-52 Emulator for Osborn Exec
Does anyone know of a Kermit program with a VT-52 emulator for the Osborne
Executive. If one does not exist, does anyone know of a VT-52 emulator for
the same?
------------------------------
Date: 20 Jan 88 00:37:50 GMT
From: johnm@ncr-sd.SanDiego.NCR.COM (John McDaid)
Subject: One Character Checksum
Keywords: Checksum
Can anyone give me a short, CLEAR explanation on how a one character checksum
is derived in Unix Kermit? Mail me directly with the answer. I thank anyone
who gives this any effort.
John McDaid
John.McDaid@SanDiego.NCR.COM
[Ed. - Add up the 8-bit values of all the bytes from the length field to the
last data byte. Truncate the sum to 8 bits. Call this S. Then:
X = S & 63; /* Low order 6 bits */
Y = S & 192 >> 6; /* High order 2 bits, shifted to right */
C = (X + Y) & 63; /* Add them together, keep low 6 bits of result */
CHKSUM = C + 32; /* Make it printable */
More concisely (as it is actually done in C-Kermit):
CHKSUM = tochar((s + ((s & 192)/64)) & 63);
This was originally designed, back in the days of 8-bit processors, to work
on machines that could only do 8-bit arithmetic. The idea was to have all
the bits of an 8-bit sum contribute to a printable ASCII single-character
checksum.]
------------------------------
End of Info-Kermit Digest
*************************
-------