[net.micro] What are KERMIT and UMODEM anyway?

msk@mtplx1.UUCP (09/19/83)

Hello, net.world.

We have just started reading the USENET.  We have seen some discussion on
KERMIT (in various flavors to  communicate  between  UNIX  and  IBM  PCs)
and UMODEM.

I would like to request information on  these  two  products.   Any  info
about availability, price, machines that they will run on, and any  other
general information is desired.

Now a more general question (but perhaps answered by the above products):

Does anyone have  a  good  system  for  communication  between  UNIX  and
DEC Rainbows?  How about UNIX and  other  PCs  (personal  computers,  not
necessarily IBM)?

Well thank you for reading this.  It seems to me that long  time  netters
may already know all this info but new netters may derive some benefit.

PLEASE MAIL ME RESPONSES SO AS NOT TO CLUTTER UP THE NEWS GROUPS!

Thanks in advance!!!

                                            Morris Kahn
                                            Micro/Tel, Inc.
                                            ... !ihnp4!mtplx1!msk

ABN.ISCAMS@usc-isid@sri-unix.UUCP (09/27/83)

Morris, et al in Netland...

KERMIT -- ah, yes.  I just so happen to have gotten good old Kermit running
on my Decision I (Z80, 4 MHz, Freedom 100 Terminal) pushing out over a Racal-
Vadic VA212LC modem at 1200 baud over a home telephone line, through military
phone lines, through my local host, etc., etc, into the ARPAnet.

Kermit works great -- it's talking to you right now in its "dumb terminal" mode,
passing control characters, escapes, etc., with no problem.  Will try down-
loading text files, .COM files, even 32-bit DEC word files in a bit.

Took much hacking since I started with CPMGENERI.ASM (from FTPing from
COLUMBIA-20's <KERMIT> directory.  Keyed in the entire 100+ pages of
assembly language (yeah, dumb), proofread, debugged, found typos.

Learned how to set my Multi I/O Board up as terminal instead of modem for P
port 3, learned how to make my own RS232 cables and ALL about RS232 (since
I had never interfaced with a modem before), wrote reader/punch drivers in
Kermit itself (since my CBIOS doesn't have them, nor does it implement
IOBYTE, so the GENERIC KERMIT wouldn't work).  In other words, a fine, fine
time was had by all -- but KERMIT WORKS on a Decision I!  And I, a bloody
novice and hacker extraordinairre, did it!  

Works just fine; will try with Unix, etc.  The key is -- you can ALWAYS
find a KERMIT out there running in his own Unix, IBM, whatever, environment,
and the Kermit's can talk to each other for file transfers, etc.  I haven't
heard much traffic on what different Kermit's are talking, or how well the
dumb terminal program works in other nets.  However, it's completely trans-
parent on the ARPAnet and my local host; and I can always drop out of the
ARPAnet, play on the micro a while, and come right back up (provided my
host didn't time me out and AUTOLOG me off!)

Message me more if interested.  My patches to GENERIC KERMIT will be sent
to COLUMBIA-20 and KERMIT there, rather than clog up the net any more.

Sorry, guys, for the VERY large message -- but I'm really kind of excited
at my success (I couldn't even spel asemblur last April!).

Regards from Toad Hall
David Kirschbaum
SGM, USA
HQ XVIII Abn Corps

ABN.ISCAMS@usc-isid@sri-unix.UUCP (09/27/83)

KERMIT -- a program that (1) can act as a dumb terminal to talk over modems
with anything pretty much, and (2) a program that can talk with other KERMITs
that run in a different environment.  The KERMITs don't know that, so they can
talk to each other for file transfers, etc.

And it works.  I'm presently typing this on my Freedom 100 terminal, hooked
to my Decision I Z80 micro, pushing out port 3 to a Racal-Vadic VA212LC
1200 baud modem, over a private phone line to a port on our local host
here at Fort Bragg for the ARPAnet.

My KERMIT was downloaded in hardcopy by FTPing from COLUMBIA-20's
<KERMIT> directory.  I got the CPMGENERI.ASM (GENERIC KERMIT) version,
and keyed in all bloody 100+ pages!  Found out my Morrow CBIOS DOES NOT
implement the IOBYTE GENERIC KERMIT relies on, so I had to learn all about
my Multi I/O board, ports, etc.  Then had to reconfigure the I/O board as a
terminal instead of modem (easy).  Then had to write reader/punch drivers
(stealing code from my CBIOS like the CONIN and CONOUT) in assembler into
KERMIT, pick and choose which parts of KERMIT to use (it's full of IF-END
statements for a bunch of machines) ... but I GOT IT WORKING!!

Dumb terminal works fine -- pushes out control characters, escapes, etc.;
is completely invisible.  Yet, at any time I can drop out of my link with
the mainframe, play micro for a while in CP/M, and come right back up
to where I was.  Will try downloading and uploading all sorts of files next
(.COM, .HEX, text, even 32-bit words from the DEC just to drive my terminal
crazy).

Don't have much info on exactly what different environment KERMITS are
talking to each other, but mine sure works great on the ARPAnet!

I'll be posting details and patches to COLUMBIA-20 <KERMIT> as to the
how-to and wuffos of the Decision I implementation.

If this works, I'll be glad to upload the entire bloody patched and kludged
TOADKRMT.ASM or .HEX or .COM or .PRN (if you'll pay the phone bill) to
anyone who wants it.  (Subject to the usual public domain constraints,
of course -- and all full credit to Columbia, Bill Catchings, BG Eiben, Frank 
da Cruz, Daphne Izoar, Bernie Eiben, et al.)

David Kirschbaum
SGM, USA
HQ XVIII Abn Corps and Toad Hall