home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
archives
/
dec20.zip
/
k20mit.ann
< prev
next >
Wrap
Internet Message Format
|
2003-12-21
|
6KB
From slogin@optonline.net Sun Dec 21 13:03:53 2003
Flags: 000000000001
Return-Path: <slogin@optonline.net>
Received: from mta11.srv.hcvlny.cv.net (mta11.srv.hcvlny.cv.net [167.206.5.86])
by cherimoya.cc.columbia.edu (8.12.10/8.12.10) with ESMTP id hBLI3qef004379
for <fdc@columbia.edu>; Sun, 21 Dec 2003 13:03:53 -0500 (EST)
Received: from optonline.net (mstr3.srv.hcvlny.cv.net [167.206.5.14])
by mta11.srv.hcvlny.cv.net
(iPlanet Messaging Server 5.2 HotFix 1.16 (built May 14 2003))
with ESMTP id <0HQ9006NTBIG8D@mta11.srv.hcvlny.cv.net> for fdc@columbia.edu;
Sun, 21 Dec 2003 13:03:52 -0500 (EST)
Received: from [167.206.5.76] by mstr3.srv.hcvlny.cv.net (mshttpd); Sun,
21 Dec 2003 12:57:57 -0500
Date: Sun, 21 Dec 2003 12:57:57 -0500
From: slogin@optonline.net
Subject: DEC-20 Kermit version 5.1(182) Available
To: Tops-20 Wizards <TOPS-20@lingling.panda.com>
Cc: Frank da Cruz <fdc@columbia.edu>
Message-id: <53db0053b2fa.53b2fa53db00@optonline.net>
MIME-version: 1.0
X-Mailer: iPlanet Messenger Express 5.2 HotFix 1.16 (built May 14 2003)
Content-type: text/plain; charset=us-ascii
Content-language: en
Content-transfer-encoding: 7BIT
Content-disposition: inline
X-Accept-Language: en
Priority: non-urgent
X-Priority: 5 (Lowest)
X-Spam-Score: 0.339 () NO_REAL_NAME
X-Scanned-By: MIMEDefang 2.35
Columbia University announces a new version of DECSYSTEM-20 Kermit
available for use. The DECSYSTEM-20 was one of the original Kermit
programs created in the early 1980's. The source is available at
ftp://kermit.columbia.edu/kermit/d/k20*.* which can be accessed via
anonymous login.
Updated release notes for this version can be found in
ftp://kermit.columbia.edu/kermit/d/k20mit.txt. This is version
5.1(182) and incorporates the following changes/fixes:
1) Automatic TVT-Binary detection/negotiation. Kermit will try to
determine if it is running on a Internet TVT line and enable
network binary mode if so. If PANDA MTOPR%s exist to do this,
then they will be used.
Otherwise, the older method of "sending magic characters to the
TAC" is used. TVT negotiation can be over-ridden and either
forced on or off. This code has been slightly enhanced to make an
attempt to flush some TCP buffers.
2) References to the ARPAnet in the online documentation have been
changed to Internet.
3) Some minor bugfixes concerning error handling. In particular,
Kermit was trying to do a CFIBF% of .FHSLF (400000). This doesn't
work and will produce a 602030 error, "Line is not active". It
works much better to use the JFN ...
Users of older PANDA distributions may be interested in upgrading to
this version as the shipped release of Kermit with PANDA was 4.1(174)
which dates from 1985. The following changes/enhancements have been
entered since then:
[180] Buffered packet input instead per-character BIN% + better
statistics.
[179] Long packets:
. Added new symbol MAXBUF (10KB)
. Increased packet buffer sizes to MAXBUF
. Changed .setpk to allow sizes up to 9000 (near theoretical
max)
. Added CAPAS, WINDO, MAXLX1, MAXLX2 field support to rpar
and spar.
. Added long-packet sending/reading to spack/rpack.
. Removed hardwired ^d94 packet-length references.
. Speed improvements by up to a factor of 20.
. Reduce default delay to 1 sec.
[178] Allow passwords to come from TAKE files.
[177] Condition the line correctly.
[176] Allow commands to servers to be issued from TAKE file.
Also, pass capabilities down to inferior process correctly.
[175] Delete any dot at the end of outbound file names (like
'makefile.').
-In particular, edits 179 and 180 provide substantially improved
performance. However, do NOT attempt to use packet sizes of over 94
bytes on a real KL-10 with an RSX20F PDP-11 handling communications:
you will crash the front end.
Some users an the KLH-10 platform may notice some differences in
performance, depending on whether the 20 is uploading or downloading
data. An upload (the 20 sends a file) seems to be greatly slower than
a download (the 20 receives a file). This is not unique to Kermit, I
have also observed it with FTP (see below) on a 100BaseT switched
infrastructure.
Independent tests by Frank have shown that he had to shorten the
packet sie to 500 bytes for sending and 200 for receiving; whereas he
could use 9K packets in both directions with the XKL (and no
retransmissions).
I can offer no explanation for this behavior as I no longer have
access to anything but a KLH-10. If anyone else does have access to
alternate platforms and is willing to provide an account, I'd be
interested in talking with you about possibly doing some more testing.
Comments (rants, raves) concerning edits 181 and 182 can be sent to me
as I did the work of merging in the PANDA changes with the main line
and the bug fixes. In depth questions about Kermit itself or older
edits can be referenced to Frank da Cruz.
There were a number of reasons for producing this version of Kermit.
The PANDA MTOPR%s being the primary ones. I got involved because I
needed a reference transfer program to test FTP against. I've been
tinkering with the TOPS-20 FTP server on and off for the past few
months to make it work with some more 'modern' clients.
In this case, these would be Emacs ange-ftp mode (I'm attempting to
transmorgrify the VMS mode into a Tops-20 one) and the innovative
Windows Interner Explorer. Thus far, I have added support for PWD,
SYST, CDUP, SIZE and MDTM to it.
--T
PS: No systems programmers were harmed or became unemployee
during edits 180 and 181.
Pps: Happy DEC20+1 Day!