home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.20020314-20021006
/
000065_fdc@columbia.edu_Wed May 1 10:37:19 EDT 2002.msg
< prev
next >
Wrap
Text File
|
2002-10-06
|
9KB
|
192 lines
Article: 13356 of comp.protocols.kermit.misc
Path: newsmaster.cc.columbia.edu!news.columbia.edu!news-not-for-mail
From: fdc@columbia.edu (Frank da Cruz)
Newsgroups: comp.protocols.kermit.misc
Subject: Re: a bug on GNU/linux: speed reset to unintended value occasionally.
Date: 1 May 2002 10:37:10 -0400
Organization: Columbia University
Lines: 175
Message-ID: <aaouim$di0$1@watsol.cc.columbia.edu>
References: <3CAFF81C.8039CBF8@yk.rim.or.jp> <3CCB4F33.4F2B2C62@yk.rim.or.jp> <aahd9h$qii$1@watsol.cc.columbia.edu> <3CCF46F4.DE916308@yk.rim.or.jp>
NNTP-Posting-Host: watsol.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 1020263832 13865 128.59.39.139 (1 May 2002 14:37:12 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 1 May 2002 14:37:12 GMT
Xref: newsmaster.cc.columbia.edu comp.protocols.kermit.misc:13356
In article <3CCF46F4.DE916308@yk.rim.or.jp>,
Ishikawa <ishikawa@yk.rim.or.jp> wrote:
: Frank da Cruz wrote:
: > ...
: > Kelly-Bootle, Stan,
: > "680x0 Programming By Example",
: > Howard W Sams & Co, Indianapolis IN (1988),
: > ISBN 0-672-22544-1.
: >
: > In this case it was the Alpha Micro version written by Rob Rubendunst:
: >
: > ftp://kermit.columbia.edu/kermit/c/am*.*
:
: So I was quite impressed at the assembly program listed above.
: Never realized that the directory contains source code files
: for a totally different architecture...
:
Most of the early Kermit programs (CP/M, DEC-20, PDP-11, IBM 370, MS-DOS)
were written in assembler, I don't think I have ever counted the number of
different assembly languages represented in the archive, but it is a good
sampling. We used to keep lists of Kermit programs sorted by various
categories: date, platform, OS, language, etc, but these lists do not always
distinguish different assemblers:
ftp://kermit.columbia.edu/kermit/a/aavlng.txt
: > And somewhat more sensationally:
: >
: > Stoll, Clifford:
: > "The Cuckoo's Egg: Tracking a Spy Through the Maze of Computer Espionage",
: > Doubleday, New York (1989),
: > ISBN 0-385-24946-2
:
: I have read this book. But it was quite some time ago, and can't recall
: KERMIT mentioned in this book. Oh well.
:
In the hardcover edition: pages 45 ("The hacker wasted no time. He issued
commands to show all the active users and any background jobs running. Then
he fired up Kermit." [Paragraph explaining what Kermit is... "the Esperanto
of computers"] ... "I watched as the hacker used Kermit to transfer a short
program into our Unix computer..."]; pages 90-93 [a more detailed explanation
of Kermit protocol, and the measurement of round-trip time of Kermit packets
to estimate the hacker's distance]; 149-156 [the keychain gambit].
: These days, a certain large software company decided to
: regard serial ports as "legacy hardware" and seem to
: discourage the use of them.
:
Of course this is unfortunate, but on the other hand it leaves a niche
for makers of software and devices that are good for something besides
shopping and making pie charts.
: > http://www.columbia.edu/kermit/ckcplm.html#x3
: >
: I read the page and it was quite refreshing to read
: these do's and dont's.
:
: I think programming KERMIT is quite comparable to embedded computer
: programming today. Often the development tools for embedded computer
: systems are under preparation for a new embedded systems and so
: assembler/compiler/linker and realtime OS kernel itself (if it exists at
: all!) is somewhat unreliable, and so we need to take a very careful and
: conservative approach in coding.
:
Quite true. Even today we try to make C-Kermit compatible with old
compilers and linkers, because many people still have them. It was only
recently that we had to give up on supporting 16-bit architectures in
C-Kermit (but of course G-Kermit still supports them, not to mention
MS-DOS Kermit and many others).
: (When I talk of embedded computer systems, I mean computer systems
: embedded in *small* appliances such as toaster, microwave oven,
: refrigerator, TV, VCR, and many other home appliances as well as
: automotive engine control system and other industrical controller. Some
: embedded systems for industry control system may use, say, Sun ULTRA
: pizza box, but most of these small systems use a tiny single printed
: circuit board with a version of 8, 16 or 32 bit computers.)
:
Right, and as you say, the programming requirements are quite rigid; not
what you learned in school :-) We often get requests for an embeddable
Kermit protocol implementation, and I'm happy to say that soon we will
have one.
: One thing that I have been wondering is this.
: How is KERMIT project sustained.
:
Whenever this topic comes up, it results in a flame war. Everybody has
strong opinions about licensing.
In the early days, we had a group of programmers who were paid to write
Kermit software, so of course we gave it away (but we retained copyright).
Soon other sites began to write their own Kermit programs and send them
back to us. Naturally, we worked with them to ensure interoperability,
a compatible command language, etc. This was great, but it was also
more work for us.
Later, it was no longer our job to develop Kermit software or coordinate
developers at other sites, or take care of the archive. But the world
still wanted and expected us to do the work, and we enjoyed it. So Kermit
work was done nights and weekends. This might be OK for students, but
it's not good for grown-ups with families.
Eventually (1986) we convinced our management to create the Kermit
Project. It would coordinate development, take care of the archive,
package and distribute the software, run a mailing list, publish a
newsletter/journal, but NOT write software. The project was to pay for
itself by charging a distribution fee for Kermit tapes and disks.
Later, as the demand for mail-order tapes and disks dimished, we wrote
some books, and our business model became something like: "the software is
free for your own use but please buy the book, and if you want to sell or
give the software to customers or clients, you *must* include the book".
This worked for a few years, but people didn't like it because in most
cases they did not want a book.
Finally in 1995, I joined the Kermit Project full time and we hired
another full time developer, Jeff Altman, with the understanding that the
Kermit Project must pay for itself out of revenue. We released Kermit 95
as a commercial product that year. People understand this much better; no
lengthy explanations are required, no arguing about "save a tree", etc;
it's quite straightforward.
Except for Kermit 95, most Kermit software is still "free" in the sense
that you can download it -- source code and all -- for your own use or
use within your company, but you can not make a product out of it or
include it with a product without a license; in other words, if you make
money (or otherwise gain commercial advantage) from our work, we deserve
a share. We believe this is fair and reasonable.
Our current business model is spelled out here:
http://www.columbia.edu/kermit/commercial.html
: Does Columbia University accept/solicit donations to
: Kermit project?
:
We would like to have donations, but this is not an adequate business
model, especially in these hard times.
: And I am curious and nosy. What is your, er, regular
: job at Columbia?...
:
Now it is Kermit, but many others before that; you can read a little bit
about this here:
http://www.columbia.edu/acis/history/
: KERMIT is one of the few (if any) powerful serial terminal program on
: UNIX with good track record under noisy and flakey connection. (tip is
: a joke as a terminal program. UUCP is only for batch transfer/invocation
: remotely.)
:
Thanks. Of course Kermit is much more than a serial-port terminal program.
Unfortunately many people don't know that; their conception of Kermit has
not changed since they first looked at it 10 or 20 years ago. It's hard
to change first impressions.
: So having KERMIT project afloat is clearly a merit to a sizable number
: of programmers all over the world. (Right, many now use Windows and
: there are good terminal emulators with KERIMT protocol support.
:
But aside from Kermit 95, I don't think there are any good Windows terminal
emulators with *good* Kermit protocol support!
: But as
: far as I can tell UNIX has KERMIT only. Maybe KERMIT was so good that
: nobody bothered to write other serial terminal program.)
:
There are a couple others; see:
http://www.columbia.edu/kermit/kermit.html
for a discussion of this topic, and also for more about licensing.
- Frank