home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Columbia Kermit
/
kermit.zip
/
newsgroups
/
misc.19950929-19951130
/
000323_news@columbia.edu_Mon Nov 6 00:54:30 1995.msg
< prev
next >
Wrap
Internet Message Format
|
2020-01-01
|
7KB
Received: from apakabar.cc.columbia.edu by watsun.cc.columbia.edu with SMTP id AA23066
(5.65c+CU/IDA-1.4.4/HLK for <kermit.misc@watsun.cc.columbia.edu>); Sun, 5 Nov 1995 19:54:39 -0500
Received: by apakabar.cc.columbia.edu id AA22591
(5.65c+CU/IDA-1.4.4/HLK for kermit.misc@watsun); Sun, 5 Nov 1995 19:54:36 -0500
Path: news.columbia.edu!watsun.cc.columbia.edu!fdc
From: fdc@watsun.cc.columbia.edu (Frank da Cruz)
Newsgroups: comp.protocols.kermit.misc
Subject: Re: how to get DOS kermit c source code?
Date: 6 Nov 1995 00:54:30 GMT
Organization: Columbia University
Lines: 102
Message-Id: <47jmc6$m1t@apakabar.cc.columbia.edu>
References: <45pk9f$so3@info.bta.net.cn> <47etn2$eq7@Mars.mcs.com> <47g716$s75@apakabar.cc.columbia.edu> <47jejo$lkh@Mercury.mcs.com>
Nntp-Posting-Host: watsun.cc.columbia.edu
Apparently-To: kermit.misc@watsun.cc.columbia.edu
In article <47jejo$lkh@Mercury.mcs.com>, Leslie Mikesell <les@MCS.COM> wrote:
: >Good points! But I still don't understand your insistence that the
: >software (aside from Kermit 95) is not freely available. See my other
: >posting of today.
:
: In article <47g716$s75@apakabar.cc.columbia.edu>,
: Frank da Cruz <fdc@watsun.cc.columbia.edu> wrote:
: Availability isn't the only issue. If kermit isn't *already* at the
: other end you have a problem. What is your estimate of this being
: the case the first time someone wants to communicate?
:
That's a good question. I don't think anybody knows the answer. But in
each particular case, you will find (a) no Kermit at all, (b) a Kermit
implementation that doesn't even work, (c) a minimalistic Kermit that barely
works but is slow as molasses, or (d) a modern up-to-date Kermit.
Cases (b) and (c) result from a lack of control in the early days -- just
the kind of "freedom" you are advocating now. Of course (d) would be our
preference, but using 20-20 hindsight, I think I'd prefer (a) to (b) and (c),
since all the crummy implementations out there are what have given many
people the mistaken idea that Kermit protocol and software themselves are
intrinsically crummy -- including people who should know better, such as major
modem manufacturers (read any modem manuals lately?).
: >: But BSDI, freeBSD and netBSD seem to be going strong.
: >:
: >And each off in its own direction. Look at how each of these needs
: >separate code in Kermit to support. Wouldn't it be nice if the world
: >were more consistent. There is something to be said for centralized
: >coordination and management, and that's what we aim to provide.
:
: That would be pretty much irrelevant if you allowed those distributions
: to include a working kermit binary. Someone else would do it. Especially
: if you would move the code that deals with tty lines into a separate
: module so it could simply be replaced for each major variation instead
: of being a nightmare of #ifdef's.
:
I challenge you to do a better job. Do you have 10 spare years? Seriously,
it is the age-old tradeoff between modularity and code sharing. If we
split the ckutio.c module off into 400-500 different copies, then any
functional improvement we make in one place might well need to be made in
hundreds of others.
That's not to say there couldn't be any improvements, of course. Only that
the scope and complexity of this project are way beyond anything anybody
ever expected at the beginning, and so some of the various modules have grown
in, shall we say, unexpected ways. And at each step along the way, it has
almost always been not only easier, but SAFER, to add some new arcane UNIX
system-dependency within #ifdefs than to rewrite 50,000 lines of code.
By safer, I mean satisfying somebody's need for the new thing without
breaking support for hundreds of other platforms in the process.
If we had it all to do over again, maybe we'd also design C and UNIX
themselves to make just a wee bit more sense.
: >Our code is freely distributed too. The question is how can it be
: >REdistributed? The reasonable restrictions we have placed on
: >commercial redistribution arose out of necessity to preserve the
: >Kermit Project, because without them we were being devoured by
: >profiteers. Perhaps it is a matter of opinion whether we have
: >chosen the right way, but in the final analysis we have to make
: >the decision.
:
: It is freely distributed to people who have ftp access. Why do
: those people need kermit other than for terminal emulation? The
: people who need it for file transfers don't have it and can get
: other products easier. If that is the way you wanted things, then
: I guess you made the right decision, but it sure doesn't follow the
: spirit of the old kermit documents.
:
Well, we're beginning (again) to go in circles here. I freely and openly
confess that the spirit in which the original Kermit documents were
written no longer totally applies. The world is a meaner place now -- to
us and to everybody. It would be easy for us to say (as you want us to),
"OK, it's free again, anybody can do whatever they want to with it", and
then walk away and get new jobs and "let the hundred flowers bloom."
Before you knew it, you'd have 1000 incompatible divergent strains with no
way on earth to make improvements to all of them, and not long after that
you'd begin to see failures to interoperate, as clever programmers
improved the protocol.
: >: Besides, at the moment the most popular communications platform
: >: is probably Windows 3.x running a dial-up or network winsock
: >: which seems to be a gaping hole in the kermit product line.
: >:
: >Granted. At least two projects were started to fill this gap, but
: >were not completed. That's one of the pitfalls of free software
: >-- with very few exceptions, you can't count on people completing
: >projects when you aren't paying them anything (for an even more
: >graphic illustration of this point, look at the history of Mac
: >Kermit). Would anybody like to volunteer to take on the Windows
: >3.1 Winsock / MS-DOS Kermit project? Five or ten years ago there
: >would have been plenty of takers.
:
: I suspect that there would still be takers if the resulting product
: could be used by the person/company that did the port.
:
Who said it couldn't? If somebody wants to do this work and also be
able to profit from it, all they need to do is talk to us. We are
reasonable people.
- Frank