home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Columbia Kermit
/
kermit.zip
/
newsgroups
/
misc.19950528-19950726
/
000062_news@columbia.edu_Mon Jun 5 13:18:58 1995.msg
< prev
next >
Wrap
Internet Message Format
|
2020-01-01
|
6KB
Received: from apakabar.cc.columbia.edu by watsun.cc.columbia.edu with SMTP id AA16931
(5.65c+CU/IDA-1.4.4/HLK for <kermit.misc@watsun.cc.columbia.edu>); Mon, 5 Jun 1995 09:19:17 -0400
Received: by apakabar.cc.columbia.edu id AA07043
(5.65c+CU/IDA-1.4.4/HLK for kermit.misc@watsun); Mon, 5 Jun 1995 09:19:13 -0400
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: Kermit MAC??
Date: 5 Jun 1995 13:18:58 GMT
Organization: Columbia University
Lines: 105
Message-Id: <3qv083$6r1@apakabar.cc.columbia.edu>
References: <3qnon9$asf@steel.interlog.com> <3qtbvi$a4o@news.doit.wisc.edu> <3qtfnp$gca@apakabar.cc.columbia.edu> <3qtt35$gji@news.doit.wisc.edu>
Nntp-Posting-Host: watsun.cc.columbia.edu
Apparently-To: kermit.misc@watsun.cc.columbia.edu
In article <3qtt35$gji@news.doit.wisc.edu>,
Glenn R. Howes <grhowes@students.wisc.edu> wrote:
>As for a more fully featured Kermit tool, a) there are
>at least 4 commercial tools of varying functionality, b) the
>freely available one is good enough for most people, c) most
>freeware communications programmers seem to be concentrating
>on TCP/IP applications. If any young programmer is reading,
>writing one will take about 3 months of weekends, buy
>Frank's book "Kermit: A File Transfer Protocol", get the
>current specs from their ftp site at Columbia, invest
>in "Inside the Macintosh Communication Toolbox" and get the
>specs on CTB 1.1 which should be on the apple ftp site. Also,
>don't write it by cutting and pasting, write it from scratch
>using the protocol specs, you'll be happy you did later.
>
Perhaps. But the real Kermit protocol engine is already inside
Mac Kermit. 80% of a real VT320 emulator is there too. 90%
of the Kermit script language is there too. The character-set
translation is there too. I can see why people who live in
the Macintosh universe would want something less monolithic and
more modular, but that is putting a rather fine point on it.
We already have something that almost works, and that shares
code with literally hundreds of other Kermit implementations
and an incredibly diverse range of platforms. The point being
that if separate Kermit tools, translation tools, VTxxx tools,
etc, were constructed, they would immediately become orphans.
There is one and only one shared nucleus of common code. We
simply can't afford to maintain lots of code bases.
Anybody in the Macintosh programming community who would like
to see Macintosh Kermit reach fruition, by whatever means, is
encouraged to take up where our last generation of volunteers
left off.
>Then base it around OpenDoc which is extremely cross platform
>and will protect us from the code bloat of Microsoft and OLE.
>
More "standards"... I have to admit I have not even looked at
OpenDoc. Why should I? Without even knowing what it is, let me
take a guess: it is some new "standard" put forward by yet another
"consortium" of "major players" which is so complex that the only
way anyone could incorporate "compliance" with it into one's
applications would be by licensing proprietary libraries and
bloated development environments, and which, no matter how
portable it is said to be, will not come close to covering the
400+ platforms covered by C-Kermit, the code base for Macintosh
Kermit. And which probably requires megabytes of memory and disk,
etc etc. Maybe I'm jaded, but don't these "standards" pop up and
then fall by the wayside on a yearly basis? Am I the only one who
sees the entire software industry grinding to a halt as it chases
the holy grail of the latest self-proclaimed "standard" from
Microsoft, Apple, IBM, or various consortia that fill the trade
publications with glorious proclamations of future cooperation and
openness, only to melt away before anybody has noticed?
>Otherwise, look at NetScape, which seems to be perfectly at
>home whether it is in Windows, MacOS, or X.
>
Of course, we are looking very carefully at the successful
cross-platform apps.
>Maybe that is your problem, no Mac programmer worth his bits
>wants to write a program that looks like it would be at home
>in DOS land.
>
Mac Kermit might have a somewhat crude user interface, but from
the very beginning it has always been a true Macintosh point-
and-click interface. Recently we added on C-Kermit's command-line
interface in an optional "command window". Why? Primarily for
scripting and dialing. Not only can script programs now be run in
Mac Kermit, but they are the very same script programs that run in
UNIX, VMS, OS/2, AOS/VS, VOS, OS-9, and on the Amiga and the Atari
ST, and to a large extent also DOS and Windows. Thousands of
sites have a huge investment in their installed base of Kermit
scripts, and Mac Kermit's command-line interface "leverages" (as
they say) that investment to the large installed base of Macs.
That's not to say that Mac Kermit's graphical interface can or
should not be improved. It needs a LOT of work. Most of the work
that is needed could be done by a competent Mac programmer in a
weekend. But we have not had a competent Mac programmer working
on the project in several years.
There is also a handful of show-stopper bugs in Mac Kermit, of the
sort that could only be found and fixed by a good Mac programmer,
one who is familiar with the quirks of the many Mac OS releases,
hardware platforms, and ROMs.
So the situation seems to be: countless thousands of people at
thousands of sites WANT Mac Kermit to reach release level, but no
Macintosh programmer will to touch it because it is not "pure".
>Unfortunately, a lot of the software I unwittingly buy is
>built with just this mentality. [coding to the lowest common
>denominator]
>
Portability is not a "mentality" -- it's the way we provide code
that works on more platforms than, probably, any other nontrivial
program on earth. A good Mac programmer, such as yourself, who
took on Mac Kermit would no doubt do a bang-up job, using all the
available tools and features in such a way as not to compromise
the quality for Mac users, but still maintaining the portability
of the protocol and other common modules to the other platforms.
- Frank