Pleasant@Rutgers.arpa (01/20/83)
HUMAN-NETS Digest Thursday, 20 Jan 1983 Volume 6 : Issue 3
Today's Topics:
Administrivia - TCP Where Are You?
Technology - WorldNet (4 msgs),
Programming - Unix (3 msgs)
----------------------------------------------------------------------
Date: 17 Jan 1983 2307-EST
From: Mel <Pleasant@RUTGERS>
Subject: Administrivia
Hi folks, just wanted to let everyone know that the Human-Nets
digest has made it through the TCP changeover. There are a few
things everyone should know. Persons running on systems that now
support the TCP protocol should continue to use the addresses
HUMAN-NETS@RUTGERS for submissions to the digest and
HUMAN-NETS-REQUEST@RUTGERS for administrative requests. If your
site is still using the old NCP protocol, you'll have to send
messages to me through an NCP-to-TCP gateway. Submissions should be
mailed to HUMAN-NETS%RUTGERS@ECLC - administrative requests to
HUMAN-NETS-REQUEST%RUTGERS@ECLC. During the changeover, some of you
might not have gotten a digest or two. The last issue in December
was Volume 5 Issue 113. If you've missed any issues, just drop me a
note and I'll get those issues to you promptly.
-Mel
------------------------------
Date: 31 December 1982 04:00-EST
From: Jerry E. Pournelle <POURNE @ MIT-MC>
Subject: The changing face of Micro-computing...
I make a living writing about that subject (actually, I don't; I
make a living writing science fiction. But I do get paid pretty
well to write about the future of micros) so I won't spend a lot of
time on this, but:
The notion that "It's too late for the parents" is goofy.
Bill and Sibyl Grieb have packed classes at everywoman's Village on
using computers; they teach CP/M and customization and all that.
The problem with adult learners are the ones I had: no one seems to
know how to explain things in English. You have to learn a lot more
than you really need to in order to be able to do much of anything.
Some of us, though, are trying to change that, and a few of us are
not only doing something about it, but getting paid to.
Patience: it took far longer for the "average citizen" to
learn enough mechanics to be able to be comfortable using cars than
it is taking for people to get used to computers. The High Priest
mentality in which one accepts whatever a highly paid computer
technician tells you, is dying away in industry already, and the
micro world ain't going to let it get a foot hold...
------------------------------
Date: Sunday, 2 January 1983, 22:48-EST
From: David Vinayak Wallace <Gumby at MIT-OZ>
Subject: The changing face of Micro-computing...
Date: 31 December 1982 04:00-EST
From: Jerry E. Pournelle <POURNE @ MIT-MC>
(I re-ordered some of your statements)
The notion that "It's too late for the parents" is
goofy. Bill and Sibyl Grieb have packed classes at everywoman's
Village on using computers; they teach CP/M and customization
and all that.
Patience: it took far longer for the "average citizen"
to learn enough mechanics to be able to be comfortable using
cars than it is taking for people to get used to computers.
I think this example shows why it IS too late. You can always find a
few exceptions to the rule; these exceptions are a tiny (>.5%) part
of the actual computer user. I doesn't take mechanical knowledge to
drive a car, it takes mechanical knowledge to modify or repair an
automobile. most people (me included) would rather buy a vanilla
car and just have it work all the time. I change the oil (more than
most people do), but have a mechanic do the real work.
The problem with adult learners are the ones I had: no one seems
to know how to explain things in English. You have to learn a
lot more than you really need to in order to be able to do much
of anything. Some of us, though, are trying to change that, and
a few of us are not only doing something about it, but getting
paid to.
The High Priest mentality in which one accepts whatever a highly
paid computer technician tells you, is dying away in industry
already, and the micro world ain't going to let it get a foot
hold...
I'm biased, but I don't think this will really work. You can't
discuss complex concepts without the proper language. I agree that
there is a bit of High Priest mentality and that there is no good
effort to teach the JARGON, but every "normal language" explanation
of anything having to do with computers comes out muddled,
long-winded, and ultimately, unclear..
Some of us even do it for free.
david
------------------------------
Date: 2 Jan 1983 1738-PST
From: GRANGER.RS%UCI@UDEL-RELAY
Subject: WorldNet Services
I'm a kind of new-kid-on-the block and need to get knowledgeable
about Arpa, WorldNet, and all the neatsy-sounding things you folks
are always talking about on here...so could anyone recommend and
tell me how to get hold of some kind of background document on
everything? Is there something on line on one of these hosts? I
would sure appreciate it.
I have some ideas about services which could be classified in the
pipe-dream category, but would like to withhold them until I gain a
bit more knowledge about the proposed goings-on.
Thank you,
Rob Segelbaum
------------------------------
From: "KENNETH G. GOUTAL at ELMO c/o" <DEC-HNT at DEC-MARLBORO>
Date: 9 Jan 1983 1503-EST
Subject: WorldNet services poll
Ref: Robert Maas in HUMAN-NETS #5:112
"Consumer information exchange"? Gak! The lawyers will get richer
on that one! Consider all the gut reactions and undocumented
experiences that people would be committing to 'paper', right out in
front of God and everybody; lawsuit city, here we come! A great
idea, though, if we could avoid that problem. I'm not really sure
where I'm getting this feeling, but seems as though it's pretty
dangerous to say what you think about a product (or person or
company) in public. Naturally, if people stick to facts, they'll
probably be okay, but the net as it has been evolving does not seem
to engender sticking to facts -- people happily flame on about
anything and anyone, as if they were sitting in the privacy of their
dens (which they are), talking with friends (which they are).
Unfortunately (?), the network is a glass house. Have I got the
legal picture all wrong? If not, how do we get there from here?
Any legal types reading this?
(Yes, I know, this isn't really WorldNet, but this is where it's
starting.)
(Say, Robert Maas, whatcha doing out at Stanford? I thought you
were at MIT-MC.)
--Kenn
------------------------------
Date: 11 January 1983 12:10-EST
From: Gail Zacharias <GZ @ MIT-MC>
Subject: VAX VMS vs UNIX
As far as I am concerned, the major difference between Unix and VMS
is that there are Unix simulators for VMS, and no VMS simulator for
Unix. So if you get VMS, you can take advantage of the features of
either system, and run programs developed for either one.
------------------------------
Date: 11 Jan 1983 2041-EST
From: ZALESKI at RU-GREEN at RUTGERS (Mike Zaleski [Secular Humanist])
Subject: UNIX and the Rest
Re: Comparisons of UNIX and Tops-20
I read the comparisons of UNIX and various other operating systems
on Human-Nets with some interest and would like to add my own
thoughts.
First, I am not sure comparisons of this sort are meaningful. I
view UNIX more as an environment. The OS, shell (command
interpreter), compilers, and tools all interface together in a
mostly nice way. My experience with Tops-20 at two different sites
is that this is not completely true under Tops-20. However, the
fault (if one is to be laid) is that most sites have a variety of
non-Dec standard tools which, not surprisingly, do not always know
about each other.
Second, UNIX differs from other systems I've used in the kinds of
system programs it includes. Programs like "grep" or "find" could
be written for other machines, yet I only see them on UNIX. The
lack of these tools on other machines can be gotten around (like
with PCLs under Tops-20), but the task then becomes troublesome.
(Example: How can you do a directory listing of files with one
specific protection under Tops-20? UNIX provides a number of tools
which make this easy.)
Third, the ever popular UNIX I/O redirection is also something which
can mostly be done with little extra programming effort on other
machines. True, there is a difference between little effort and no
effort, but I have yet to meet a machine which constrained its
output to the terminal. The only aspect of UNIX I/O which I have
not encountered on any other machine is ability to set up
communications "pipes" between two programs. This allows, for
example, the trivial implementation of an Emacs command to run a
program (like a compiler) and save the output (error messages) in a
buffer. (If anyone knows how to do this under Tops-20, please send
me mail about it!)
Fourth, one aspect of UNIX that I do like and which has not
previously been mentioned is that there is not a ton of expensive
documentation. Standard 5.0 UNIX documentation consists of 3
manuals (one of which is a collection of console messages unneeded
by the average user), and a two volume set of "Documents for use
with UNIX" (consisting of memos on the various more sophisticated
UNIX tools). The C book is also handy.
Fifth, regarding the initial-state "problem" with UNIX. At least
UNIX lets the user do something about it, like resetting the erase
and kill characters. This can be done automatically at login.
Tops-20 does not provide any such facility that I know of. By the
way, there are various "editing shells" for UNIX floating around
which would (or should allow) resetting any function to any key.
This, however, is not part of standard UNIX.
Incidentally, UNIX system managers can (and do) custom configure
their systems. There is a file in /etc which is executed on login
time. I am actually quite happy that when I log onto an alien
system, at least I know that the erase and kill characters will be
where I expect them (even if I don't like that location). All
things considered, I don't consider this initial state thing a big
deal.
Some quick thoughts on other UNIX items:
- The Control-D logout is, if I am correct, going to be changed in
an upcoming UNIX release.
- Not all sites have the source code online, contrary to what one
message might have implied.
- Performance of UNIX vs VMS vs ... I'm sure a test can be
generated to "prove" it either way. Supposedly, 5.0 UNIX shows
improvement in this area.
- Berkeley UNIX and Bell Labs UNIX are not identical. When reading
a message from someone flaming about or raving about some UNIX
feature, remember that it may not be Bell (or Berkeley) UNIX.
- To the person who asked for a comparison of VAX VMS and UNIX.
This question could take volumes. One suggestion: Attend a DECUS
conference and look for the UNIX users and VMS users and talk to
them. (My personal recommendation is UNIX.)
UNIX is a trademark of Bell Laboratories. DECUS, VAX, and Tops-20
are trademarks of Digital Equipment Co. (Keep those lawyers happy.)
(Personal opinions of Michael Zaleski, UNIX Systems Development
Dept, Bell Labs, Murray Hill.)
-- Mike
------------------------------
Date: 12 January 1983 05:44-EST
From: Robert Elton Maas <REM @ MIT-MC>
Subject: Re: Unix Initial State / EMACS initial state bad??
I find that the initial state of EMACS is quite usable. With the
freeze-file-version and compare-windows added, it's almost
wonderful, the only major pain being rubout (the most common noise
character on dialup lines) being non-undoable delete-backwards,
requiring customization in a few input characters. Keyboard macros
are very nice and I don't consider them to be system customization
since you can do them at any time without knowing anything about the
internals, and they work completely at the source (keyboard input in
this case) level.
I've used EMACS only here on ITS and there (IMSSS) on TENEX which
doesn't have freeze-file-version (too old a version, it's standard
now) nor compare-windows (I miss it every so often there). When I
refer to initial state I refer to the TENEX version since it has
less than the ITS version, doesn't seem to have any of the extra
packages that have become standard here.
------------------------------
End of HUMAN-NETS Digest
************************