human-nets@ucbvax.ARPA (12/12/84)
From: Charles McGrew (The Moderator) <Human-Nets-Request@Rutgers>
HUMAN-NETS Digest Wednesday, 12 Dec 1984 Volume 7 : Issue 81
Today's Topics:
Computers and People - Compact Disks vs. Books (3 msgs),
Computer Networks - Notification of Mail Delivery,
Computers in the Media - Pilot for New PC show?
----------------------------------------------------------------------
Date: Fri, 7 Dec 1984 17:02 EST
From: Jim Aspnes <ASP%MIT-OZ@MIT-MC.ARPA>
To: Wayne McGuire <MDC.WAYNE%MIT-OZ@MIT-MC.ARPA>
Cc: AIList@MIT-MC, asp%MIT-OZ@MIT-MC.ARPA, zbbs%MIT-OZ@MIT-MC.ARPA
Subject: Don't capture my soul in your little silver box
Ah, yes. The ultimate record; the convenience of having one's every
written thought, shopping list, and love letter archived away with an
army of daemonic librarians to call them back. It eliminates so many
of the problems of human memory: vagueness, forgetting, the "creative"
aspects of recall. But among these it eliminates a most important
feature. Privacy.
There has not yet been a government which has not chosen at some time
to pilfer the files of dissidents, journalists, opponents, and/or
writers in order to expand its power or silence opposition. Consider
now our 100-disk, world-hopping information superman. It's 1995, and
his conscience forces him to join a possibly-socialist group
protesting, say, a nuclear freeze or equal rights for gays. He comes
home one afternoon to his study, finds his archives gone, and a note
on his desk saying simply: "We've got a file on you that fills 100
optical disks. Watch your step."
This is not to say that archives would not be useful, nor that the
Feds can't raid your space-consuming pen-and-paper file cabinets. I
just hope that the optical-disk sorceror's apprentice is rigged with
unbreakable encryption (and it had better still be unbreakable thirty
years from now), and is encased in thermite. And even then I'd still
put my money on the Thought Police.
It's sad that computers, which have so much potential, have so much of
it invested in the purposes of the authorities. I wonder if some day
we'll be looking back at what we did in the 1980's the way many atomic
physicists ended up remembering the 1930's.
Jim Aspnes (asp%mit-oz@mit-mc)
------------------------------
Date: Mon 10 Dec 84 00:29:18-EST
From: Wayne McGuire <MDC.WAYNE%MIT-OZ@MIT-MC.ARPA>
Subject: Optical Disks Vs. Paper Media
To: zbbs%MIT-OZ@MIT-MC.ARPA
It seems certain that the optical disk will heavily cut into the
use of traditional books and other paper media, such as newspapers and
magazines.
Imagine if we had a portable machine somewhat along the lines of
the Data General/One, but smaller, lighter, and with a much crisper
and more readable visual display. Imagine also that this machine
possesses 3MB of RAM, a 3 1/2" 3MB micro-floppy disk drive, a 3 1/2"
20MB micro-hard disk drive, and a 2" .5GB optical disk drive (IBM is
rumored to be working currently on 2" optical disks). It is not
far-fetched to think that a machine with this general range of
specifications will be available within a very few years.
Imagine then that, on a train or at the beach or anywhere, one
could easily insert into this machine a 2" optical disk containing
_War and Peace_, the last month of _The New York Times_, or the last
ten years of the journal _Artificial Intelligence_. (It would be an
easy matter to carry around 20 or 30 of these tiny disks in a
briefcase.) Imagine also that this portable system for reading texts
was fitted with the following capabilities:
1) One could easily search the full-text of large documents, by
complex combinations of terms or concepts, to pinpoint specific
passages of interest.
2) One could easily copy passages of interest from optical disk,
to floppy disk, and later to a master archival optical disk at one's
home base. These passages would be automatically documented with full
bibliographical data.
3) One could easily annotate passages in the optical disk
read-only text. This system might use a split screen: in the upper
half of the screen would appear the original text (let's say James
Joyce's _Ulysses_), and in the lower half one could compose and edit
an annotation of unlimited size. The precise passage annotated (even
just a phrase or word or diacritical symbol) would be marked in the
upper screen, and linked by a unique identifier to the annotation.
The annotations would perhaps be stored on free space on the same disk
which holds the original read-only text. Later, when rereading
_Ulysses_, one would have the option of reading an unmarked cleantext
of Joyce's epic, or a mode which displayed (perhaps in a margin or
between the lines) pointers to all the annotations for given passages.
One could then split the screen, and call up previous annotations in
the lower half. (I suppose there is no reason why one could not also
annotate previous annotations, and annotations to previous
annotations, and so on ad finitum.) With this method the reader could
handily record and retrieve and analyze all his or her thoughts about
a text like _Ulysses_ over an entire lifetime.
One argument for paper media is that they are more portable than
computers, but is that really the case? Scanning, annotating, and
copying passages from ten volumes of a journal on a train, in a
library, or even at home would certainly be much easier and more
efficient with a Data General/One-type machine and a 2" optical disk
than with the hardcopy original. Even reading a single newspaper or
paperback book, especially if one is in the habit of making
annotations, could be more convenient with an optical dynabook than
with a paper copy.
Before the end of this century optical disk-based notebook-size
portable computers will probably replace traditional paper documents
as the medium of choice for knowledge workers and perhaps the general
public as well. (The great majority of the general public will
probably be knowledge workers in any case.) The advantages of the
optical disk over paper media are overwhelming.
-- Wayne McGuire (wayne%mit-oz@mit-mc)
------------------------------
Date: Sun 9 Dec 84 23:11:51-PST
From: Ken Laws <Laws@SRI-AI.ARPA>
Subject: Re: Optical Disks Vs. Paper Media
To: MDC.WAYNE%MIT-OZ@MIT-MC.ARPA
I personally doubt that there will be many knowledge workers
in, say, another century. Most people will be employed in
social work and service industries, if they are employed at
all. (See Nils Nilsson's AI Magazine article on economics for
the alternative, namely that machines work and people receive
investment income.)
The purpose of knowledge work is to extract useful information from
knowledge streams in which it is implicit or scattered. This is
currently done by humans because 1) we don't have good algorithms for
extracting "obvious" patterns in naturally occurring data, and 2) we
don't have sufficiently powerful parsers and knowledge representations
to decode human-generated knowledge streams. The first problem will
be solved by the development of better knowledge-based pattern-
recognition algorithms. The second will be solved by both better
parsing tools and by constrained data generation so that knowledge
streams contain more formalized knowledge. Progress in the first
task will mean fewer human-generated knowledge streams, simplifying
the second task.
There may come a time when the only knowledge work left for people
is that for which automation is not economically justified, e.g.,
the analysis of ancient texts. That era will quickly fade as
frustrated "knowledge workers" with powerful tools and little to do
take up the automation of other people's hobbies as their own hobby.
-- Ken Laws
------------------------------
Date: 9 Dec 1984 2013-GMT
From: CCD-ARG (on Dundee Tech DEC20) <CCD-ARG%dct@ucl-cs.arpa>
Subject: Automatic notification of mail delivery
The U.K. standard JNT Greybook mail which is based on RFC822 has
defined the additional header field "Acknowledge-to:". Any full spec
mail system in the U.K. academic community should return an
acknowledgment back to any specified receiver when it either places
the message in the users mailbox or, in some implementations, when the
recipient actually reads the message. This is fine in theory but in
practice many mail servers just ignore this field !
Here's the relevant section from the greybook protocol document.
========================
5.1. Delivery Notification
In exceptional circumstances, a user may require positive confirmation
that a message has been delivered. This will be indicated by the
insertion of an optional field:
"Acknowledge-To" ":" mailbox
; address to
; acknowledge to
It is recommended that the header also contain a Message-ID: field.
The server will then generate an acknowledgement message to be sent to
the address specified. To determine the full source route to be used,
the guidelines of appendix E on the handling of trace information
should be followed. This address should not be used for error
notification or for any other purpose. An implementation should take
steps to prevent acknowledgement messages from generating error
notification messages, as discussed in Appendix I.
The acknowledgement is used at a user level, and is not intended
for automated analysis. It should indicate exactly what has happened
to the message (e.g. that the message has been placed where the user
can read it, or that the user has read the message). It may be
appropriate to send more than one acknowledgement message. The text
of the acknowledgement message will contain the address or addresses
being acknowledged, and the value of the Message-ID: field (if
present). Use of a References: field is suggested. Automatic
generation of acknowledgement requests is not recommended, as such
messages are not needed in most cases. It is noted that this field
may cause problems in the area of distribution lists, although this
problem is beyond the scope of this protocol specification.
========================
Put simply, this means that if I send a message to say FRED@DCT from
host DDXA and wish to know if it actually arrives (pretty pointless in
this case as its not involving the message being routed through lots
of gateways) then I would send a message somthing like this:
Date: Sun 9-Dec-84 8:05PM-GMT
From: someone <Someone@DDXA>
To: Fred@DCT
Subject: Testing acknowledge
Message-ID: <Unique-Garbage-1234567890@DDXA>
Acknowledge-to: someone@DDXA
Message text as usual
If you want to try it for yourself you could try mailing to me at
Alan%DCT@UCL-CS.ARPA with an Acknowledge-to field and see what if
anything you get back ! It should manage to trace a route back to Arpa
sites although its never been tested...
Alan Greig
Computer Centre
Dundee College of Technology
Dundee
Scotland
------------------------------
Date: 12 December 1984 01:12-EST
From: Steven A. Swernofsky <SASW @ MIT-MC>
Subject: [CRISSE: forwarded]
MSG: *MSG 3512
Date: 12/11/84 14:33:35
From: CRISSE at MIT-OZ
Date: 11 Dec 1984 14:36 EST (Tue)
Message-ID: <CRISSE.12070634601.BABYL@MIT-OZ>
From: Crisse Ciro <CRISSE%MIT-OZ@MIT-MC.ARPA>
To: *mac%MIT-OZ@MIT-MC.ARPA
THURSDAY, DECEMBER 13
CHANNEL 2, 9:00PM
"THIS COMPUTER THING"
This is a pilot for what may be a 13 show series concerning personal
computers. The show features Robin Young and our very own Randy
Davis, so if you've a chance, please make a note of it and tune in!
(Profuse apologies to those in AI who have already received this msg.)
------------------------------
End of HUMAN-NETS Digest
************************