Human-Nets-Request%rutgers@brl-bmd.UUCP (07/27/83)
HUMAN-NETS Digest Monday, 25 Jul 1983 Volume 6 : Issue 37
Today's Topics:
Query - Is Technology Worth It?,
Technology - Re: Text and Sound for Messages,
Computers and People - Personal Information Systems (3 msgs)
----------------------------------------------------------------------
Date: 25 Jul 1983 10:54:17 EDT (Monday)
From: Erik Sherman <esherman@BBN-UNIX>
Subject: Civilizing technology?
The debate over the usefulness of WorldNet has uncovered a question.
Has any technological development fundamentally changed men and women
for the better? If so, what was the development, what was the change,
and how can you demonstrate the change? If not, why is further
technology desirable in and of itself?
Erik Sherman (ESHERMAN@BBN-UNIX)
------------------------------
Date: 24 Jul 83 11:37-EST (Sun)
From: Steven Gutfreund <gutfreund.umass-cs@UDel-Relay>
Subject: Secretaries and Managers
There was a note in here a few days ago about a manager who preffered
to use his secretary as a reminder/calendar system than his terminal.
There was a conclusion drawn that he had some sort of "secretarial
dependency disease" or computer xenophobia.
I would put it to you that the real reason is a much more common one
in office situations: blame and responsibility. If the secretary
forgets to remind him, there is someone to blame, if he forgets to
look at his terminal, he can only blame himself.
Covering your ass is a very common office politics trick. Furthermore,
some people perfer to trust thinking people who have an understanding
of the importance of the reminder, than a dumb machine.
In the long run, until your machines become well rounded
psychological substitues for secretaries, you will have an uphill
battle.
- Steven Gutfreund
------------------------------
Date: 22 July 1983 12:45 EDT
From: Robert Elton Maas <REM @ MIT-MC>
Subject: Fear and loathing of personal information systems - oracle?
Date: Thu, 21 Jul 83 12:01 EDT
From: MJackson.Wbst@PARC-MAXC.ARPA
Given a wide
data base, one would expect that even simple queries would yield
more than one response; would not the multiplicity of "answers"
tend to educate the user toward a more inquiring (not to say
skeptical) attitude? I am assuming that the information-retrieval
mechanism would not give "oracular" responses but rather a series
of pointers into relevant files.
This depends on the type of system (fully-automatic such as Dialog or
Mycin, fully-human with computer merely being a communications medium
such as HUMAN-NETS, or truly hybrid systems such as computerized
conferencing or Hypertext or Generalized Computer Dating). Early
fully-automatic systems for general info retrieval would have to be
like Dialog, having no capability for understanding English or other
natural language text in enough detail to synthesize an answer. They
would indeed give just a list of citations, and hopefully present the
text onine rather than require you to order microfiche like Dialog
currently does (either you order fiche, or spend time looking thru
local libraries, or you don't see the full text at all). But advanced
systems should be able to collate the available answers to your
question and construct a summary such as "most sources indicate
Einstein's general relativity is valid, but some alternate theories
have been proposed and not yet refuted" which indicates the
most-likely-correct answer but also indicates the margin for doubt.
This relieves the user of having to read all the cited articles and
construct the summary in hir mind. The citations would of course be
available if the summary wasn't sufficient for the user's needs.
Mostly-human systems would suffer the opposite problem currently.
Whoever the "expert" is, everybody believes that person. Like if
Lauren Weinstein says the quality of CBS Teletext in Los Angeles is
shoddy, everybody takes that as fact. Even if somebody else says it
was good, Lauren is believed. Typically there won't be enough experts
to have a true difference of opinion that the user will believe,
rather the one expert will be believed absolutely, and without a way
to check the expert's alleged facts, well what can you do? With
systems that direct you to an expert, rather than posting your query
on a whole mailing list of random people, you'll get only one reply
and there'll be no room for alternative views, so this effect of
believing the expert will be even worse. But eventually with good
systems that send your query to more than one expert and which allow
you to look up references, this problem will be alleviated.
Eventually the two systems will converge. You'll get a summary,
written by an expert or a computer, you sometimes won't know which;
and you'll get a list of references, facts and expert opinions cited
by the summarizer entity (expert-human or computer).
By the way, I would not like a system that ALWAYS replied with two
opposing answers, one from respected scientists and one from
flatworlders or occultists for example. This "point/counterpoint"
method of disseminating knowledge, such as on various TV programs and
in the ballot proposition booklet in California, usually results in
two extreme views, neither of which is correct. I'd rather have one
generally-accepted answer with alternative views listed as secondary.
Example of query:
Does vitamin C cure the common cold?
Example of point/counterpoint answer:
Yes it does -- Linus Pauling
No it doesn't -- AMA
Example of what I'd prefer:
The question is hotly debated, but some general conclusions seem
warranted. Vitamin C strengthens the membranes causing lessening of
symptoms, but doesn't totally stop the virus. The extent to which it
lessens symptoms is still up in the air, from hardly at all to very
much. <citations to various experiments and opinions and arguments>
Example of query:
Is the Earth flat?
Example of point/counterpoint:
No, spaceflight photos show clearly the Earth is round - NASA
Yes, spaceflight is staged in Disney studios - flatworld society
Example of what I'd prefer:
The question is firmly decided in the negative. The Earth is a
nearly-spherical body orbiting the Sun. The Earth is so large (7600
miles in diameter) that to a person standing on its surface it appears
flat. There is one group claiming the Earth is flat, but other than
claiming all evidence of roundness is falsified, they have no valid
point. <citations to flatworld society claims, and citations to
massive list of experiements and practical matters that depend on
round Earth and to list of experts to all agree it's round>
------------------------------
Date: Mon, 25 Jul 83 09:34 EDT
From: MJackson.Wbst@PARC-MAXC.ARPA
Subject: Re: Fear and loathing of personal information systems -
Subject: oracle?
It seems to me that after the two systems converge (dumb computer and
mostly-human) you have the same problem as with the mostly-human
system. The list of references, facts, and expert opinions cited will
be selected and summarized by one "expert," although in this case it
may be a computer. I'm a bit dubious of the proposition that the
computer expert is, a priori, less subject to bias (whatever that is)
than the human.
Since most queries are not either/or, suppression of a
minority/unfavored viewpoint can be pervasive while remaining
relatively subtle.
Example of query:
What are the effects of Vitamin C on humans?
Example of response from AMANet:
Vitamin C is necessary for life. In inadequate amounts. .
.<extensive discussion of deficiency diseases, but no mention
of the common cold>
Example of response from Pauling InfoService:
A major effect of Vitamin C, given in adequate quantities, is
the suppression of the common cold. . .
The solution that comes to mind is to accomodate multiple "expert"
services on the net, so that the user could shop around and sample the
prejudices of several. Of course, just as many persons read only the
magazines that reflect their personal viewpoint, some users would
focus solely on the expert service that told them what they wanted to
hear.
Example of query:
Do UFOs exist?
Example of response from a popular, hence profitable, expert service:
UFOs definitely exist; visits from space creatures are
well-documented, but the evidence has been systematically
suppressed by the Air Force. <citations to bogus reports of
saucer rides, romantic encounters with aliens, and anti-USAF
diatribes>
-- National Enqiry Service Executive
The drawbacks are obvious--but I see no acceptable alternative; other
options lead to objectionable pruning or have excessive potential for
abuse. Of course, some of us make a point of reading literature from
"the other side" even when we hold strong opinions. And on-line
access to sources tends to speed up the task of uncovering
falsification, misstatement, and questionable interpretations. This
kind of diversity would help keep the system as a whole from being
viewed as an oracle--defending against (but hardly eliminating) some
of Ron's major concerns.
"Objectivity" tends to be a myth; I suspect the best one can do is try
to put the options, and the tools to evaluate them, in the hands of
the users.
Mark
------------------------------
Date: 24 July 1983 17:31 EDT
From: Zigurd R. Mednieks <ZRM @ MIT-MC>
Subject: "You will be asked to leave the future immediately."
We already have the means to study what will happen when some people
cannot use modern tools. The modern tool I'm refering to is the
library. Dynabook is to the library what a Vic20 is to an 1130, it's
something you can cart around with you and is much easier to use.
The people who can't use libraries can afford to use what is free.
Money and our economic system are not the problem. The problem is that
even today there are large numbers of people who just cannot read and
unless some dictator decrees that all illiterates be shot, the problem
won't go away. But it will get worse: I was listening to NPR news
recently, not the sort of news show that often admits that there are
problems that more public spending won't cure. The feature I was
listening to was about a job placement program. What nearly made me
gag on my oatmeal was the casual remark that about half the people in
the placement program had "reading deficiencies". When you work every
day at a job where if you just apply yourself a bit more diligently
you'll find that bug, it becomes difficult to accept that fact that
there are problems without solutions.
What is even more depressing is that being illiterate before the
printing press was invented was a common condition, before the recent
explosion of technology it didn't mean you could not make a living as,
say, a laborer, but now that human muscle is a vanishingly small part
of what creates wealth, illiteracy can be more crippling than
blindness.
Solutions? Not from me. Perhaps we should just let the welfare state
mentality take over. Let's give then bread and circus, we can afford
it. Let's just whoop it up 'till the barbarians invade.
Cheers,
Zig
P.S. Isn't it great how "Cheers" just expresses the right thing every
time? Thank you Roger Duffey, wherever you are.
------------------------------
End of HUMAN-NETS Digest
************************