Human-Nets-Request%rutgers@brl-bmd.UUCP (07/21/83)
HUMAN-NETS Digest Monday, 18 Jul 1983 Volume 6 : Issue 34
Today's Topics:
Reply to Query - Command Syntax(es?),
Computers and the Law - property rights for stored data
Technology - text and sound for messages (2 msgs)
Computers and People - Personal Information Systems
----------------------------------------------------------------------
Date: 15 Jul 83 01:12:04 EDT (Fri)
From: Mark Weiser <mark.umcp-cs@UDel-Relay>
Subject: Re: Command Syntax(es?)
As to whether the syntax should match the system or the tool:
its basically a matter of analyzing the user population. If there
are features of the tool which use syntax which everyone using the
system must know (like how to backspace a character), or if the
tool will (can) only be used by people who are pretty familiar with
the system, then the tool should conform to the system.
On the other hand, if the tool is one which someone might only
use the system for, and not be familiar with any other aspects of
the system, then the tool can afford to offer uniform syntax
regardless of the system. This pretty much means that the tool
must supply a complete operating environment. For instance,
apl and emacs environments ought to be the same everywhere, because
they insulate you from you operating system and because
you might use the machine for only them. A compiler or a mailer
would be different (not so sure about the mailer).
------------------------------
Date: 14 July 1983 03:06 EDT
From: Steven A. Swernofsky <SASW @ MIT-MC>
Subject: property rights in stored data
Chip,
Do you have any more info on this topic? Is the legislature insane,
or is this an attempt to foist the problem off on the courts? Do you
know the story of the Mass. "right to privacy" bill?
-- Steve
------------------------------
Date: 14 July 1983 10:21 EDT
From: Robert Elton Maas <REM @ MIT-MC>
Subject: text and sound in the network?
Date: 14 July 1983 03:15 EDT
From: Steven A. Swernofsky <SASW @ MIT-MC>
It's silly to use sound as a backup for text. Just send the sound
directly!
If infinite data storage were free so the sound could be kept forever,
and if listening to sound were as efficient as reading text (remember
people can read or skim-read at over 700 wpm while they can talk at
only 200 wpm), then your statement would be valid. But for permanent
storage of memos and documents voice is too expensive and slow to
access. Besides, if you never convert it to text or other abstract
form you can't index it by keywords for later access by subject.
It's also silly to require a secretary-proofreader. Just display
the text for the speaker as he speaks, and allow him to be his own
proofreader. Adding a secretary just makes for another error in
theloop.
I agree, except that a lot of executives think their time is too
valuable to waste doing that so they have secretaries do it for them.
(My boss is an example. For a simple reminder message to himself he
calls his secretary in for dictation instead of sending himself a
message on his terminal.)
------------------------------
Date: 15 July 1983 23:01 EDT
From: Steven A. Swernofsky <SASW @ MIT-MC>
Subject: text and sound in the network?
Robert,
[SASW]
It's silly to use sound as a backup for text. Just send the
sound directly!
[REM]
If infinite data storage were free so the sound could be kept
forever,. . ., then your statement would be valid. But for
permanent storage of memos and documents voice is too expensive
and slow to access.
I was thinking of short memos or phone messages. (1) Short messages
of this kind are not usually kept forever. (2) Short messages don't
take very much storage space. Speech research has shown that the
amount of storage required is not so huge anyway. (3) Long documents
would be transcribed to text for review and revision before they were
sent anyway.
Besides, if you never convert it to text or other abstract
form you can't index it by keywords for later access by subject.
So what? Short messages (as noted) aren't usually saved and aren't
usually worthwhile to access by keywords anyway. I just find that
they get in the way.
Another good reason to send the sound directly is that inflection and
other nuances of speech are difficult (if not impossible) to transmit
via text.
It's also silly to require a secretary-proofreader. Just
display the text for the speaker as he speaks, and allow him
to be his own proofreader. Adding a secretary just makes for
another error in the loop.
I agree, except that a lot of executives think their time is too
valuable to waste doing that so they have secretaries do it for
them.
This is the wrong way to think about the topic. Provide the
capability and SOME people will take advantage of it. When those
people find it to be convenient, they will spread the word and it will
become more common. Planning for prejudice to last forever is also
silly.
A short note on execs vs secys -- it's LOTS cheaper for my secretary
to do dull work like this than for me to. I had in mind the situation
where the spoken version was too ambiguous to transcribe, not the
typical proofreading situation. In the ambiguous-verbiage case, the
exec will have to clear up the text somehow anyway (as by telling his
secretary what to insert in place of the meaningless drivel he
caused). So having him do it immediately isn't too unreasonable.
-- Steve
------------------------------
Date: 18 Jul 83 14:41:56 EDT
From: Ron <FISCHER@RUTGERS.ARPA>
Subject: Fear and loathing of personal information systems
I am trying to consider the consequences of widespread use of personal
information systems. Imagine a large segment of the world population
owning the rough equivalent of Alan Kaye's (sp?) Dynabook, a portable
information manipulation device about the size of a notebook. It
would include every spiffy accouterment known to computer science.
Imagine them interconnected by the infamous "Worldnet" concept so
often lauded here.
Here we have a machine that can hold encyclopedia, technical manuals,
histories, images, sounds, and communicate them to anyone else with a
similar machine, etc.
Aside from the question of what happens to those who cannot own such a
device...
My basic concern was: with such ready access to information would our
desire to ask questions be stifled? Knowing that an answer is only a
keypress away might keep one from making that movement, the same way
that having reference texts on a shelf lulls one into a false sense of
"defacto understanding."
At another end of the spectrum: might academia fall into the trap of
rearguing each other's theses indefinately, with a significant
percentage no longer doing original or empirical work? Redundancy is
a healthy thing in research. By asking the same question again new
viewpoints may be discovered. By challenging existing ideas, directly
or indirectly, we come upon new truths.
Are there limits to this media that might have an impact on our way of
thinking? Might limits be imposed that would somehow do this? The
latter is something of a horror scenario and may be ignored by the
sensitive respondent... :-)
Further: might we become conceptually "xenophobic?" I.e. afraid of
things that our little boxes could not explain?
Please consider carefully. I believe that this is a deeper problem
than "meets the eye" and also has a serious component of reality.
I worry that easy access to incomplete systems of information may
cause a television-like stupor to descend upon us, ultimately
decreasing our involvment and interaction with knowledge.
(ron)
------------------------------
End of HUMAN-NETS Digest
************************