From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!mips!pacbell.com!att!linac!mp.cs.niu.edu!rickert Tue Apr  7 23:22:27 EDT 1992
Article 4738 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!mips!pacbell.com!att!linac!mp.cs.niu.edu!rickert
>From: rickert@mp.cs.niu.edu (Neil Rickert)
Subject: Re: Language as Technology: A Phenomenological Study
Message-ID: <1992Mar26.130807.18717@mp.cs.niu.edu>
Organization: Northern Illinois University
References: <1992Mar25.080515.20086@a.cs.okstate.edu> <1992Mar25.185007.21788@mp.cs.niu.edu> <1992Mar25.225555.41966@spss.com>
Date: Thu, 26 Mar 1992 13:08:07 GMT
Lines: 80

In article <1992Mar25.225555.41966@spss.com> markrose@spss.com (Mark Rosenfelder) writes:
>In article <1992Mar25.185007.21788@mp.cs.niu.edu> rickert@mp.cs.niu.edu (Neil Rickert) writes:
>>  The most important aspect of human language is that it is a digital system.
>
>I presume you mean "mostly digital", or "in the aspects which interest me,
>digital"?  Language does have analog features: e.g. stress, pitch, tempo,
>length of pauses, intonation contours.

 I agree those are analog features.  It is not so certain they should be
considered part of language, but if you wish to so treat them you will
surely agree that they are relatively unimportant, given that they do
not show up in printed language, and are not very apparent with some
speakers.

>>  Ideas can be transmitted with body language, physical demonstration,
>>etc.  You don't need language for that.  Digital language allows
>>modelling/simulation, and allows the modelling to be separated from
>>physical artifacts.  In effect this makes abstract ideas possible.  
>
>What's unclear to me in your remarks is why the digital nature of language
>is necessary for modelling or for abstract ideas.  Simulation is often
>just as possible and just as symbolic with analog methods: e.g. you
>could model voltage with water pressure; or you use brightness on a CRT
>to represent some numerical quantity.  And I don't understand the
>relation of abstraction to digitality at all.

 Yes, you can model voltage with water pressure.  It is very expensive to
do so, since it requires the construction of elaborate equipment.  It is
very difficult to achieve more than three or four digits of precision
without using digital methods.

 Imagine that you wanted to plan tomorrow's dinner without digital means.
You might draw some lines in the sand to represent the table.  Find some
large rocks to represent the place settings, some smaller rocks to represent
the meat course, perhaps some twigs to represent the vegetables, etc.  It
becomes an elaborate production.  It is far too much effort to be worth
doing.  It is the digital nature of language which allows you to construct
the model in your mind without depending on physical artifacts.

>>Digital
>>language also greatly enhances memory quality and memory density [compare
>>the density and quality of a conventional audio tape and a DAT].
>
>This seems simply false; digital HDTV, requiring a billion bits per image,
>only becomes efficient relative to analog competitors once a good deal of
>sophisticated data compression has been done.

 I admit I haven't been following HDTV standards very closely.  But lets look
at what can be done with a billion bits per image.  We might use 24 bits
to represent the color.  This allows 16,777,216 distinct colors, which is
surely much more than the human eye can discriminate.  That leaves enough
bits for more than 40 million pixels.  We can construct a picture with
5000 visible scan lines each containing 8000 pictures.

 Assuming a scan rate of 30 images per second, as in conventional TV, an
analog TV with similar resolution circuit will require a band pass in the
video amplifier covering all frequencies from 150 KHz to at least 1 GHz.
I would really like to see your design for a video amplifier which did this
but avoid any resonances or high freqency cut off effects which would degrade
the picture.  For that matter, I would like to see a coax cable capable of
this bandwidth with sufficiently good impedance match accross the full
bandwidth that there is no serious echo problem to degrade the picture.

 But then I was talking about memory, not about display.  How much magnetic
tape will this take compared with the tape requirements for the digital
signal.  Oh, and by the way, how are you intending to design your
recording head so that its magnetic coils handle the full video bandwidth
smoothly without resonance?

 In fact, in trying to give an example where analog does better than
digital you have actually come up with something which is, for all
practical purposes, outside the limits of available analog technology
today.


-- 
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
  Neil W. Rickert, Computer Science               <rickert@cs.niu.edu>
  Northern Illinois Univ.
  DeKalb, IL 60115                                   +1-815-753-6940


