RE: SUO: Ballot Comment - 3D versus 4D.
Matthew,
I think the situation is more subtle here - as we have discussed a number
of times.
I think that the en- view is more suitable for NL work as it reflects NL's
technological origins. E.g. Linguists explaining the way language works
deals with the human brains small input buffer size (not a necessary
problem for silicon brains). This may make it more suitable for *human*
communication - but not necessarily machine communication.
Regards,
Chris
-----Original Message-----
From: West, Matthew R SITI-GREA-UK [SMTP:Matthew.R.West@is.shell.com]
Sent: Tuesday, September 04, 2001 10:53 AM
To: John F. Sowa; Horn, Graham
Cc: 'pat hayes'; standard-upper-ontology@ieee.org
Subject: RE: SUO: Ballot Comment - 3D versus 4D.
Dear John,
If we get as far as a 4D ontology within the SUO I
don't see why this perspective should not be supported.
I would like to comment on this.
> 5. I agree with Pat that the per- descriptions are often more prolix
> than the en- descriptions for many common ways of talking about
> the world. That is partly because people have developed a large
> vocabulary of common expressions that would require very extended
> translations if mapped into the primitives of the per-
> perspective.
En- descriptions are strongly influenced by language, and the primary
use of language is to communicate, rather than to store, knowledge.
Evolution has meant that language has developed to be very efficient
at communication, i.e. the minimum size of exchange to convey what is
required. However, this is based on assuming the maximum amount of
context knowledge on each side. This context knowledge is an essential,
but usually implicit, part of any communication.
My conjecture is that whilst en- is more efficient for communication,
per- will turn out to be more appropriate for storage of the whole (what
is communicated plus the context). This arises from the high level of
reuse that is achieved from the more context neutral nature of per-
descriptions.
Matthew West
Principal Consultant
Shell Information Technology International Limited
Shell Centre, London SE1 7NA, United Kingdom
Tel: +44 20 7934 4490 Other Tel: +44 7796 336538
Email: matthew.r.west@is.shell.com
Internet: http://www.shell.com
> -----Original Message-----
> From: John F. Sowa [mailto:sowa@bestweb.net]
> Sent: 03 September 2001 16:33
> To: Horn, Graham
> Cc: 'pat hayes'; West, Matthew R SITI-GREA-UK;
> standard-upper-ontology@ieee.org
> Subject: Re: SUO: Ballot Comment - 3D versus 4D.
>
>
> Pat, Graham, and Matthew,
>
> I agree with Pat on the general issues, but I have a somewhat
> different way of expressing them. As I have said in some previous
> notes (and my KR book), I prefer Whitehead's 4D representation,
> which accommodates many different ways of describing physical
> entities:
>
> 1. W's basic physical entities are processes (which he calls
> actual occasions or actualities). Each such entity has a
> very short duration (not a zero-lenth instant, but a short
> interval that depends on granularity).
>
> 2. Enduring physical objects are extended processes that are
> characterized by various "forms of definiteness", such as the
> geometrical shape of a crystal or the DNA of a plant or animal.
>
> 3. You can talk about variations over time or over space by looking
> at the first derivatives with respect to time coordinates or space
> coordinates. A slow change has a small continuous derivative,
> and a so-called discrete change has a very large derivative,
> which may be considered a discontinuity.
>
> 4. For both the en- and per- methods of description, the processes
> characterized by "forms of definiteness" are important. They are
> called "physical objects", and they can be described as "changing"
> through time, if one wants to use that way of talking. But the
> underlying actual occasions are identical from every viewpoint,
> and they form the basis for relating one description to another.
>
> 5. I agree with Pat that the per- descriptions are often more prolix
> than the en- descriptions for many common ways of talking about
> the world. That is partly because people have developed a large
> vocabulary of common expressions that would require very extended
> translations if mapped into the primitives of the per-
> perspective.
>
> 6. Whitehead is often criticized for inventing his own
> highly technical
> vocabulary, but to a large extent, that criticism misses
> the point.
> W. was trying to develop a way of talking that could express
> the concepts that were significant for both the en- and per-
> vocabularies. Therefore, he had to coin many new words to express
> the highly general concepts that were missing from both.
>
> For an introduction to Whitehead's point of view, you can start with
> his 1920 book, _Concept of Nature_, which is available on the web:
>
> http://paradigm.soci.brocku.ca/~lward/Whitehead/White1_pref.html
>
> His vocabulary in this book isn't as technical, since he was just
> getting started in developing this point of view. (He had already
> written his book _Principles of Relativity_, in which he presented
> his own formalism as an alternative to Einstein's. For all results
> that were observable at that time, Whitehead's method and Einstein's
> gave equivalent results, but there are some newer results which
> agree better with Einstein's version than with Whitehead's. But
> in any case, W. certainly knew how to represent 4D.)
>
> For his later work, see the following books (all available in
> paperback):
>
> Science and the modern world (1925)
>
> Process and Reality (1929)
>
> Adventueres of Ideas (1933)
>
> Modes of Thought (1938)
>
> Process and Reality is the most technical, but also the most complete
> presentation of his ontology. If you find it hard to read,
> don't worry.
> That is what happens to most people, including professional
> philosophers.
> But the other books are more accessible, and after going back to PR
> from time to time, you begin to see the need for a level of concepts
> that can relate the en- and per- terminologies to each other.
>
> Whitehead is noted for making statements that are widely quoted out
> of context -- not incorrectly, but without the underlying
> philosophical
> issues he was trying to make. But they still make good
> reading anyway.
> Following is a list of some of them:
>
>
> http://www-groups.dcs.st-andrews.ac.uk/~history/Quotations/Whi
tehead.html
John Sowa