RE: PDKB States of Matter
Dear Colleagues,
There seems to be an assumption being made here I would like to challenge.
This is that the objectives of this group are met if we can agree on an
ontology that covers some of what we call common sense.
This is not true for me at least. I am working in a business, engineering,
and scientific environment in which there are a number of different views of
the world which are used for different purposes. To bring things together in
this environment I need an ontology that underlies and can explain the
relationship between the different views of the world that legitimately
exist.
Regards
Matthew
============================================
Matthew West
Operations & Asset Management
Shell Services International
H3229, Shell Centre, London, SE1 7NA, UK.
Tel: +44 207 934 4490 Fax: 7929
Exchange e-mail: Matthew.R.West@is.shell.com
Compuserve e-mail: Matthew_West@compuserve.com
http://www.shellservices.com/ <http://www.shellservices.com/>
============================================
-----Original Message-----
From: apease [mailto:apease@teknowledge.com]
Sent: 30 June 2000 22:07
To: Peter Szolovits; pdkb@ldl.HealthPartners.com
Cc: standard-upper-ontology@ieee.org
Subject: Re: PDKB States of Matter
Pete,
Thanks for your points on my informal axioms, especially the reversal from
what was intended on (b). While I don't speak for the whole Standard Upper
Ontology effort, I would suggest that it have the aim of codifying possibly
naive default rules which could be made more correct or specific in theories
which build on the SUO. In order to keep the SUO from becoming an even
bigger undertaking than it already is, I think we have to resist a
temptation to make every area correct and complete. That does beg the
question though of how correct or how complete must it be to be a viable
standard.
Ideally, I think we would codify theories which, much as Newtonian
mechanics is a special limiting case of Einsteinian mechanics, are useful
and good enough for the majority of situations.
You point out another issue in that the axioms below assume other factors
such as acceleration are 0. I think that's actually a reasonable
assumption. To use one of Steven Pinker's examples (although he was making
a different point) we might have an axiom that says if a person goes to the
store, then his head goes with him. That's a reasonable assumption even if
he could be decapitated while riding the bus and arrive at the store without
his head.
Adam
At 04:49 PM 6/30/2000 -0400, Peter Szolovits wrote:
Just a couple of comments on Adam's informal axioms.
b is backwards
d is true only if we control for pressure. Generally, it's difficult to
axiomatize "common sense" binary relationships when the underlying reality
involves more terms. E.g., PV=nrT means that any pairwise relationship
between variables holds only if you hold the others constant.
The notion of contained liquids and gases "conforming" to their containers
or to gravity-selected parts of their containers is tricky to state
correctly. First, in gravity or under acceleration, gas in a container may
not have a uniform pressure. This has implications for b and g. Second, by
Einstein's arguments, you can't treat gravity and acceleration separately.
Thus, a circular tank spinning with its axis perpendicular to gravity with
liquid in it will produce a parabolic surface as the top of what the liquid
will conform to. (This observation led to a very cheap way to make
excellent infrared telescopes by letting resin cure in a tub spinning at
constant velocity!)
Sacks and Doyle [1], for example, have made a distinction between AI
theories that try to capture "naive physics" intuitions of the man on the
street, which may often miss lots of subtleties and therefore be wrong, vs.
"qualitative physics", which is in line with accepted theories but abstracts
away from particular magnitudes. To what purpose is the PDKB representation
constructed?
--Pete Sz.
[1] E. P. Sacks and J. Doyle, Prolegomena to any future qualitative
physics, Computational Intelligence, vol. 8, 1992.
At 09:06 PM 6/29/00 -0700, apease wrote:
a. Gases conform to the shapes of their containers
b. A gas escapes from a container if it is not closed and the pressure
outside the container is larger than the pressure inside
c. If the volume of a container is greater than the volume of a liquid in
the container then the liquid conforms to the shape of the portion of the
container that has a volume equal to the volume of the liquid and is in the
direction of the vector of gravity.
d. For a particular substance, the temperature at which it is a gas is
higher than the temperature at which it is a liquid and in turn than which
it is a solid.
e. Default statement: substances have non-zero temperature ranges in
which they exist in each of the three states
f. Gas volume is proportional to temperature and inversely so to pressure
g. A liquid escapes from a container if it is not closed and the opening
in the container is between the surface of the liquid and the terminating
point of the vector of gravity, and the pressure of any gas in the container
is greater than the pressure of any gas outside the container minus the
partial pressure of the liquid on the area of the opening in the container
(a bit too complicated, I hope this is right)
-----------------
Adam Pease
Teknowledge
(650) 424-0500 x571