Doug, Sean,
I agree generally with your comments, but would add some additional
comments: (01)
[Sean Barker] > > A chair is not defined by its form, but by the processes
that humans apply to create and use it (02)
> > ...
>> However there is nothing of itself that makes a chair a chair (think
> of
> > the 60's fad for pod shaped chairs hanging from the ceiling).
> > Basically, chair is an artefact designed and manufactured for a
> single
> > adult
> > human to sit on, clear of the floor. This definition eliminates
> > - Rocks
> > - Logs
> > - Tables (which can be sat on)
> > - Benches, pews, settles, sofas
> > - Cushions
> > However, the definition would need a few tweeks to eliminate car
> seats,
> > ejector seats, lavatories, bosun's chairs, etc.
>
[DF] > ... stools, bicycle seats, ... Wouldn't a back support be needed.
> Why exclude children's chairs? My main point here is that various
> existing ontologies would have different descriptive derinitions.
> Most would have defined their local terms on the basis of object
> features.
> This definition is based on the intent of the designer & manufacturer
> instead of intrinsic properties of the object.
>
The only stable way I have found thus far to logically specify the
meanings of artifacts is to describe them as objects that are *designed* to
serve a particular purpose. In some cases there may also be necessary
properties. Thus we might have an ontology type labeled "Chair" which is a
physical object that can be moved by one person, and is designed to support
one person in a sitting position on a horizontal surface, and has a back
support. That would include folding chairs, even when they are folded. It
would exclude seats designed to be bolted to a floor, or seats too heavy to
move (perhaps a stone throne?), and stools, and some modernistic objects
that the manufacturers call "chair". No problem. What people **call** a
chair is a different problem from what ontology entity might be labeled a
"Chair". If someone needs a chair-like object that has different
properties, it can also be included in the ontology, under a different
ontology label, and if the user desires, can be mapped to the term "chair"
in any terminology used in any context. (03)
The point here is that what a user chooses to map to a linguistic label is
completely arbitrary to that user. The entities in the ontologies should be
as well defined as practical, and whether one or another or several of those
ontology entities happens to suit the terminological needs of some user, is
a matter for the linguistic mapping. So Wittgenstein may not be able to
find any common property in things called "game" but we can still have an
ontology entity called "game" that has as its subtypes all of the much
better defined "games" (sports, board games, hide-and-seek) that people
play; in effect it would be the union of all those other types, even though
they have no common properties. Though I would not include psychological
"mind games" within such an umbrella category, I would put them elsewhere.
It isn't necessary - if any user wants to refer to *any* ontology entity as
a "game" they only need to create the proper mapping to the ontology
entity(s) that they want to refer to. (04)
I do think it is a good idea to try to make ontology labels as informative
as possible, to avoid misleading the ontologist or database developer. But
we know that no label will be fully satisfactory to everyone, and that is
not important in ontology development. It is only important to include
clear logical specifications of the ontology entities people want to use in
their applications, so that they will have entities to map to the labels
used in their applications. I try to use ontology labels in their most
common sense, and if that is still potentially misleading, leave out the
common words and use more specific terms. So in the COSMO there is no
entity called simply "Process" because that is used by different communities
in different senses. But any user can map the word "process" to any
entity(s) whose logical specification(s) seems to fit their intended
meaning, or if none exists, they can create a new one with the desired
logical structure. (05)
Pat (06)
Patrick Cassidy
MICRA, Inc.
908-561-3416
cell: 908-565-4053
cassidy@xxxxxxxxx (07)
> -----Original Message-----
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-
> bounces@xxxxxxxxxxxxxxxx] On Behalf Of doug foxvog
> Sent: Tuesday, February 16, 2010 6:32 PM
> To: [ontolog-forum]
> Cc: ontolog-forum@xxxxxxxxxxxxxxxx
> Subject: Re: [ontolog-forum] Chairs - was Foundation ontology, CYC, and
> Mapping
>
> Sean Barker wrote:
> > Doug Foxvog wrote
>
> >> "The referent for many terms would have to be defined in NL. And
> the
> >> boundary conditions for terms would be hard to define. Can we
> formally
> >> define a cat? Certainly one could formally define a chair, but each
> >> ontology to be mapped to would probably have a different definition.
> It
> >> might be nice to describe fuzzy boundaries for concepts."
>
> > One might define cat through what cats are in themselves, a
> particular
> > group
> > of living creatures, defined by the genetic boundaries that prevent
> > breeding
> > with other creatures.
>
> This definition relies on an unknown genetic defintiion. It can not be
> used to determine whether a given living being is a cat.
>
> > However there is nothing of itself that makes a chair a chair (think
> of
> > the 60's fad for pod shaped chairs hanging from the ceiling).
> > Basically, chair is an artefact designed and manufactured for a
> single
> > adult
> > human to sit on, clear of the floor. This definition eliminates
> > - Rocks
> > - Logs
> > - Tables (which can be sat on)
> > - Benches, pews, settles, sofas
> > - Cushions
> > However, the definition would need a few tweeks to eliminate car
> seats,
> > ejector seats, lavatories, bosun's chairs, etc.
>
> ... stools, bicycle seats, ... Wouldn't a back support be needed.
> Why exclude children's chairs? My main point here is that various
> existing ontologies would have different descriptive derinitions.
> Most would have defined their local terms on the basis of object
> features.
> This definition is based on the intent of the designer & manufacturer
> instead of intrinsic properties of the object.
>
> > A chair is not defined by its form, but by the processes that humans
> apply
> > to create and use it (the forms of life in which it is involved).
> This is
> > what I mean when I say that "semantics is defined by behaviour". We
> have
> > common semantics (we have made the same differentations) up to the
> point
> > where we have common forms of life.
>
>
> > If when I ask the robot grocer to give
> > me "three red apples" it gives me three red apples, then I can be
> happy it
> > understood the request - in saying that the semantics of "three red
> > apples"
> > is understood, all I am saying is that it behaved in the correct way.
> I
> > have
> > no need of the concept of "concept" to describe its understanding.
> >
> > That is, semantics is about the behaviour of systems - if you want to
> talk
> > about the meaning of signs, I think (please correct me) we are
> talking
> > semiotics. I suspect that much of the argumentation of recent threads
> > (primitives, Longman's dictionary, etc) is a proxy arguement between
> > semantics and semiotics.
> >
> > Sean Barker, Bristol
>
>
> =============================================================
> doug foxvog doug@xxxxxxxxxx http://ProgressiveAustin.org
>
> "I speak as an American to the leaders of my own nation. The great
> initiative in this war is ours. The initiative to stop it must be
> ours."
> - Dr. Martin Luther King Jr.
> =============================================================
>
>
> _________________________________________________________________
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
> Shared Files: http://ontolog.cim3.net/file/
> Community Wiki: http://ontolog.cim3.net/wiki/
> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
> (08)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (09)
|