|
At 3:14 PM -0400 3/11/08, Patrick Cassidy wrote:
And yet, John, I believe that I
understood every word you said (955 of them
in the note, including the previous quotes) in exactly the sense
you
intended it.
How could you possibly know this?
And some of it was not just the
basic vocabulary. Is that hard
to believe?
Lets say it isn't, notwithstanding the Quinean objection.
How could I do it without a mental
model very similar to yours?
By having a mental model which supports enough entailments to
agree with yours on those entailments. This is a MUCH weaker condition
than being identical.
Pat H
Pat
Patrick Cassidy
MICRA, Inc.
908-561-3416
cell: 908-565-4053
cassidy@xxxxxxxxx
> -----Original Message-----
> From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-
> bounces@xxxxxxxxxxxxxxxx] On Behalf Of John F. Sowa
> Sent: Tuesday, March 11, 2008 2:57 PM
> To: [ontolog-forum]
> Subject: Re: [ontolog-forum] Ontology similarity and accurate
> communication
>
> Dear Matthew, Pat, Pat, and Sean,
>
> The following point raises an important question about degrees
of
> closeness:
>
> PC> My point was that the language learning process is
sufficiently
> > *similar* in different people learning the same native
language,
> > that the process supports the ability of learners
to develop a
> > common internal ontology (of unknown structure) that
is very close.
>
> Children (and adults) learn a large vocabulary, but the words
have
> a very flexible meaning, which results in an open-ended number
of
> word senses, each of which has a different set of axioms.
>
> SB> The claim that we simply learn an ontology of fundamental
concepts
> > is not one I would be comfortable with. We learn that
the elements
> > of earth and water fall, while fire rises - concepts
like weight,
> > and near-the-Earth's surface are secondary, and the
latter quite
> > outside most people's experience.
>
> I agree. Civilization had developed to a very high degree
before
> Aristotle introduced his ontology. That ontology and most
of the
> others that anyone has proposed are *abstractions* from the
way
> language is actually used. Those ontologies might be useful
for
> many computer applications, but there is no evidence that
anything
> like Cyc or other computer systems is going on inside the
minds
> (or brains) of children and adults who use language.
>
> MW> I suggest that the 3D/4D interpretation of continuant and
occurrent
> > mentioned below is a well known example of
contradiction between
> upper
> > ontologies.
>
> Yes. That illustrates the point that just agreeing on the
choice of
> words in a natural language does not force agreement on any set
of
> detailed axioms. As we can see from any large dictionary,
there are
> many different word senses for each word, which require very
different,
> and as Matthew notes, usually incompatible axioms.
>
> PC> My opinion that our mental models for the basic terms are
over
> > 99.9% in agreement is based on personal observation of
the high
> > accuracy of communication, when using the basic
words.
>
> I have no idea where you got that percentage or what it
means.
> Furthermore, communication in any NL is definitely *not*
accurate
> without a great deal of explanation, clarification, and
negotiation.
>
> Anyone who has ever been a teacher (at any level from
Kindergarten
> to graduate school or lectures on specific topics) knows that
very
> little of what the teacher says gets through to the students on
the
> first try. Discussion, questions, repetition, exercises,
exams,
> tutorials, and a wide variety of readings are essential.
Even then,
> only a few students really master the material -- and the best
ones
> are *not* the ones who memorize the teacher's words.
>
> PH> DOLCE and BOF both require the categories of continuant
and
> > occurrent to be disjoint: nothing can possibly be both
a continuant
> > and an occurrent in these
ontologies. Other ontologies (I have one,
> > and I think the same is true of Cyc) allow both
categories wit
> > pretty much the same properties of the respective
types, but allow
> > them to overlap.
>
> I agree. In my KR ontology, I made the point that the
dividing line
> between the two is task dependent. If you're skiing on it,
a glacier,
> for example, is an object (continuant). But a geologist
would view
> it as a process (occurrent) that melts, flows, breaks apart,
and
> acquires new layers at the top.
>
> PC> Clearly, if some entity is an instance of one
"occurrent" but not
> > an instance of another "occurrent", the
meanings differ -- they are
> > using the terms in different senses.
>
> That's fine. But similar issues arise with nearly every
word in the
> language. The main thing that native speakers of the same
language
> agree on is the basic vocabulary. The senses of those words
change
> with every application. Just look at any dialog by
Plato.
>
> The main point is that nearly every word (especially all the
common
> ones) has an open-ended variety of senses -- which the
linguist
> Alan Cruse aptly called 'microsenses'. Each of those
microsenses
> can be axiomatized for a particular task, but each task
requires
> a change of word senses that may involve a total
restructuring
> of the axioms for all the task-related terms.
>
> Even in the same so-called field, such as medicine, basic
words,
> such as heart, kidney, blood, skin, or bone, are used in very
> different ways with different axiomatizations by a patient,
> a nurse, a general practitioner, a specialist, a pharmacist,
> a microbiologist, etc. They may agree in a vague sense
about
> that thumping thing in the chest, but their axioms are very
> different.
>
> Fundamental principles:
>
> 1. Communication between two agents (human, computer,
or whatever)
> requires agreement at the *task*
level or even the level of
> individual *messages* -- but *not*
at any kind of global level.
>
> 2. Different agents talking about different tasks
with other agents
> may use very different
axiomatizations for the same terms.
>
> 3. No two agents *ever* require a global alignment of
their
> ontologies in order to communicate
effectively. The only
> agreement necessary (or even
possible) is at the level of
> the task they are doing.
>
> 4. When multiple agents are cooperating on the same
task,
> e.g., surgeons, nurses,
anesthetists, patient, etc.,
> any given agent may use very
different ontologies in
> communicating with each of the
other agents.
>
> 5. Even for the same two agents, their choices of
ontologies
> may differ widely when they are
cooperating on different
> tasks.
>
> 6. When misunderstandings arise (as they inevitably
do), the
> agents switch to a metalevel of
questions, explanations,
> clarifications, and negotiations in
order to align subsets
> of their ontologies for the
specific task they are doing.
>
> John
>
>
>
>
_________________________________________________________________
> Message Archives:
http://ontolog.cim3.net/forum/ontolog-forum/
> Subscribe/Config:
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 Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>
_________________________________________________________________
Message Archives:
http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config:
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 Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
--
---------------------------------------------------------------------
IHMC
(850)434 8903 or (650)494 3973 home
40 South Alcaniz St.
(850)202 4416 office
Pensacola
(850)202 4440 fax
FL 32502
(850)291 0667 cell
http://www.ihmc.us/users/phayes
phayesAT-SIGNihmc.us
http://www.flickr.com/pathayes/collections
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: 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 Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (01)
|