Re: [ontolog-forum] Universal Basic Semantic Structures

[William Frank <williamf.frank@xxxxxxxxx>]

I agree with all your comments about this Ed, except that 

1. you seem to be saying that descriptions may be vague, that is, leave it unclear whether something fits the description or no, while classifers aren't.    It seems to me that we can take any description, such as "its raining", and lave that as vague or make it as razor sharp as we like. 

But then, I am delighted that: 

2. you seem to be talking about what I would call natural types, that describe some essense of the concept from the point of view of the given domain.  (For example, in your example, hardness might be a property of a mineral, but the best typology comes from crystaline structure, which determines these other attributes.   And, as science progresses, the typology changes. 

Horray, I say!!!!    

This is a critical feature not only of science, but also of every systematic domain of human endeavor, from police work to filmaking.    

Only, when I said "classifier", I mearly meant "specification that allows me to determine whether an individual satisfies the specification or not".  I did not mean natural type, or what would be a sort in a many sorted logic with a sensible semantics.   So, I agree with everything you said when it comes to natural types, and think that they are the essence of an ontology.     


Now, IUML conflates a/ natural types as I have described them b/ classifiers, in my sense, c/ the peice of langauge in which the classifier is specified (confusing the specification with the equivelence class of specifications that must apply to the same things -- such as cartesian and polar coordinates for some set of points) and d - a template, like a driver's licence application form, that can be used to instantiate an instance of some artifact (confusing instantiation with 'instance of'" --  while only eggs can instantiate chikens, not chiken 'classes', a bilogist can use a chicken specification to determine if something IS a chicken).  UML calls ****all*** of these 4 things by one name: "class".  This is a historical reason for our confusions.  I think that without a common language that distinquishes between these things, we will continue to talk at cross purposes.

On Thu, Sep 27, 2012 at 1:55 PM, Ed Barkmeyer <edbark@xxxxxxxx> wrote:

William Frank wrote:

Pat, my point is that when you describe something, you are ipso facto classifying it. That is just what a description does. It specifies the set of things that fit the description, even if that is a singleton.  

I think what Pat is saying is that a  well-defined 'classifier' is a razor.  You can apply it to decide unequivocally whether a given thing satisfies it or fails to.  A 'description' is not a razor.  A 'description' is merely some set of properties and observations that are true of the object or category, and may or may not be sufficient to differentiate the thing from other things that are similar but not intended.

In particular, I think you HAVE classified the thing, as something locallizable in space and sherical in shape, and occupying a particular position at a particular time.   this distinquishes it from Pakistan, topology, hurricane Dora, my brother, and many other things that do not fit the classifier you created.   

Indeed.  But that is hardly the primary mechanism for distinguishing basketballs from Pakistan, and it may or may not distinguish basketballs from planets.  The point of classifiers is to create a set of well-defined categories for a certain domain of interest. 

Put another way, descriptions include properties the things are noticed to have.  Classifiers are necessary and sufficient properties for being such a thing, which often do not include properties commonly appearing in descriptions of such things. 

To use a particularly NIST example, many  properties of materials are a direct consequence of the crystal structure of the material.  Therefore, a classifier for a material may specify its composition and crystal structure, which is a hard razor, while a typical description of the material will report its commonly observable physical properties, or properties that have been empirically measured.  While the description is very useful for engineers contemplating the use of the material for some particular purposes, those properties are not necessarily distinctive of the material, especially in comparison to its close relatives.

But, on reflection, I think my point was beside you all's point.  Because you all are discussing rules for **constructing** an ontology, and one of the rules is, we have a universe of discourse of atomic zero order things, and each of those has to have a classifier that it fits. 

Yes, in that many ontologies involve the development of rigourous classification systems (where that is possible and appropriate to the intended use).

So, this is a good rule for the completeness of an ontology, only it is not a rule about the world, where everything DOES have a classifier. It is a rule about how to build a human aritifact, a formal ontology.

I don't follow.  It seems to me that "things of interest" to a given sentient group have "classifiers" imposed on them by that sentient group.  But there are many things that don't have classifiers other than "uninteresting thing" or "something I never heard of" except for vary narrow interest groups.  And in many cases, we tend to be utterly unaware of the existence of things outside our ken.

-Ed

"There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy."
  -- a different William (Shakespeare), "Hamlet, Prince of Denmark"

-- 
Edward J. Barkmeyer                        Email: edbark@xxxxxxxx
National Institute of Standards & Technology
Manufacturing Systems Integration Division
100 Bureau Drive, Stop 8263                Tel: +1 301-975-3528
Gaithersburg, MD 20899-8263                Cel: +1 240-672-5800

"The opinions expressed above do not reflect consensus of NIST, 
 and have not been reviewed by any Government authority."

Wm

>
> Pat Hayes
>
> ------------------------------

------------------------------
> IHMC                                     (850)434 8903 or (650)494 3973
> 40 South Alcaniz St.           (850)202 4416   office
> Pensacola                            (850)202 4440   fax
> FL 32502                              (850)291 0667   mobile
> phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
>
>

 

--
William Frank

413/376-8167

_________________________________________________________________
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
 

-- 
William Frank

413/376-8167


This email is confidential and proprietary, intended for its addressees only.
It may not be distributed to non-addressees, nor its contents divulged,
without the permission of the sender.
_________________________________________________________________
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    (01)