Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: lattice of ontology



John,
All the indicated good points, especially this one: 'Everybody who develops 
an upper ontology has very different and inconsistent axioms at the topmost 
levels.''
 But the real reason of the SUO nonachievement is that the extant 
computational entity typologies are implicitly shaped by the old-fashioned 
philosophical traditions, where reality is separated into two or three 
disjoint divisions of the same rank and status. These broad categories are 
either concrete individuals (contingent things as physical objects and 
events) and universals (necessary things, or abstractions as classes, 
states, qualities, and relations). Or thay may be the concrete and the 
abstract (conceptual, conceptional, ideal, theoretical entities, or 
universals) intermediated with collections (or classes or sets) of concrete 
things (concrete universals). Written in abbreviated forms, the above may be 
presented as the CC (concrete and conceivable) schema or the CCC (concrete, 
collective, and conceivable) perspective of reality; what is commonly 
expressed in the natural languages as concrete names (individuals), 
collective names (collections), and abstract names (classes or masses). Then 
most existent general and upper ontologies, like the Semantic Web and SUO 
candidates, can be assigned to the CC or CCC taxonomies, limited by Nothing 
or Nonentity at the bottom and the concept of (individual) Thing or Entity 
at the top.

It appears that the main theoretical obstacle to the standard reference 
framework lies in the uncritical acceptance of the dichotomous 
classification of entities into realities and abstractions, like as tangible 
(objects), intangible (processes), or theoretical entities 
(representations). When individuals (the concrete) together with sets 
(abstract representations or collections of individuals) are viewed 
equipotent with ontological structures (universals, kinds, generic things, 
or classes of entities), the researchers are doomed to create false 
assumptions while trying to construct a formal general ontology language.

In effect, we need to consider the things in the right order of their 
existence and presentation, neither as an equal-order absolute separation 
into abstract and concrete nor in the inverse order, from the concrete 
objects, properties, events, and specific relations to the abstract 
ontological classes of substance, state, change, and relation. The order of 
things here makes all the difference. As the sociologists observed, a young 
woman personality may be quite different depending on the order of 
occurrences of her life experiences: becoming a mother, a college 
graduation, and becoming a wife.

However  complicated, in the bottom the matter looks quite simple: to have 
SUO we need an adequate representation of reality done not from the specific 
perspectives of formal logic, cognitive sciences, linguistics, natural 
sciences, or information sciences, but from one unifying ontological (world) 
position.

Just don't neglect the world as the only true reference frame, otherwise all 
your efforts lead to naught, constantly repeating the same error like the 
ONTAC Forum concocting from SUMO, DOLCE, and CYC ingredients an indigestible 
mixture.


Azamat

>
> ----- Original Message ----- 
> From: "John F. Sowa" <sowa@BESTWEB.NET>
> To: "Azamat Abdoullaev" <abdoul@CYTANET.COM.CY>; <velman@COX.NET>; 
> <standard-upper-ontology@listserv.ieee.org>
> Sent: Tuesday, December 27, 2005 6:05 PM
> Subject: Re: lattice of ontology
>
>
>> Azamat,
>>
>> I think most people have come to some conclusion like that:
>>
>>> Maybe because we all are becoming accustomed to the thinking
>>> that the SUO is a pipe dream and better to leave behind the
>>> listing and all associated with it.
>>
>> When this group was founded in 2000, most of us had some hope
>> that something useful could be accomplished, and I think we
>> have learned a lot.  Following are my conclusions:
>>
>> 1. Everybody who develops an upper ontology has very different
>>    and inconsistent axioms at the topmost levels.
>>
>> 2. Those inconsistences at the top make it impossible to share
>>    anything at the lower levels with any other ontology whose
>>    lower levels depend on assumptions made at the top.
>>
>> 3. Yet people have been communicating successfully for
>>    thousands of years with very few abstract assumptions
>>    about top-level entities, such as time, place, object,
>>    process, etc.
>>
>> 4. Database systems have been interoperating successfully for
>>    about 40 years with very few axioms or assumptions about
>>    the top levels.
>>
>> 5. The most successful sharing in *all* fields -- science,
>>    engineering, medicine, business, etc. -- has been based on
>>    *terminology* at lower levels with very few, if any axioms
>>    about the upper levels.
>>
>> 6. On the other hand, we do need axioms (and programs, which
>>    are essentially compiled axioms) in order to do any kind of
>>    detailed reasoning, computation, and problem solving.
>>
>> 7. Therefore, we should make a clear distinction between the
>>    vocabularies or terminologies, which have very few axioms,
>>    and the problem-oriented reasoning and computational
>>    systems.  For general purposes, sharing should be based on
>>    the terminology.  For reasoning and computation, the axioms
>>    should be introduced at the lower problem-oriented levels.
>>
>> In short, the hope of finding a commpon set of axioms at the
>> upper levels is *DOOMED*.  The SUO work over the past five years
>> has been interesting, and we all learned a lot.  But the most
>> important thing we leared is that assuming a fixed and frozen
>> set of upper level axioms does not promote interoperability.
>> Instead, the axioms introduce irrelevant contradictions that
>> are a major barrier to communication and sharing.
>>
>> John Sowa
>>
>
>