Thursday, February 28, 2008 5:19 AM, John wrote (01)
> Azamat,
>
> There are a great many people who have interesting ideas about
> ontology. But until any large ontology has been tested on a
> significant number of major applications, I have no faith in
> its consistency, adequacy, and extensibility. (02)
That's correct. An extensive experimental test in all field of knowledge is
a necessary condition for proving the validness of a single ontology.
>
>> Now look at the CYC upper level ontology. It allows some hybrid thing
>> named
>> 'a partially intangible' thing, a hybrid of abstract, notional,
>> intangible
>> and concrete, physical, tangible. As the instances of individuals, along
>> with the physical objects named spatial and temporal things, we meet here
>> the so-called intangible individuals that encompass time intervals,
>> attribute values, and mathematical objects as relations, as well as
>> situations and events. Only sets and collections are admitted as purely
>> mathematical things. And many other strange things.
>
> I have criticized those features of the Cyc upper ontology since I
> first began to study it about 18 years ago. I have discussed them
> with Doug Lenat and his colleagues many times over the years. During
> that time, they have made many changes, and they have made strong
> arguments for the features you criticize. I have not been fully
> convinced by their arguments, but I must admit that their choices
> cannot be so easily dismissed as you have claimed.
>
> Over the years, Lenat has hired many outstanding researchers in logic,
> linguistics, philosophy, artificial intelligence, and related areas.
> There are other good people working at top-rated institutions around
> the world, but on the whole, I would rate the Cyc group as one of the
> best, if not the best of its kind, in the world. (03)
I have no doubts about the high scientific quality of these people. They all
must be brilliant minds in their fields of expertise. The creators of the
project must be paid homage for their intellectual courage to start such an
enormous high-risk enterprise. (04)
What I doubt, the approach, the way they attack such unprecedented problems
as creating a common sense knowledge base. One can not build a good house
without a good design. Here you need a great conceptual design, uniform
ontological design, single conceptual framing, a consistent and
comprehensive top ontology. Lacking this attribute raises questions about
the validity of the whole enterprise regardless how brainy researchers and
engineers are hired and how extensive and expensive the implementation has
been performed. (05)
Take another good example you've mentioned recently, WordNet. I view the
project as a genuine linguistic deed made by great linguistic minds. But
look at their top tree-like scheme, entity (physical entity, abstract
entity). Now what they have as the abstract entities: otherworld, set,
quantity, communication, relation, group, attribute, psychological feature.
And what they have as the physical entities: thing, physical objects, cause,
substance, physical process.
It is a mix-up.
1. Thing is NOT a physical entity, but vice versa is the case.
2. Relation is NOT an abstract entity, like spatial and temporal relations,
like causal relations of forces.
3. Event is NOT an abstract entity or psychological feature, but a kind of
real change.
4. Communication is NOT an abstract entity, but a kind of relationship, etc. (06)
Having an inconsistent top ontology, however large-scale and costly, will
end up with inefficient knowledge applications.
The same analysis can be done for the CYC top ontology, with the like
results. (07)
> But as I have said, no ontology, not even Cyc, has demonstrated
> that it is adequate to support all the claims that have been made
> for an ontology. For example, see
>
> http://www.jfsowa.com/pubs/challenge.pdf
> The Challenge of Knowledge Soup
>
> > Real world knowledge applications ask for the unified representation
> > of the world, whatever you like to name it, global ontology, standard
> > ontology, etc.
>
> You have made that claim repeatedly, but there is *zero* evidence
> in its favor. If you look at the entire history of mathematics
> and science, there is no single foundation for any of the subjects.
> Whitehead & Russell made a good stab at it. The Bourbaki published
> many volumes of books. But most practicing mathematicians today
> consider foundational studies to be irrelevant to their daily work. (08)
The whole history of science consists in attempting to found more general
theoretical representations of objective laws and patterns, guiding the
world. The universality of mathematics had been accepted since Euclid and
Nicomachus, who put quantity with its key species, multitude and magnitude
as its subject matter. While Descartes, Whitehead , Russell extended the
mathematical universe by introducing order and relation. Its universality
implies a single axiomatic foundation regardless your practicing
mathematicians disregarding the mathematical foundation. (09)
> The state of physics is much worse. See _The Road to Reality_ by
> Roger Penrose, for a very good overview. (010)
His understanding of reality is very narrow and specific. imho, this book
hardly makes here a good argument. (011)
And every other field
> of science is far more chaotic than math & physics. When you
> go to the social sciences, economics, philosophy, business, etc.,
> there is no such thing as an extended chain of deductive reasoning
> that could stand up to any kind of empirical test. A unified
> foundation for those fields would be ludicrous, and if anybody
> tried to enforce any such foundation, it would be disastrous. (012)
Try and look at the situation from the other side. The natural and social
sciences are disorganized, they are uniformly unordered, lacking single
conceptual order and ontological and methodological arrangement. (013)
azamat abdoullaev
> _________________________________________________________________
> 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
> (014)
_________________________________________________________________
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 (015)
|