Re: [ontolog-forum] Ontologies and individuals

[Alex Shkotin <alex.shkotin@xxxxxxxxx>]

John,

Nicola reminds me 2 different definitions of ontology:

1. >.

2. >.

I prefer (2) as it keeps all knowledge together:-)

And it looks like Sandro keeps in mind (1).

Alex

2012/12/11 John F Sowa <sowa@xxxxxxxxxxx>

Alex, William, Nicola, and Hassan,

Scientific method:  Science begins by collecting data about individuals:

  1. By induction, scientists form generalizations from the data.

  2. By abduction, they form hypotheses about the data.

  3. By deduction, they make predictions from the combined data +
     generalizations + hypotheses.

  4. By observation, they test those predictions to generate more data.

  5. By further induction, abduction, testing, and belief revision,
     they keep refining their generalizations & hypotheses to form
     what they call theories.

  6. After a theory has been confirmed by systematic testing on a
     very large sample of combinations, they declare it to be a law.

  7. But any law of science might be falsified when it is tested on
     new kinds of data that hadn't previously been available.

Observation:  Any of the theories and laws of science could be used
as an ontology or part of an ontology.  But ontologies about artifacts
and human conventions can become true by fiat:  somebody declares some
particular theory to be a standard for that domain.

AS

> About individuals. We have some 100 000 rock samples around the world
> collected by field geologists. Some of these samples are from the Moon.

> All of them are very important individuals of petrography ontology...

>
> By the way if we look inside Big Data, it is mostly facts about huge

> numbers of individuals

Yes, but that kind of data is the starting point for science.

WF

>  Here you have supplied a great example of the absolute need for
> individual in a domain ontology in a big way.

Yes, but the induction phase might replace most or even all of the
data about individuals with general statements about types.

NG

> the point is not so much whether ontologies "contain" individuals,
> but whether they should include *facts* concerning specific individuals.

Yes, but the requirements can differ from one domain to another.

As Alex pointed out, Peano's theory about the integers requires some
"facts" about a certain privileged individual named '0'.  In general,
any theory that has recursive axioms requires a starting point, such
as 0, and some facts about that starting point.

An ontology for geological information requires a privileged individual
named 'earth'.  It also requires some facts about the earth in order
to define the geological features and coordinates.

HAK
> summing up the current W3C's (useless) standard epidemic: http://xkcd.com/927/

That cartoon about the proliferation of standards confirms Tim B-L's
requirements for diversity, heterogeneity, and interoperability.

Some conventions (AKA standards) are useful for coordinating diverse
human activities.  But more important than any specific convention is
a general framework that can accommodate any or all conventions and
facilitate interoperability among them.

There is an existence proof that such a framework is possible:
Every natural language enables people with different points
of view to collaborate on a common goal.  The challenge is to
develop a computational framework with that kind of flexibility.

John

_________________________________________________________________
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

_________________________________________________________________
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)