W3C home > Mailing lists > Public > public-xg-eiif@w3.org > April 2009

Re: Standard ontological relations for our voc -leverating Obo relations

From: Gary Berg-Cross <gbergcross@gmail.com>
Date: Mon, 27 Apr 2009 11:05:55 -0400
Message-ID: <f867f9b20904270805g4ded1540q2cee1771f422849b@mail.gmail.com>
To: Guido Vetere <gvetere@it.ibm.com>
Cc: paola.dimaio@gmail.com, public-xg-eiif <public-xg-eiif@w3.org>, public-xg-eiif-request@w3.org
Guido,

Here is a quick response to your remarks  - we may want to dialog on this to
reach an understanding and get back to the group.

>It seems that OBO relations are defined over 'continuants' (i.e., roughly,
objects) and 'occurrents' (processes) but I don't find a 'foundational
ontology' where these two classes are defined (maybe I miss something?)


In the OCO Foundry ontologies there is a the top level division
corresponding to the ways the entities exist in time. As I believe you know
this follows Barry Smith's Snap vs. Span perspctives, but seems compatible
with DOLCE.



So they have ‘Continuants’ endure through time. ‘Occurrents*’ *(processes)
unfold through time in successive stages.



Continuants are divided into physical things, on the one hand, and qualities
and functions, on the other. The latter are dependent continuants: a quality
such as the shape of a fly’s wing depends for its existence on, and endures
through time in tandem with, the wing that is its bearer; a function, such
as the function of an enzyme to catalyze reactions of a certain type,
similarly endures through time in tandem with the enzyme itself and exists
even when it is not being exercised in any instance of that reaction.

You can see a basic discussion of this at:

www.openehr.org/wiki/download/attachments/196630/NBT_OBO.pdf?version=1

On your 2nd remark
>I would suggest avoiding the mix of is_a with first-order relationships
like 'part', since the former has a specific logical import (subclass of)
which is natively axiomatized in any description logic like owl.

My take on this is that while we start with a representational languages
given relations we develope the ones we need using them and just be clear
what the semantics of the various part relations are (proper part vs
contained in).  That is, we aren't limited to the "natively axiomatized "
ones as long as we provide sutible axioms.

Gary Berg-Cross,Ph.D.
gbergcross@gmail.com
http://ontolog.cim3.net/cgi-bin/wiki.pl?GaryBergCross
SOCoP Executive Secretary
Principal, EM&I Semantic Technology
Potomac, MD
301-762-5441
On Mon, Apr 27, 2009 at 6:45 AM, Guido Vetere <gvetere@it.ibm.com> wrote:

>
> Gary,
> two quick remarks:
>
> It seems that OBO relations are defined over 'continuants' (i.e., roughly,
> objects) and 'occurrents' (processes) but I don't find a 'foundational
> ontology' where these two classes are defined (maybe I miss something?)
>
> I would suggest avoiding the mix of is_a with first-order relationships
> like 'part', since the former has a specific logical import (subclass of)
> which is natively axiomatized in any description logic like owl.
>
> Cordiali Saluti, Best Regards,
>
> Guido Vetere
> Manager & Research Coordinator, IBM Center for Advanced Studies Rome
> -----------------------
> IBM Italia S.p.A.
> via Sciangai 53, 00144 Rome,
> Italy
> -----------------------
> mail:     gvetere@it.ibm.com
> phone: +39 06 59662137
> mobile: +39 335 7454658
>
>
Received on Monday, 27 April 2009 15:06:34 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 15:06:35 GMT