RE: SUO: Monolithic ontologies (was ontology as science)
Chris,
At 11:04 PM 7/2/2003 +0100, Chris Partridge wrote:
>Adam,
>
>I can appreciate your frustration - but I think it is missing the issue.
>
>The first stage is to get an overview, a context. When this has been
>attempted for SUMO - it has not gone smoothly. E.g. is it 3D or 4D? We are
>told 3D. We start looking at the detail of the axioms and they seem to
>suggest otherwise. A number of us ask questions - and we still have not got
>to the bottom of this. We are stuck at the first stage.
That's because you're not looking at the axioms. It's unsurprising that
the general-purpose and imprecise labels don't match the axioms. You've
gotten an overview label - it's 3D - now look at the details to get a real
understanding of the formal model. That you find some things that support
a 4d view should also not be surprising.
>So it is not a simple matter of division of time - but one of a natural
>progression. If one gets stuck at the first stage, then one will spend most
>of one's time at that stage. One should not expect anything different.
Then don't get stuck at the 50,000 foot level. Do some work on the ground.
>Also, there is another important point. We want to understand the impact of
>different choices, given that we have a potential range of ontologies to
>'merge'. So, it becomes useful to see how the ontologies have implemented
>the choices.
Indeed, but the only way to do that is look at the axioms, preferably with
a theorem prover on hand. Try some proofs and see how they come
out. Change an axiom and see what that does to your
conclusions. Otherwise you'll never understand the impact of your choices
any more than you would writing software and never compiling or executing it.
Adam
>Regards,
>Chris
>
>-----Original Message-----
>From: Adam Pease [mailto:adampease@earthlink.net]
>Sent: 02 July 2003 22:44
>To: mail@chrispartridge.net
>Cc: standard-upper-ontology@ieee.org
>Subject: RE: SUO: Monolithic ontologies (was ontology as science)
>
>
>Chris,
> Yes, labels and summaries can help, and they can be useful context for
>proceeding towards deeper understanding. It's just that the balance is off
>here. If we spent 10% of the time on the list discussing general
>principles, and 90% of the time working on embodying those principles in
>axioms, then the balance would probably be right. Unfortunately, right now
>the ratio is something like 99.999% and 0.001%. After several years, it's
>time that most comments reflect concrete additions or changes to a
>particular document, and not high-level discussion.
>
>Adam
>
>At 10:33 PM 7/2/2003 +0100, Chris Partridge wrote:
> >Adam,
> >
> >I beleive we have covered this ground before.
> >
> >It is much easier to review something as large as SUMO if there is a
>general
> >principle to guide you through it. So, for example, Nicola made the similar
> >point about understanding the choices that motivate an ontology in his last
> >email.
> >
> >And, for a SUMO example, Ian has in a recent email effectively said that
> >SUMO is eternalist rather than presentist - making no distinction between
> >the tenses. This is a useful and helpful architectural principle to know -
> >and helps us to understand SUMO better. It illuminates a raft of axioms.
> >
> >If there is a similar principle relating to 3D and 4D - this would also be
> >helpful. It is also not an uncommon principle to consider in ontologies -
> >e.g. Nicola's DOLCE is explicitly 3D and Matthew's EPSITLE is explicitly
>4D.
> >
> >The problem, for me at least, is that trying to get to these principles in
> >SUMO is like pulling teeth. And without the context they provide it is very
> >difficult to "get a mutual understanding, much less agreement". Getting
> >"down to details and comment[ing] on particular terms and axioms" is not an
> >easy route to understanding if the underlying principles are not clear.
> >
> >This is not to decry the value of the axioms etc., looking at them is a
> >natural next step once one has some grasp of the context.
> >
> >In this regard, ontologies are no different from any other large system of
> >knowledge.
> >
> >Regards,
> >Chris
> >
> >
> >-----Original Message-----
> >From: adampease@earthlink.net [mailto:adampease@earthlink.net]
> >Sent: 02 July 2003 20:19
> >To: Chris Partridge
> >Cc: standard-upper-ontology@ieee.org; Patrick Cassidy
> >Subject: RE: SUO: Monolithic ontologies (was ontology as science)
> >
> >
> >
> >Chris,
> >
> >At 06:40 PM 7/2/2003 +0100, Chris Partridge wrote:
> >
> > >Adam, (Pat,)
> > >
> > >In response to Ian's claim that you refer to ("I think Ian commented
> >earlier
> > >that he considers SUMO basically 3D.") we entered into a discussion in
> >which
> > >it seemed that he was actually commiting to a 4D position for SUMO (a
>point
> > >Matthew West has also made) - though whether Ian accepts this is not
> > >entirely clear as he has not answered the last message.
> > >
> > >My personal suspicion is that concern about these architectural issues -
> > >e.g. making a clean choice on 3D and 4D - was not considered in the
>design
> > >of SUMO.
> > >
> > >There seem to me two possible positions here.
> > >1) An architecural approach that makes choices about 3D and 4D (and
> > >eternalism and presentism, etc.) is not required - and was not taken.
> > >2) It is required and was taken.
> > >
> > >If SUMO adopted 1), then it would be interesting to known what braod
> > >principles architected it - if any (apart form Ian's naturalism :-) ).
> >
> >I'd say #1. But really, I think this whole conversation is off track. If
> >we are to focus on a document, let's get down to details and comment on
> >particular terms and axioms. We could talk forever at an informal level
> >and never get mutual understanding, much less agreement. "Let us
> >calculate!"
> >
> >Adam
> >
> >
> >
> > >Regards,
> > >Chris
> > >
> > >
> > >-----Original Message-----
> > >From: owner-standard-upper-ontology@majordomo.ieee.org
> > >[mailto:owner-standard-upper-ontology@majordomo.ieee.org]On Behalf Of
> > >Adam Pease
> > >Sent: 02 July 2003 02:22
> > >To: Patrick Cassidy
> > >Cc: standard-upper-ontology@ieee.org
> > >Subject: Re: SUO: Monolithic ontologies (was ontology as science)
> > >
> > >
> > >
> > >
> > >
> > >Pat,
> > >
> > >At 05:27 PM 7/1/2003 -0400, Patrick Cassidy wrote:
> > >
> > > >Adam,
> > > > Concerning your statement:
> > > >
> > > > > That entities do not change their class membership over time,
> > > > > and that classes themselves are reserved for those things which do
> > > > > not change, is an important guideline.
> > > > >
> > > >
> > > > First, a clarification: Do you consider SUMO 3D or 4D or both?
> > >
> > >I think Ian commented earlier that he considers SUMO basically 3D. That
> > >sounds right to me, although you're welcome to ask him for elaboration.
> > >
> > > > The specific question related to your statement above is
> > > >whether roles (student, dentist) must therefore be treated
> > > >only as relations, or whether they can also be classes.
> > > >The latter would seem to require that membership change
> > > >over time.
> > >
> > >Those sorts of roles have been treated as properties. Take a look at
> > >SocialRole
> >
> ><http://ontology.teknowledge.com:8080/rsigma/SKB.jsp?req=SC&name=SocialRo
> le>&>skb=SUMO>
> > >
> > > > If roles cannot be classes, then do you consider SUMO
> > > >classes like "Food" and "BiologicallyActiveSubstance"
> > > >as *not* being roles which might change for a particular object?
> > >
> > >I'm guessing that you're concerned about cases like the same substance
> > >being used as a floor wax or a dessert topping, but as I see it, it would
> > >still be a food, just as a table could be used as a chair, or a TV as a
> > >hammer, but they would still be tables and TVs, respectively.
> > >
> > >Adam
> > >
> > >
> > >
> > > > Pat
> > > >
> > > >=============================================
> > > >Patrick Cassidy
> > > >
> > > >MICRA, Inc. || (908) 561-3416
> > > >735 Belvidere Ave. || (908) 668-5252 (if no answer)
> > > >Plainfield, NJ 07062-2054 || (908) 668-5904 (fax)
> > > >
> > > >internet: cassidy@micra.com
> > > >=============================================