Re: SUO: a silly question about the new modular architecture
I agree with Adam, if i understand his diagram.
Does this mentograph accurately represent what you are talking about?
http://robustai.net/mentography/modulesConforming.gif
So if I change any axiom in a module, I would need to create a new module
that would be termed 'non conforming', then any other modules which were
dependant on my new module would also classed nonConforming.
Is that the way you all see it? If so, then might we have axioms in some
conforming module which would precisely define both of the two relations:
dependant and derivedFrom ?
Seth Russell
----- Original Message -----
From: "Adam Pease" <apease@ks.teknowledge.com>
To: "West, Matthew R SITI-GREA-UK" <Matthew.R.West@IS.shell.com>; "Seth
Russell" <seth@robustai.net>; "SUO" <standard-upper-ontology@ieee.org>
Sent: Thursday, September 06, 2001 7:22 AM
Subject: RE: SUO: a silly question about the new modular architecture
> Matthew,
> It still seems as though you and Seth are describing different cases.
I
> agree completely that many standards have modules and conformance
> classes. What I'm questioning is whether it makes any sense for someone
to
> develop their own module which is not in the standard and yet which
> addresses the same topic as an existing module in the standard. That
would
> seem to me not to conform to such a standard.
> As a simple example, say that we have the following modules and
dependencies
>
> structural
> / | \
> / | \
> / | \
> parts time-3d|4d units & measures
> \ /
> holes
>
> A consumer of the ontology could pick either the 3d or 4d modules and
still
> conform. A consumer could add a new domain specific ontology on
> earthmoving operations which used all the modules in the standard and
still
> conform. However, if the consumer decided to change some of the axioms in
> the units and measures ontology I would say that he or she was not
conforming.
>
> Adam
>
> At 11:12 AM 9/5/2001 +0200, West, Matthew R SITI-GREA-UK wrote:
> >Dear Adam,
> >
> >It is really quite easy, you comply where you use
> >modules from the SUO. I would expect a number of conformance
> >classes to be developed, one might be a 3D set of modules,
> >another might be a 4D set of modules. This is all standard
> >standards stuff.
> >
> >
> >Matthew West
> >Principal Consultant
> >Shell Information Technology International Limited
> >Shell Centre, London SE1 7NA, United Kingdom
> >
> >Tel: +44 20 7934 4490 Other Tel: +44 7796 336538
> >Email: matthew.r.west@is.shell.com
> >Internet: http://www.shell.com
> >
> >
> > > -----Original Message-----
> > > From: Adam Pease [mailto:apease@ks.teknowledge.com]
> > > Sent: 04 September 2001 17:51
> > > To: West, Matthew R SITI-GREA-UK; Seth Russell; SUO
> > > Subject: RE: SUO: a silly question about the new modular architecture
> > >
> > >
> > > Matthew,
> > >
> > > At 10:39 AM 9/3/2001 +0200, West, Matthew R SITI-GREA-UK wrote:
> > > >Dear Adam,
> > > >
> > > >Conceptually I think there is no difference between
> > > >having two standardised concepts/axiom sets
> > > >and choosing between them and
> > > >one standardised one and one non standardised one
> > > >and choosing to use the non-standardised one.
> > >
> > > In terms of the mechanism for performing this choice but..
> > >
> > > >There is a formal (standards) difference, but I expect
> > > >most extensions/changes to SUO to come about initially
> > > >in this way, and it seems healthy to me provided the
> > > >limits of what is standardised are known.
> > >
> > > ...this is key. At an extreme, what would conformance mean
> > > if one could
> > > accept the top term Entity and substitute one's own terms and
> > > definitions
> > > for the rest? Conformance would be meaningless.
> > >
> > > Adam
> > >
> > >
> > > >Matthew West
> > > >Principal Consultant
> > > >Shell Information Technology International Limited
> > > >Shell Centre, London SE1 7NA, United Kingdom
> > > >
> > > >Tel: +44 20 7934 4490 Other Tel: +44 7796 336538
> > > >Email: matthew.r.west@is.shell.com
> > > >Internet: http://www.shell.com
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: Adam Pease [mailto:apease@ks.teknowledge.com]
> > > > > Sent: 31 August 2001 20:37
> > > > > To: West, Matthew R SITI-GREA-UK; Seth Russell; SUO
> > > > > Subject: RE: SUO: a silly question about the new modular
> > > architecture
> > > > >
> > > > >
> > > > > Matthew,
> > > > > I think Seth's question was somewhat different although
> > > > > he'll need to
> > > > > confirm. He seemed to address changing one axiom in the
> > > standard to
> > > > > another version that he preferred. You seem to be addressing
> > > > > selection of
> > > > > one of two alternative theories that are both part of a standard.
> > > > >
> > > > > Adam
> > > > >
> > > > > At 02:30 PM 8/30/2001 +0200, West, Matthew R SITI-GREA-UK wrote:
> > > > > >Dear Adam and Seth,
> > > > > >
> > > > > >In ISO TC184/SC4 we have been doing precisely the same
> > > sort of thing
> > > > > >for at least 10 years with data models (using different sets of
> > > > > >axioms to describe things).
> > > > > >
> > > > > >A particular example is in geometry. There are two types
> > > of geometry
> > > > > >commonly used to describe engineering artifacts, BREP
> > > and CSG. Which
> > > > > >one you use depends on your purposes. It is all fine
> > > unless you think
> > > > > >that it is possible to convert from one to the other,
> > > which at least
> > > > > >generally it is not. The rest of the model can be the same.
> > > > > >
> > > > > >
> > > > > >Matthew West
> > > > > >Principal Consultant
> > > > > >Shell Information Technology International Limited
> > > > > >Shell Centre, London SE1 7NA, United Kingdom
> > > > > >
> > > > > >Tel: +44 20 7934 4490 Other Tel: +44 7796 336538
> > > > > >Email: matthew.r.west@is.shell.com
> > > > > >Internet: http://www.shell.com
> > > > > >
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: Adam Pease [mailto:apease@ks.teknowledge.com]
> > > > > > > Sent: 29 August 2001 20:03
> > > > > > > To: Seth Russell; SUO
> > > > > > > Subject: Re: SUO: a silly question about the new modular
> > > > > architecture
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Seth,
> > > > > > > I'd say that although you could do this, then you wouldn't
> > > > > > > be using the
> > > > > > > standard. A good analogy would be to another standard like
> > > > > > > XML. You could
> > > > > > > decide to define a new special character in the language if
> > > > > > > you want, but
> > > > > > > then tools that depend upon a particular syntax wouldn't
> > > > > > > work. You'd have
> > > > > > > to write new tools, thus limiting the value of the standard
> > > > > > > for your work.
> > > > > > > I believe this would hold true to nearly the same extent
> > > > > > > even if we were
> > > > > > > to adopt a theory lattice approach. Let's say that you
> > > > > worked in a
> > > > > > > manufacturing consortium that agree to use the "SUO 4d
> > > > > > > Temporal Ontology"
> > > > > > > "Structural Ontology" and "Mereology1" modules from a
> > > modular SUO
> > > > > > > standard. If you decided to change one of the axioms in the
> > > > > > > mereology
> > > > > > > module then your product might not work with others that
> > > > > > > agreed to use the
> > > > > > > standard.
> > > > > > > I don't believe that there are any existing tools that
> > > > > > > would perform
> > > > > > > this sort of theory compilation nor has anyone volunteered to
> > > > > > > build them
> > > > > > > but that's another story.
> > > > > > >
> > > > > > > Adam
> > > > > > >
> > > > > > > At 10:34 AM 8/29/2001 -0700, Seth Russell wrote:
> > > > > > >
> > > > > > > >I must admit to being new at this ... but can I ask a silly
> > > > > > > question anyway
> > > > > > > >...
> > > > > > > >
> > > > > > > >Let's say I don't like the way physical entities are defined
> > > > > > > in the ~new
> > > > > > > >proposed~ modular architecture and want to substitute my own
> > > > > > > definition, but
> > > > > > > >still want to use the rest of SUMO. How do I do it?
> > > > > > > >
> > > > > > > >For instance suppose I want to define a physical entity as
> > > > > > > something that
> > > > > > > >has *any* physical relation to some other entity. Like:
> > > > > > > >[1] (connectedVia Knee Thigh Leg)
> > > > > > > >[2] (instance connectedVia physicalRelation)
> > > > > > > >... so based on this, there would be an axiom from which we
> > > > > > > could infer that
> > > > > > > >Knee, Thigh and Leg were physical entities.
> > > > > > > >
> > > > > > > >Do I just substitute all the axioms on the node
> > > > > > > >http://ontology.teknowledge.com:8080/rsigma/SKB.jsp?req=SC&sk
> > > > > >b=Merge-WordNet
> > > > > > >&id=95
> > > > > > >with my own, and then thread through all the other axioms
> > > > > that are in
> > > > > > >libraries that I have chosen and replace the axioms that
> > > > > refer to the
> > > > > > >constants defined on that node ? Will SUO provide me with
> > > > > tools to help in
> > > > > > >this process?
> > > > > > >
> > > > > > >Seth Russell
> > > > > >
> > > > > >Adam Pease
> > > > > >Teknowledge
> > > > > >(650) 424-0500 x571
> > > > >
> > > > > Adam Pease
> > > > > Teknowledge
> > > > > (650) 424-0500 x571
> > > > >
> > >
> > > Adam Pease
> > > Teknowledge
> > > (650) 424-0500 x571
> > >
>
> Adam Pease
> Teknowledge
> (650) 424-0500 x571
>
>