SUO: RE: should a standard specify...
Dear Jim,
A nice demonstration of some of the issues I was trying to
draw to Jean-Luc's attention.
I will cut out most of the message and try to give some responses
against the summary at the end.
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
> SUMMARY
>
> * I see some difficult practical issues of how a human or a machine
> can determine whether (certain subdomains of particular) ontologies
> stand a chance of being able to interoperate with (certain
> subdomains of)
> other ontologies.
MW: Yes, I think this is difficult. In ISO 18876 I have tried to set
down some basic elements of what I think you need to do.
>
> * The main difficulties seem to be: (1) how to identify the relevant
> ontological primitives (e.g., terms, definitions, axioms)
> used by an ontology;
> and (2) how to use these primitives to extract a particular
> domain-related
> portion of the ontology.
MW: In general it is a reasonable bet that all concepts in an ontology
are primitive (in that ontology) unless they are explicitly defined in
terms of other concepts (e.g. intersection of sets). For example, take
some subtypes of vehicle, like lorry, car, motorcycle, van, coach.
If these are not subtypes of something else as well, then they must have
some concept hidden in them that differentiates them from other vehicles,
and that concept is (for this ontology) primitive, i.e. not further
defined. On the other hand, elsewhere in the ontology, there may be the
concept of "red car" which is defined as the intersection of "red things"
and "car". Now you know that "red car" is not primitive.
MW: One of the problems is that it is quite arbitrary what an ontology
might choose to take as its primitives. Further most ontologies do not
really concern themselves with which concepts are primitives. They are
much more concerned about which concepts are useful for the problem
they are designed to help with.
MW: Since there is no reason why ontologies should choose the same
primitives, it is prudent to assume that they have none, or only a few
in common.
>
> * Being able to identify, extract, and manipulate these
> primitives _seems_
> key to being able to assess whether two ontologies can interoperate
> (in virtually any plausible sense of that word). But it may
> indeed not be
> appropriate to suppose that one could determine whether two
> ontologies
> can interoperate based on accessing the ontological primitives and
> assertions used by the ontologies to talk about a particular domain.
> To those who hold this view, can you please elaborate?
MW: I certainly do not think this is sufficient, for the reasons I give
above, i.e. that it is very unlikely that two ontologies would share the
same
primitives, even if they were about the same area. The result is that
you need to do some reasoning about how the primitives of one ontology
relate to the primitives of another ontology.
MW: I find I can do this manually, but I could not define how to automate
this process. For me this is the process of integrating the ontologies,
essentially generating the mapping between them of what corresponds.
>
> * Currently, as shown by the examples of SUMO, Cyc, and
> ISO15926, different
> ontologies have different ways of representing these
> primitives, and in many
> cases it is difficult to determine from the ontology itself,
> how these
> primitives are identified and can be made available. Currently, these
> determinations must be done manually (I think) and in an ad hoc way.
MW: I agree, this is an art at present.
> Perhaps the SUO standard can address how to accomplish make such
> determinations programmatically.
MW: As we gain experience with the art, we may learn how this can be
(semi) automated.
>
> * It is possible that a single, agreed-upon ontology language
> will go a long
> way in helping to identify and make available these ontological
> primitives and the subsequent extraction of domain-specific
> subontologies.
MW: Not really. It just removes one small barrier. Indeed, creating
cross compilers into a common language would be relatively easy.
> But it seems that some specific mechanism that is different from a
> common language, and which belongs perhaps to the ontology's
> specification,
> needs to be provided. Such a mechanism needs to be, well,
> mechanical -- capable
> of being done by a machine with limited sets of seeded input.
MW: Well yes, if all the concepts in an ontology were first defined
in terms of the concepts of some integrating ontology, then it would
be easy. The problem is now to create the integrating ontology ...
>
> * What that mechanism is, what the ontological primitives are, and how
> this mechanism can be used in practice to identify and make
> available these
> primitives and to allow certain subontologies to be extracted
> -- I suggest
> that all this should perhaps be investigated in the context of our
> standards work.
MW: This would certainly be useful.
>
> * If these issues have already been addressed and solutions for them
> have been found, please let me know.
MW: And me too.
>
> --------------------------------------------------------------
> ---------------
>
> Any and all comments appreciated.
>
> Thanks,
>
> Jim
>