Re: SUO: Multi-Source Ontology (MSO) Draft Ballot Question
Adam,
The primary purpose of the Multi-Source ontology is
to support the development and integration of the
various resources and theoretical proposals that
the SUO group has been talking about for nearly
4 years.
Right now, we have several "starter" documents
that have been approved. Those include several
resources:
SUMO, OpenCyc, and Matthew's Express-based
ontology.
There are other resources that we have discussed
and many of us are using, but they are not yet
included in the officially approved starter set:
WordNet, Dolce, and my KR ontology.
Then there is the IF Framework, which was the
first "starter document" to be approved, and
which provides the theoretical methods that
can accommodate all of the above, in principle,
but only parts of which are implemented.
We have also approved a motion to support multiple
ontologies, such as all of the above, and integrate
them by taking advantage of some or all of the
theoretical work that has been done with IFF.
What we are lacking is an implemented set of tools
that can take advantage of the resources and the
theoretical work and integrate them into a single
usable package.
Now, Philippe Martin has come along with well-
developed a tool set that can be used -- and which
he did use -- to integrate multiple ontologies.
That is an important step toward the goals of the
SUO Working Group. It is something that everybody
has been talking about, but Philippe is one of the
few people who has actually done the hard work.
MSO might not yet be complete, but Philippe has shown
a willingness to do two things that most of the other
people have only talked about:
1. Build actual tools.
2. Use the tools to integrate the resources that
other groups have contributed.
Some people have done parts of #1 or #2, but Philippe
has done both, and he has promised to do much more of
both along the lines that the SUO WG is talking about.
Summary: The MSO does not replace SUMO, OpenCyc,
the Express ontology, the IF Framework, Dolce, or
many other things. But it is the first serious
attempt to put them together by someone who is
ready and willing to do the really hard work.
I believe that we should take advantage of what Philippe
has already done on the MSO and work together as a group
to assist him in adding whatever any of us thinks should
be added to it. If MSO doesn't yet include the SUMO
axioms, then we should put them on the to-do list for
the next version of MSO.
John Sowa