All, (01)
Since I have been asked privately, and I have some ideas, I want to
suggest a concrete plan of action that follows from the Web conference. (02)
(1) Requirements gathering
From the Web teleconference, and its references and contributions,
we have a lot of material that identifies the requirements for UoM
models that were perceived as important to various groups. We need to
turn that material into an explicit catalogue of "requirements" as a
single document.
We should probably divide the requirements into "minimal" (sine qua
non), highly desirable, and WIBNI ("wouldn't it be nice if...").
Someone needs to volunteer to edit/draft this document.
We can expect the first circulated draft to focus the attention of
this group and create some debate. There will follow further drafts. (03)
(2) Language selection
Since we are going to propose a standard ontology, it should be
documented in one or more standard languages.
Technically, we have at this time only 3 good choices: CLIF, RDF,
and OWL. But it should be noted that "OWL" is an umbrella for several
languages, one of which (OWL/Full) might be considered the best choice
for an appropriate RDF dialect. We need to make choices among these for
the normative ontologies.
These languages have very different expressive powers. I suggest
that we choose one axiomatic form and one (extended) DL form, and do all
the formal ontology work in exactly those languages.
We also need a non-normative graphical representation, to enable
rapid comprehension. The ODM Profile for OWL (using UML tools) suggests
itself, but I usually use a more vanilla UML form for presentation of
basic concepts. And something adequate that is supported by
web-available tooling (like Protegé) is a good alternative.
I suggest that we choose a useful graphical form supported by
available tools and use it, exclusively, for presentation and discussion
in the group. Further, I suggest that we will include non-normative
diagrams in this language in the proposed standard, as an aid to reader
comprehension. (04)
(3) Available ontological models
From the presentations, we know that there are basic UoM ontologies
in DOLCE, SUMO, OpenCyc, BFO, and probably others. In addition, we know
there are basic UoM models in UCUM and UnitsML.
We should identify the specific set of reference ontologies and
other knowledge sources that we will use, each supported by a named
expert who is a member of this working group. We can add to the set as
additional resources come to light.
We should extract the UoM parts of these ontologies into a
repository and look at them side-by-side. (The basic VIM concept set is
only a dozen concepts or so. It should not be necessary to include all
the infrastructure on which the UoM part is built -- the idea here is to
identify the UoM concepts that are captured.)
That means: individuals who are familiar with each ontology will
volunteer to be the "named expert", do the abstraction, and put the
result in a *document* in a Wiki repository.
(If we get wrapped around an ontology repository that only accepts
ontologies in OWL, it is of no use, because each of these extracts will
be in the language of the upper ontology.)
We should then create a table of all the distinct UoM concepts that
appear in any of these resources, with one row for each concept, and one
column for each of the resources, such that the row/column cell
identifies the "equivalent" of that concept in that resource.
(We can create the table as we go. The first document in the Wiki
populates some set of rows and one column, the next populates a second
column and possibly extends the set of rows, etc. We can later work
with the table to identify formerly unperceived "equivalents".) (05)
(4) Initial draft
Once we have a catalogue of the UoM concepts in the major reference
ontologies, all the UoM concepts that appear in all (or most) of the
upper ontologies are obvious candidates for immediate formulation in our
chosen languages. And those formulations can be guided by the
formulations in the various extracts.
What will remain are the debatable axioms and concepts, and the
issues from the requirements gathering activity that are not
satisfactorily addressed in the available work. So, when we reach this
stage, we can sort out a further program-of-work. (06)
I believe these three general activities can proceed in parallel, at
least for a while. (07)
So we have some job openings (the pay is really poor: no money, no
prestige, a lot of work, some vilification; we depend on your altruism,
or some other motivation): (08)
- chief requirements editor (1)
- vice requirements editor (1-3)
- language selection committee (5-7)
- ontology extraction manager (table management, expert goader)
- reference ontology experts (1-2 per resource)
- chief ontology formulation editor (1)
- vice ontology formulation editor (1-2)
- Working Group co-Chairs (2), or Chair and Vice Chair
- Wiki manager (1) (09)
Agreement? Suggestions? Alternative proposals? (010)
Volunteers? Resignations? ;-) (011)
I got the impression that Frank Olken and Howard Mason were nominated as
the WG co-chairs. But someone can correct that. And I will leave to
Peter the identification of his position in the WG. (012)
I will nominate myself as a "requirements editor". I will be happy to
be an Indian ("vice editor") if someone else wants the big headdress. (013)
-Ed (014)
P.S. If this sounds like a W3C procedure, it is not accidental. (015)
--
Edward J. Barkmeyer Email: edbark@xxxxxxxx
National Institute of Standards & Technology
Manufacturing Systems Integration Division
100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528
Gaithersburg, MD 20899-8263 FAX: +1 301-975-4694 (016)
"The opinions expressed above do not reflect consensus of NIST,
and have not been reviewed by any Government authority." (017)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/uom-ontology-std/
Subscribe: mailto:uom-ontology-std-join@xxxxxxxxxxxxxxxx
Config/Unsubscribe: http://ontolog.cim3.net/mailman/listinfo/uom-ontology-std/
Shared Files: http://ontolog.cim3.net/file/work/UoM/
Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?UoM_Ontology_Standard (018)
|