I agree. I would add that for the most part, the terms needed in a given
industry are already owned by communities of practitioners and standard
bodies in that industry, so all that's needed is to offer to those
groups a way to enhance their work with an ontology which they would own
and manage. Whether that lives on their own servers with links to some
common management framework, or lives on some commonly managed resource
with links to their existing standards materials (e.g. messages and so
on) will depend on where each of those groups is currently at. (01)
So for example in the financial industry, once the ISO 20022 version 2
interbank messaging standard is available (which it isn't yet), this
will have an explicit semantic component defined in the standard. The
content will be managed by a Registration Authority appointed by ISO,
likely to be someone like SWIFT. The methods for translating the
semantics to various physical message formats will almost certainly
follow the OMG's MDMI standard (I just came off a call with SWIFT and
OMG on this) and the semantics model itself will be either in native OWL
or ODM, most likely ODM. (02)
So assuming that the financial industry is fairly typical, the question
is not one of finding some green field site where no-one has heard of
semantics and offering to create a great ontology, it's a question of
providing some way of inter-relating the more basic concepts from which
the semantics models in each industry are built, so that you end up with
ontologies that have some formal relations among them. For instance in
the current draft semantics model for the financial industry we have had
to use a lot of primitive concepts for financial (based on XBRL), time,
geography, math, legal and so on. Most of these we've had to make up for
ourselves but would expect to be able to find an interoperable and
respected world of semantics models from which to draw these concepts.
This is the missing piece at the moment. (03)
In other words it's in the formal semantics inter-relations that I would
see this community being able to add something, not so much in the idea
of modeling semantics for people. Each industry will want to do that,
and there has been a huge explosion in the appreciation of semantics at
least in the financial industry, and probably in others. (04)
I don't know if that is in line with how others see things here? (05)
Mike (06)
John F. Sowa wrote:
> Ron,
>
> I want to point out that my proposal requires a minimal amount
> of funding to get started.
>
> > I am not suggesting that we should build a full OR. I believe
> > that there is a project currently underway. I hope that the
> > functional requirements that are outlined below will be considered
> > in that process.
>
> I am not proposing that we begin by developing *any* ontologies.
> The starting work requires less effort than this group puts into
> a hotly debated email thread:
>
> 1. Define the operators that relate the theories in the hierarchy.
> Adolf Lindenbaum kindly did the theoretical work for us about
> 80 years ago, and I summarized it in Sections 2, 3, and 4 of
> the following paper:
>
> http://www.jfsowa.com/logic/theories.htm
>
> 2. The next point is to develop a set of policies about how
> to handle contributions, relate them, evaluate them, etc.
> As Elisa said, the OBO has a well developed framework that
> does much of what we need, and we could start by identifying
> what they have accomplished that we can adopt, and what more
> we would like to add.
>
> After these two points have been established (or even during the
> debate about them), Peter Yim or anybody else on the list that
> wants to do so could set up a Foundation Ontology wiki.
>
> We would also need a cute logo, design graphics, and a URL with
> an appropriate name that is dedicated to the Foundation Ontology.
> That can also be set up in parallel.
>
> The ontologies themselves would come from donations. Some of
> those that are already available as open source could be adapted
> very quickly by adding the appropriate metadata and making a
> place for them in the hierarchy. They need not be physically
> moved from their starting places, but we do need to establish
> some controls for versioning, etc., which are often minimal
> or nonexistent in open-source resources.
>
> My point is that we can begin this work today with mostly
> volunteer effort. If we do a decent job, the funding will come
> later. But we need to do something solid to demonstrate that
> this group is capable of accomplishing something. Otherwise,
> hope for multimillion-dollar grants is a pipe dream.
>
> John
>
>
> _________________________________________________________________
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
> Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
> Shared Files: http://ontolog.cim3.net/file/
> Community Wiki: http://ontolog.cim3.net/wiki/
> To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
> To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx
>
>
>
> (07)
--
Mike Bennett
Director
Hypercube Ltd.
89 Worship Street
London EC2A 2BF
Tel: +44 (0) 20 7917 9522
Mob: +44 (0) 7721 420 730
www.hypercube.co.uk
Registered in England and Wales No. 2461068 (08)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (09)
|