xml-dev - Re: local, global (was various ontology, RDF, topic maps)

[Jonathan Borden <jborden@mediaone.net>]

Simon St.Laurent wrote:
> At 11:27 AM 12/20/00 -0500, Jonathan Borden wrote:
> >>
> >    This has been the problem of structured medical terminology for the
past
> >3 decades and today we have no universal agreement about and less
structure
> >in medical records. And then the government steps in and says "We aren't
> >requiring you to do it this way but if you don't you won't get paid."
That
> >is the best way I know of achieving consensus. My proposition is, simply
> >put, that while we are cleaning up the medical claims procedures, we
seize
> >the opportunity to fix the information system.
>
> I'm aware that this is what _is_ happening, but are you really convinced
> that it's a good idea to enforce 'universal agreement' throughout the
> process?  I'm not sure ontology by fiat is blessed with any particular
> exemption that makes it a better idea overall.  Is 'seizing the
opportunity
> to fix the information system' an even larger project than dealing with
> HIPAA itself?  I suspect that interchange formats are a difficult enough
> problem by themselves - they generally seem to be.

    I don't believe that even though probably possible, it would be either
good, or practical, to enforce a universal medical vocabulary. What I expect
to occur is a patchwork of ontologies knitted together by "type",
"subClassOf", "equivalentTo" arcs etc much as the web itself is knitted
together with hrefs.

    In medicine there do exist large areas of substantial agreement what we
call "standard practice" and so we begin by standardizing what is already
standard, the difference is merely that we define an XML document standard
and readily demonstrate how this can be displayed or printed in a familiar
format via a stylesheet. With a little care this data format can be parsed
as RDF.

>
> >    Again the principle of reification, or in plain English the ability
to
> >make assertions regarding the classifications allows us to place these
> >ontologies into "contexts" or "spaces". One's regard for a particular
> >ontology is then applied to its context. As an individual, I can use only
> >those ontologies in a particular "context" and I can apply a belief value
> >(i.e. create a Bayesian network) to individual statements or entire
> >ontologies as I choose.
>
> That makes it sound wonderfully simple, but I don't think we're anywhere
> near the point where such 'reification' is the answer.  "As an individual"
> is a concept which has almost no respect in the world of software
> development and systems integration, I fear, and is a privilege currently
> available only to systems developers, not system users.

    True. To me the concept of a "service" implies that everyone gets the
same thing from a given input. The concept of an "agent" implies something
more individual. While we all exist on the same playing field, I might have
the capability of giving "my" agent "my" instructions. The semantic web
whose foundation is a feltwork of ontologies and instances exists as a
datastructure through which an agent might navigate. Currently the "user
agent" is a browser that responds to my input, but suppose this input were a
task e.g. "Purchase a Leica M6 at the best possible price" ... Rather than
go to a service at a site which I might not trust to get me the actual
"best" price, what I really want is for the information to be available on
the network and for my user agent (which perhaps I trust more than
verybestcameraprice.com ) to be able to search and then negociate prices
based upon my directives. I still need to trust the information on the web
but as an individual I want more control of the processing.

>
> > [...much good technical which does nothing to ease my concerns...]
> >
> >Again, think of RDF as an assembly language for semantic reasoning, but
> >realize that real benefits won't be achieved until we get high level
> >languages and tools.
>
> I'd like to see higher-level languages and tools which make these concepts
> available to developers other than ontology specialists, but I have
serious
> doubts about 'real benefits' which rely on large-scale agreements, RDF or
not.

    XML itself is an example of such a large scale agreement, and until the
things we are talking about are available it is all speculation about how
useful it will ultimately be. On the other hand such agreements will only be
as useful as we make them.

Jonathan Borden
The Open Healthcare Group
http://www.openhealth.org