On Fri, May 31, 2013 at 2:04 PM, Ron
Wheeler
<rwheeler@xxxxxxxxxxxxxxxxxxxxx>
wrote:
On 31/05/2013 1:31 PM, William Frank wrote:
....
RW
Can one actually construct a "wine ontology"
that will be equally meaningful in both
contexts? And equally convenient to build and
maintain?
WF
*As I believe and understand several others to
believe, one wants to specify a generic, way
underspecified wine ontology that needs to be
extended in different ways to suit the needs of
particular domains of endeavor, such as the two
excellently different ones you introduce. With,
among its goals, the ability to related the
different domains to each others.
RW
I look forward to see
how this evolves.
I would be delighted to find that my misgivings are
misplaced and that there is a way to define something that
is both true and useful in a wide range of use cases.
WF
Well, a high level, highly underspecified ontology is not
useful ***in itself**. It is a framework to start from, and
extend in different directions in different domains.
As an extreme example, consider arithmetic, which is both true
and useful in a VERY wide range of use cases.
For the trader and the back office, for the depository, for
the regulator and the investor, for the issuer and his
investment bank, a 'security' has quite different most
important properties. (And, I have noticed, they frequently
dismiss as unimportant the properties that are not important
**to them**). Yet, a careful analysis can capture a generic
model of what a security is most essentially, then specialized
along different dimensions to capture what is important to
different stakeholders, with different concerns.
Consider marriage. People in different cultures have very
different mores involving marriage. And yet, the seem to have
a way of identifying that they are all talking about the same
thing: marriage, (while generally disapproving of the
practices that are different from their own.) So, a domain
ontology for human kinship, that underspecified marriage, but
captures what is common, to all these practices, and then
adding the details for specific cases, and more details for
the different kinds of things, that, as in your example,
different stakeholders might want to know about marriage, even
in the same culture. There are some sticky points with this,
such as the fact that many concepts drift so that at one end,
there is nothing in common with another use at the other end
of the drift, even though they are tied together, one use to
the next (called family resemblance, when contemporaneous
uses. But by and large, this provides a good working
approach.)