On 4/10/12 11:26 PM, Pete Johnston wrote:
> For some other examples, see the W3C XML Schema Datatypes specification
> which uses the terminology of "lexical space" ("set of strings") and
> "value space" ("set of resources").
> I often think the Boolean example is a nice one to consider, because
Unfortunately, our use case isn't Boolean, so I am trying to understand
how we apply this to our use case, for example Gordon's case #1:
which is currently declared to be an SES in the Open Metadata Registry.
An example of a publication statement, as it appears today, is:
New York, Bowker, 1987
place of publication: New York
The order is fixed, and any of the parts can be repeated, although again
with a fixed order. In library data today, repeated elements \= repeated
place of publication: London, New York
publisher: Penguin, Bowker
but we don't know how that will be handled in the future.
> (i) both the lexical space and the value space are "small", so one can
> easily "work through the whole set"
> (ii) the concepts of "true" and "false" are fairly easy to grasp as
> two distinct concepts
> (iii) the act that two different components in the lexical space
> ("true" and "1") both map to a single component in the value space
> (true-as-concept) (and similarly for the "false" case), (I think)
> quite neatly illustrates the lexical space/value space separation and
> the idea that a mapping is being made.
[log in to unmask] http://kcoyle.net