Yet another approach (maybe less ugly, maybe not) would be to define a
Property which acts like a curried function consuming one argument
(say, Chicago) and producing another Property (distanceToChicago),
which, when applied to NewYork, gives you your distance:

place:Chicago foo:funkyDistanceFunction bar:distanceToChicago .
place:NewYork bar:distanceToChicago 1000 .

Of course, it might be nice if bar:distanceToChicago were a blank
node, but then RDF wouldn't let you use it as a predicate.

Josh


On 1/31/07, David Karger <[EMAIL PROTECTED]> wrote:
> There are lots of ugly ways to squash non-binary relations into binary
> ones; if you want a symmetric one you can do it by creating a new
> "distance measurement" resource U, with the following 4 statements:
>
> U type distanceMeasure
> U distance "1000 miles"
> U endpoint "New York"
> U endpoint "Chicago"
>
> Kelly Jones wrote:
> > What's the natural way to model "The distance between Chicago and New
> > York is 1000 miles" in RDF?
> >
> > The only way I could come up with is ugly and really models "Chicago
> > is 1000 miles distant from New York" (same semantic meaning, but makes
> > Chicago seem more important, instead of indicating a symmetric relation).
> >
> > My ugly model:
> >
> > Chicago has_property DistanceRelation#1
> > DistanceRelation#1 has_target "New York"
> > DistanceRelation#1 has_distance "1000 miles"
> >
> > In other words, Chicago has a property called "DistanceRelation#1",
> > and DistanceRelation#1 is an object that represents a distance of 1000
> > miles from New York. Of course,
> >
> > Miami has_property DistanceRelation#1
> >
> > might hold as well, so DistanceRelation#1 doesn't really seem to
> > encapsulate what I want.
> >
> > Is there a cleaner way to do this in RDF?
> >
> > (I realize Chicago/Miami aren't 1000 miles from New York-- this is
> > just an example).
> >
> >
> _______________________________________________
> General mailing list
> [email protected]
> http://simile.mit.edu/mailman/listinfo/general
>
_______________________________________________
General mailing list
[email protected]
http://simile.mit.edu/mailman/listinfo/general

Reply via email to