On Fri, 2004-04-30 at 16:14, Steve Pepper wrote:
> * Kal:
>
> | As a concrete example, lets say that I want to state that the subject
> | identifier "http://psi.techquila.com/techquila"; was created by Kal on
> | 29/04/2004
>
> You want to make an assertion about a subject identifier?
>
> In that case the subject identifier is your subject. You
> simply create a topic that represents it and then you can
> make whatever assertions you please.
>
> As to the identity of that subject, I would recommend
> creating a subject indicator/identifier pair, e.g.
> "http://psi.techquila.com/techquila-psi";.
>
Yes, thats one way I thought about doing this. What it lacks is some
sort of pointer back to the identifier being reified. I suppose that the
thing to do here is to recognise that an identifier is a non-addressable
resource, therefore you can only refer to it (in topic maps) by
reference to a subject indicating resource. In terms of interop of topic
maps, it would be good if there were some convention we could use to tie
identifier metadata back to the identifier (e.g. a conventional
occurrence type to apply to an occurrence whose value is the identifier
being reified). This has practical consequences for topic map-based
identifier registries, so it would be a worthwhile thing to do, no ?
Cheers,
Kal
--
Kal Ahmed <kal-5x+ggncFKvmB+jHODAdFcQ@xxxxxxxxxxxxxxxx>
techquila
|
