Re: Dynamically assign identifiers - UMD response

On Tue, Jul 12, 2005 at 08:32:57AM -0400, Eric Prud'hommeaux wrote:
>     Split the label space of bNodes:
>         { _!:xyz foaf:homepage ?hp }

Ron sent our reasons for preferring this solution in another email message.
(I reiterated them in yet another message.) There are other considerations
in addition to OWL DL safety.

> For instance, DL restricts triples involving anonymous classes.

In Dan's view, it won't matter, since he's argued that if a term acts
substantively like a URI, then it's not anonymous in some important sense
(which is, I think, a legitimate position -- I just don't think you *have*
to see it that way).

I'm satisfied with a purely syntactic hack (I won't say "sugar"), but others
don't seem to be.

> about them (or, at least, your view of them). Does that change if the
> addressing mechanism is in a FILTER, protocol association, or special
> _!:xyz node?

I don't know. There are other reasons why filter & protocol solutions aren't
ideal in our view (basically, skepticism about implementation acceptance and
excessive generality & power, respectively).

Kendall Clark

Received on Tuesday, 12 July 2005 13:04:04 UTC