Re: Profiling in the News: Was: RE: Taxonomies, URN's etc..

["Sean B. Palmer" <sean@mysterylights.com>]

> 3.  The internet is only one system with respect to identifiers.
> Name and location are not converging because all systems
> are not internet systems.   URIs aren't universal.

I vehemently disagree; anything that can have identity can be mapped into
URI space, whether network retrievable or not. There are many myths behind
URIs that need exploding, but the non-universal one is probably the
largest.

I agree that sometimes it is better to talk about something using their
properties rather than a single node, but you can't tell me that this rules
out any form of skolemization. That's absurd. If you can talk about it in
language, then you can talk about it using URIs.

> 4.  The uniqueifying system has to attach it's own identifying
> system to use HumanML or any dataset.  That is why there is an
> SSN in there and a PDID.  Neither are HumanML datatypes.
> Nor is a URI.  The URI only comes up because RDF depends on it.
> That is a weakness of RDF.

This is a common misconception. RDF depends on neither URIs or URI
references, it depends upon resources. These resource do not necessarily
have to be identified within a system, e.g. they could be unlabelled
existentially quantified (anonymous) nodes.

The process of identifying resources on the Semantic Web is made easier by
using URI references, but it is also useful to talk about some resource as
an existentially quantified node using a set of proerties that relate to
it... some of which may be unambiguous, either individually, or when used
as a combination of properties.

RDF does have weaknesses (of course), but it is wholly incorrect to level
this one at it.

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
:Sean :hasHomepage <http://purl.org/net/sbp/> .