cygri / void

An RDF schema and associated documentation for expressing metadata about RDF datasets
http://www.w3.org/TR/void/
14 stars 1 forks source link

Pick a Range for void:uriSpace #92

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago
Should a void:uriSpace value be a literal or a URI ?

I think it should be a literal. My understanding is that it is a string  onto 
which other strings can be concatenated to make full URIs, and that it does not 
"identify" a resource. 

People will want to do string-like things with it (concatenation, comparison), 
not URI-like things with it (identity, dereferencing).

A problem resulting from this is:

1. A publisher describes their dataset like this:
<http://data.example.org/> a void:Dataset ;
    foaf:homepage <http://data.example.org/html> ;
   void:uriSet <http://data.example.org/> ;
   void:exampleResource <http://data.example.org/things/1> .

2. Another dataset publisher wants to declare a Linkset linking to to 
http://data.example.org/, but does not know the URI, but knows the 
foaf:homepage, so following the advice in voiD guide 6.1 does:

<http://foo.com/void.ttl#dataExample>
    foaf:homepage <http://data.example.org/html> .

3. LATC dataset inventory aggregates both descriptions and smushes on 
foaf:homepage, resulting in:

<http://foo.com/void.ttl#dataExample>
   a void:Dataset ;
   void:uriSet <http://foo.com/void.ttl#dataExample> ;
   void:exampleResource <http://data.example.org/things/1> .

4. Confusion

Original issue reported on code.google.com by K.J.W.Al...@gmail.com on 14 Dec 2010 at 1:21

GoogleCodeExporter commented 9 years ago
Duplicate of Issue 91.

(I'd mark 91 as a duplicate of this here, but the spec links to 91 in Section 
4.2...)

Original comment by richard....@gmail.com on 12 Jan 2011 at 3:19