RE: Recommendations: URIs

Jeff Young wrote:

>> The simpler ISBD case would output a triple  > number> "BNB number".
Assuming a one-to-one correspondence between the
>> BNB number (a "local" identifier) and the Resource instance, there
>> would be no duplication of BNB numbers.
>
> I agree this is the best way to handle all identifier schemes that
aren't URIs. The general pattern would be this:
> 
> tbox:hasXYZIdentifier a rdf:Property;
>    rdfs:comment "has XYZ identifier";
>    rdfs:subProperty of dcterms:identifier.
>
> When creating these properties, the domain and range should probably
left unspecified unless it is being coined
> by the managing agency itself.


Good idea. This should be the minimum, although it it does not ensure
any one-to-one corresponce. Such restrictions can be expressed with
owl:FunctionalProperty, owl:InverseFunctionalProperty, and
owl:maxCardinality. But you should think very carefully before adding
such constraints.


>> I would expect different answers regarding other "local" identifiers,
including ISBNs, etc.
>
> ISBNs do have a URI form (urn:isbn), so it's probably not the best
example [...]


A good example for the problem with one-to-one correspondences:


my:x my:hasISBN "0-486-22543-7" .
my:x my:hasISBN "0486225437" .
my:x my:hasISBN "9780486225432" .
my:x my:hasISBN "978-0-486-22543-2" .
my:x my:hasISBN "9 78048 622543 2" .


No uniqueness without identifier normalization. But then you can also
craft an new URI/URN namespace.


my:x owl:sameAs urn:isbn:9780486225432

As normalization cannot be simply defined in RDF, at least a broad
identifier syntax would help:


tbox:hasXYZIdentifier a rdf:Property ;
   rdfs:comment "has XYZ identifier" ;
   rdfs:subProperty of dcterms:identifier ;
  tbox:matchesRegularExpression "^[A-Z0-9]+$"


Jakob


-- 
Verbundzentrale des GBV (VZG)
Digitale Bibliothek - Jakob Voß
Platz der Goettinger Sieben 1
37073 Goettingen - Germany
+49 (0)551 39-10242
http://www.gbv.de
jakob.voss@gbv.de

Received on Monday, 2 May 2011 17:44:18 UTC