On 24/05/11 13:29, Damian Steer wrote:
On 24 May 2011, at 10:57, Andy Seaborne wrote:
This is the latest proposal for adding datatypes for all literals:
http://www.w3.org/2011/rdf-wg/wiki/StringLiterals/LanguageTaggedLiteralDatatypeProposal
It obviously affects Jena because current code was written when plain
literals didn't have datatypes.
Reactions?
Difficult to grasp at first reading.
The only real difficulties I've encountered with plain literals were
in SPARQL (certain functions catch people out), and of course the
dreaded xsd:string.
As far as I can tell this will make SPARQL functions a bit more
regular around literals, and have a little impact on (de)serialisers.
I don't think it will have any actual impact on jena stores or even
APIs, will it? We'd lose some null return values (hurrah).
Some persistent stores might be invalidated without some compensating
code - TDB would be safe though (I think).
Andy
Not huge benefits, but not huge disruption either.
(I'd prefer xsd:string to be deprecated and no other changes
made.)
+1, that seems a more efficient use of the working group's time.
Damian
