W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > October 2002

Re: n-triples for datatype values [was: Agenda for RDFCore WG Telecon 2002-10-18]

From: Tim Berners-Lee <timbl@w3.org>
Date: Fri, 1 Nov 2002 05:29:09 +0100
To: w3c-rdfcore-wg@w3.org
Message-Id: <773321D2-ED52-11D6-ABEB-000393914268@w3.org>
> [Patrick Stickler, Nokia/Finland, (+358 40) 801 9690,  
> patrick.stickler@nokia.com]
>
> > To me, using "^^"   makes it clear that ^^ is a syntactic thing
> > whose semantics are in fact equivalent to "^"  except that
> > the formal triples representation is different.
> >
> > So Jos, you can if you want dismantle the triple into two.
> > You will have a semantically equivalent graph.
>
> Well surprise surprise. I guess my suspicions about ^^ were correct.
>

If you suspicions were that RDF was to perverted by the addition of  
extra triples in the definition of an RDF parser then you were wrong.    
Don't panic!


> I reiterate my opposition to the use of ^^ in the abstract syntax.
>

I find the use of juxtaposition very messy for the parser, and  
potentially
confusing for users.  It is much safer in the syntax to use a piece of  
punctuation.
That syntax point is completely irrelevant to question in the rest of  
your message.


> A typed literal node may *not* be "dismantled" into
> additional triples, even if it might be deemed to be semantically
> equivalent to an expansion into a bnode with datatype property
> (and I am not convinced that it is).

I am sorry, I clearly didn't distinguish well enough between things you
do inside a parser and things you do outside.
I was suggesting that one could do what Jos wanted outside the parser.

If you are not convinced that, for a given datatype, a property can  
relate a member of the value space and a member of the lexical space,  
then you must have thought of something I haven't thought of.


> If an application wishes to define rules to infer those additional
> triples, fine,

That is just what we are talking about here.  You can't stop Jos  
treating his data
in that way.


>  but the ^^ delimiter does not function in any way
> like ^ in N3.
>

Exactly it does not.   It is syntax in the RDF spec, not a triple

If they had ben the same, then I would have suggested ^ not ^^.

> I would like either for the delimiter to be removed entirely or
> for there to be an explicit statement that such "dismantling"
> of the typed literal node is not licensed by the RDF specs.
>

The RDF spec's job is to define the set of triples which corresponds to  
a given serialization.
Not to define what people do after they have got them.  Do not blur the  
line.
I was saying that Jos could do them *after* the RDF parsing stage.
This draws away from RDF spec some criticsism of it being clumsy,
demonstrating that it can be converted into a different form.

IMHO
Tim

> Patrick
>
>
> ----------------------------------------------------------------------- 
> -
> *	Next message: Patrick Stickler: "Issuette for tomorrow's aggenda"
> *	Previous message: Patrick Stickler: "Re: Datatyping literals:  
> question and test cases"
> *	In reply to: Tim Berners-Lee: "Re: n-triples for datatype values  
> [was: Agenda for RDFCore WG Telecon 2002-10-18]"
> *	Next in thread: Dan Connolly: "Re: n-triples for datatype values  
> [was: Agenda for RDFCore WG Telecon 2002-10-18]"
> *	Reply: Dan Connolly: "Re: n-triples for datatype values [was: Agenda  
> for RDFCore WG Telecon 2002-10-18]"
> *	Reply: pat hayes: "Re: n-triples for datatype values [was: Agenda  
> for RDFCore WG Telecon 2002-10-18]"
> *	Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
> *	Other mail archives: [this mailing list] [other W3C mailing lists]
> *	Mail actions: [ respond to this message ] [ mail a new topic ]
Received on Thursday, 31 October 2002 23:28:46 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:52:36 EDT