W3C home > Mailing lists > Public > semantic-web@w3.org > July 2007

Re: RDFON: a new RDF serialization

From: Henry Story <Henry.Story@Sun.COM>
Date: Fri, 27 Jul 2007 17:26:26 +0200
To: Garret Wilson <garret@globalmentor.com>
Cc: Semantic Web <semantic-web@w3.org>
Message-id: <1EE14FA2-9740-4543-A1FD-A682550E1E73@Sun.COM>



On 27 Jul 2007, at 16:39, Garret Wilson wrote:

> Henry Story wrote:
>>
>> But I think literals do serve a purpose as opposed to URLS. They  
>> are things that refer to themselves, as opposed to URI's that  
>> refer to other things.
>
> The literal "value"^^xsd.string refers to itself, but that's a  
> special case. The literal "123" in many contexts does not refer to  
> the string "123", but refers to the value 123---we just happen to  
> be using a set of digits in a particular language to describe that  
> value. The literal "123" is no different than the literal "one  
> hundred and twenty-three", if we were decide to use it to represent  
> the value 123---it's just another representation in English.

I think you missed my previous comment that

"123"^^xsd:integer .

is really syntactically very similar to

"123"^xsd:integer .

which is just short hand for

_:123 xsd:integer "123" .

Which is to say that there is a thing, which has relation xsd:integer  
to the string "123".
my guess is that the relation xsd:integer, is inverse functional and  
functional.

Now I am not sure if in this case the blank node refers to itself. I  
suppose some (Bertrand Russle for example) would say it refers to the  
set of sets of size 123, just as 2 refers to the set of pairs, and 3  
refers to the set of triples. But I am not sure how one should think  
of it in rdf.


> We could use Morse code "_...." (it appears that the international  
> version is "___.."---they're different, surprise) to represent the  
> value 8, but there's still no self-reflection going on. Literal  
> strings are just identifiers---no more than a URI, with an added  
> data type to indicate that the lexical form is only unique within  
> the domain indicated by the datatype. Which is leading me to think  
> that the correct representation of literals in a graph is something  
> like <rdfliteral:123;xsd:integer> as the URI of a normal resource,  
> with maybe an automatic rdf:lexicalForm property as David mentioned.


Well that's ok

_:123 morse:intcode "_...." .

would be ok.

The we would have

[] morse:intcode "_...." ;
    xsd:integer "123" .


Perhaps I was wrong to think that literals refer to themselves.  
Perhaps that's just strings that do that... I have not thought about  
that in detail.

Henry


>
> Garret
Received on Friday, 27 July 2007 15:26:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:41:58 UTC