Re: The rdf:JSONLiteral datatype

Hi Dominik,

(This is a personal response and I'm not speaking for the WG.)

It seems there are number of obvious objections that could be raised against introducing such a datatype and I think they would need answering (for example in a “Rationale” section attached to the proposal).

 • What is the use case for this datatype?
 • Are there examples of systems that currently use JSON literals in RDF literals?
 • Are there examples of currently published RDF data that use such JSON literals?
 • Why isn't xsd:string sufficient for representing JSON literals?
 • Given that anyone can define new RDF datatypes, why should RDF-WG do it?
 • Why is this within the scope of RDF-WG's charter [1]?

Best,
Richard


[1] http://www.w3.org/2011/01/rdf-wg-charter


On 17 May 2012, at 10:39, Dominik Tomaszuk wrote:

> Dear RDF-WG members,
> 
> I propose a new JSON datatype. Here is my proposal:
> An IRI denoting this datatype
> http://www.w3.org/1999/02/22-rdf-syntax-ns#JSONLiteral
> 
> The lexical space
> is the set of all strings which are self-contained JSON-text [RFC4627];
> 
> The value space
> is a set of JSON objects or arrays [RFC4627]. Two objects or arrays A and B are considered equal if and only if the JSON method and == operator JSON.stringify(A) == JSON.stringify(B) returns true [ECMA262-5].
> 
> The lexical-to-value mapping
> Let jsontext be a JSON object or array [RFC4627] corresponding to the literal's lexical form
> Return JSON.parse(jsontext) [ECMA262-5]
> 
> [RFC4627] http://www.ietf.org/rfc/rfc4627
> [ECMA262-5] http://www.ecma-international.org/publications/files/ECMA-ST/Ecma-262.pdf (mainly section 15)
> 
> Best regards,
> 
> Dominik Tomaszuk
>

Received on Thursday, 17 May 2012 18:23:36 UTC