RE: Moving forward with ISSUE-30 (IRI template expansion)

On 15 Aug 2014 at 11:20, Ruben Verborgh wrote:
>> As an 'off-the-top-of-my-head' suggestion - has anyone considered
>> asking the RFC6570 authors or 'community' about extending it to
>> support what we require here? Being fully RFC6570-compliant while
>> also getting datatypes and language support would be great, no?
>
> I think it was deliberately out-of-scope for RFC6570. I.e., this RFC
> assumes that the application has already decided on a datatype-to-
> string conversion method for use in the template. That makes it most
> broadly applicable.

I agree. This is out-of-scope for RFC6570.


> Before I brought this up on the mailing list, people here seemed to
> assume the same. Because indeed, it is straightforward for most
> things. 30 is just "30" and "Paris" is "Paris". And usually there
> are other ways to indicate languages (i.e., another field).
>
> However, since some applications (like mine) might have a stricter
> RDF domain where there is a difference between URIs and strings with
> the same value, it makes sense to define this. But I think it really
> belongs in Hydra, not RFC6570b.

Yep, RDF is a special case here I'd say.. but since we are based on RDF we
obviously have to specify how it is intended to work.


--
Markus Lanthaler
@markuslanthaler

Received on Friday, 15 August 2014 15:46:19 UTC