Re: BPR: Is redirection permitted?

On 10/15/12 2:22 PM, Andy Seaborne wrote:
>
>
> On 15/10/12 17:00, Kingsley Idehen wrote:
>> You end all of this confusion by loosely coupling RDF, EAV, and Linked
>> Data.
>
> A loose coupling (a meta-architecture) does not give interoperability.

The Web has a lot of interoperability and the whole thing is loosely 
coupled.


We should use RDF (which is about data representation) to obscure Linked 
Data which is about the combination of data representation and data 
access. These matters are related but not one and the same. As you know, 
modulo LOD and DBpedia, a lot of RDF documents out in the wild don't 
even conform to the key principles espoused in TimBL's Linked Data meme.

RDF is an optional (and W3C recommended) mechanism for creating Linked 
Data. That doesn't make it the sole option or an adjudication mechanism 
when the issues at hand have more to do with Linked Data specifics. 
Simple example, de-referencable URIs. That really has nothing to do with 
RDF.

> It gives a set of design principles, but it needs another step to 
> decide on the concrete instantiation of those principles.  A 
> meta-architecture gives portability of skills and ideas.
>
> The LDP needs to choose the concrete instantiation so that 
> interoperation at the byte-on-wire level happens, otherwise it's just 
> one more set of guidelines.

I don't see why or how loosely coupling RDF and Linked Data impedes that 
goal :-)


>
>     Andy
>
>
>


-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen