Re: The need for RDF in Linked Data

On 6/17/13 9:44 AM, Luca Matteis wrote:
>
> On Mon, Jun 17, 2013 at 3:33 PM, Karl Dubost <karl@la-grange.net 
> <mailto:karl@la-grange.net>> wrote:
>
>     Nobody will disagree with that. BUT it is the URI which creates
>     the Web.
>
>
> Nothing in the thing you cited suggests that URI is the only 
> technology which "creates" the Web. Like, what do you mean by 
> "creates"? There's a hole bunch of tech (even JavaScript/CSS/HTML) 
> that "creates" the Web.
>
> All I'm saying is that HTTP is strictly part of it, and we both agree 
> with that.
>
> So is RDF for Linked Data.

Linked Data is something you can produce, is extremely useful form, 
using RDF.
Linked Data is also something you can produce, without any knowledge of RDF.

The statements above make the fundamental assumption that RDF is an 
enhancement of the long established entity relationship model. 
Basically, RDF didn't drop for the sky devoid of any genealogy.

Most important of all, RDF doesn't need to own Linked Data in order to 
justify its existence. There's a lot of virtue in being able to 
incorporate human- and machine-comprehensible entity relationship 
semantics into structured data representation and publication.

As I've stated repeatedly, conflating RDF and Linked Data is ultimately 
a disservice to both. These two things have distinct defining 
characteristics.


-- 

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