Re: Content negotiation for Turtle files

On 2/6/13 9:52 AM, David Booth wrote:
> My view:
>
> On Wed, 2013-02-06 at 11:59 +0100, Bernard Vatant wrote:
> [ . . . ]
>
>> Do I have to pass the message to adopters : publish RDF in Turtle,
>> it's a very cool an simple syntax (oh but BTW don't forget to add HTML
>> documentation, and also RDF/XML, . . . .
> Please promote Turtle and actively discourage RDF/XML.  Turtle is now
> far enough along in its uptake and tooling to displace RDF/XML as a
> common denominator format for RDF, and here is real harm in doing
> anything that promotes RDF/XML, as: (a) RDF/XML is much harder for
> humans to read; and (b) RDF/XML misleads people into thinking that RDF
> is a form of XML.  I and others have many times seen people fall into
> the trap of thinking that they can use familiar XML approaches to RDF,
> and the result is painful disaster, because they have the wrong mental
> model of RDF.
>
> I think it is fine to quietly continue to serve RDF/XML if you have
> already been doing so, but please do not serve any new data in RDF/XML,
> and please do not use RDF/XML in published examples of RDF.
>
> BTW, the old "Semantic Web Layer Cake"
> http://www.w3.org/2001/09/06-ecdl/slide17-0.html
> is flat out wrong in showing XML at the base, as RDF/XML is merely one
> serialization of RDF, which is syntax independent.  Here is a much
> better version:
> http://www.w3.org/Consortium/Offices/Presentations/RDFTutorial/figures/TwoTowers.png
> It appears in this slide set from Ivan:
> http://www.w3.org/2005/Talks/1214-Trento-IH/#%28160%29
>

+10000



-- 

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