Re: Practical issues arising from the "null relative URIs"-hack

Richard,

your statement is more politically-correct, but that's not the point.

After all these efforts by the community to argue that "RDF is a data
model and not syntax" and point this out as a newbie-misconception
(should I bring up RDF/XML here?), suddenly it is fine to have a
specification that does the opposite: base itself on RDF syntax, not
data model? Most of presentations about RDF that I've seen emphasize
absolute URIs, but now it looks like we were doing it all wrong and we
can come up with Relative-RDF at a snap of a finger.

Failure by the WG to see that it is the wrong level of abstraction
baffles me. Absolute URI-based serializations (like N-Triples) and
toolkit support (RDF-compliant, mind you) goes out the window (working
with RDF on the string level does not count). How in the world is this
going to help implementations of LDP and adoption of Linked Data in
general?


Martynas
graphityhq.com

On Thu, Apr 3, 2014 at 12:37 PM, Richard Cyganiak <richard@cyganiak.de> wrote:
> On 3 Apr 2014, at 11:03, Reto Gmür <reto@apache.org> wrote:
>> and LDP doesn't base on RDF.
>
> LDP is more than POST on containers, and Turtle is part of RDF. What you mean to say is:
>
> "The POST behaviour of LDP containers doesn't base on the RDF abstract syntax."
>
> There is no need to undermine your own argument by shrouding an *entirely valid technical point* in unnecessary divisiveness and FUD.
>
> Best,
> Richard

Received on Thursday, 3 April 2014 11:07:24 UTC