Re: What Does Point Number 3 of TimBL's Linked Data Mean?

On 06/25/2013 01:30 PM, Ted Thibodeau Jr wrote:
> On Jun 25, 2013, at 12:19 AM, David Booth wrote:
>> In fact, as long as a standard mapping to the RDF model is
>> available, *any* document format can be interpreted as RDF.
> Ahah!
> So, if "a standard mapping to the RDF model is available" for my
> chosen data publication format -- which I may choose just because it
> looks pretty to me -- it doesn't matter that *I* don't know anything
> about RDF, that mapping, SPARQL, etc.

Yes!  :)

> In other words, I can publish Linked Data without RDF [ . . . ]

No, that does not follow.  :(   But you *can* publish Linked data 
without *knowing* RDF and even without knowing that you *are* publishing 
RDF -- for example if you publish JSON-LD.

> Your *interpretation* and *exploitation* of that Linked Data may
> require RDF and/or SPARQL and/or a lot more alphabet soup -- but the
> simple question of whether what I have published counts as Linked
> Data or not requires much, much less.

No, it requires the RDF to be a part of the information content of the 
document, represented in some standards-based RDF-interpretable 
serialization.  It isn't a question of what tools or technologies the 
client application uses.  Whether the serialization is Turtle, JSON-LD 
or anything else, the RDF information content needs to be there.


Received on Tuesday, 25 June 2013 18:14:21 UTC