RE: The tone of the "JSON-LD vs. RDF" debate (was re: Sub-issue on the re-definition of Linked Data)

On Wednesday, June 12, 2013 12:19 AM, Pat Hayes wrote:
> On Jun 11, 2013, at 5:02 AM, Markus Lanthaler wrote:
>> We were asked by members of the RDF WG to add this section. In fact,
Richard
>> wrote most of it, I just completed it because he got busy otherwise.
There
>> why reason it is there, is because the data models are not exactly the
same.
>> JSON-LD allows bnodes for graph labels and predicates wheras RDF's data
>> model does not.
> 
> I know there are these slight differences, and I do not mean to imply
> that they are not important. But still, the JSON-LD model is clearly a
> slight generalization of RDF. It is not something completely different.
> It could be explained, for example, by saying that its the RDF model
> with two generalizations. (The bnodes-for-predicates generalization is
> already widely known in the RDF literature, invented by terHorst in
> 2004, and could be cited.) But to re-state all the rest of the model as
> though it was something wholly new, without any reference to RDF at
> all, is (to repeat myself) deliberately misleading.

So, you are raising a new point then. You want the data model section to be
replaced with a statement saying that JSON-LD serializes the "RDF['s data]
model with two generalizations" citing terHorst for the
"bnodes-for-predicates generalization".

Is that correct? Could you please propose a concrete wording?


> And I have to say,
> that your reply here seems to me to be an almost pure example of
> tendentiousness.

I just re-read the my reply above three times and all I could find was an
(in my opinion at least) completely objective explanation of why that
section is there. I don't think it makes sense to continue these discussions
if every second sentence gets personal and offensive without any reason. So
either that stops or I stop responding to such emails.



--
Markus Lanthaler
@markuslanthaler

Received on Wednesday, 12 June 2013 00:03:23 UTC