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

On 06/22/2013 04:47 AM, Nathan Rixham wrote:
> What it means now, or at any point in time, must be inclusive to new
> in-development or in-use things, other wise it will never mean anything
> else later down the line.
>
> If you want it to mean a very specific set of things at any one time,
> then take "Linked Data" down the standardization path and give it fixed
> versions which are RECs.
>
> I don't see anybody saying "don't use RDF" or "RDF is a bad idea for
> Linked Data, use Y instead". I just see some people inferring that RDF
> precludes all other things, and other people saying why should it
> preclude everything else?

RDF does not *preclude* anything else.  Documents can certainly be both 
RDF *and* something else, such as JSON or XML or HTML.  In fact, as long 
as a standard mapping to the RDF model is available, *any* document 
format can be interpreted as RDF.

The problem is that some people are claiming that RDF is not a 
*necessary* component of Linked Data.  If you take Linked Data to simply 
mean "data that is linked", then of course RDF is not necessary.  But if 
you take the term to refer to "the Semantic Web done right" or something 
similar that is intended to support the goals of the Semantic Web -- as 
most of us apparently do -- then RDF is currently *necessary*, because 
the Semantic Web relies on having a standard, universal data model to 
enable applications to automatically meaningfully combine independently 
created data.

Think about it.  How else could applications automatically meaningfully 
combine independently created data without having special out-of-band 
information about that data?  Without a standard universal data model 
each application would have to understand *all* of the data models used 
by any of the data sources that it may wish to access -- a massive data 
integration problem.

A key insight of the Semantic Web architecture is that by adopting a 
*standard* universal data model, this integration problem can be 
dramatically simplified for applications that wish to combine Web data: 
instead of having to understand N data models, the application only 
needs to understand *one*, regardless of the data sources.  To whatever 
extent you depart from using *one* universal data model, you are 
departing from that central vision of the Semantic Web, just as to 
whatever extent you depart from using *one* universal identification 
scheme -- URIs -- you are departing from the central vision of the 
regular Web and creating walled gardens -- just like in the days before 
the Web.

Can Semantic-Web-ish applications be built that understand the multiple 
universal data models used by those walled gardens (and internally unify 
them into their own private data models)?  Of course.  But that is the 
*opposite* of what the Semantic Web is all about.

The Semantic Web is not about building applications that are so smart 
that they can understand a plethora of different data models.  It is 
about enabling applications that are so *stupid* that they don't *have* 
to understand a plethora of different data models.

> ps: I'd be very wary about saying that any web tech is ".. *the*
> universal ..", many of them are Uniform, non are truly universal,even
> within their specific domains,

I beg to differ.  URIs (or their extension, IRIs) -- the standard 
universal identification scheme for the Web -- are so fundamental to the 
Web, that IMO if you do not have URIs you do not have the Web.  URIs are 
what make it *the* Web, as opposed to a collection of fiefdoms, which is 
what the Internet was before TimBL invented the Web.

> and any such claims will always be
> disagreed with by somebody as they are always untrue claims and
> alternatives always exist.

Really?  Can you show me an existing alternative to URIs, that you can 
write on the side of a bus and 20 strangers can read and, within 30 
seconds, all view the same document?   Similarly, can you show me an 
existing alternative to RDF, that allows applications to automatically 
meaningfully combine independently authored data, without out-of-band 
knowledge of that data?  Perhaps alternatives to URIs and RDF exist in 
theory, but I have not seen them in practice.

In short, if a document does not have a URI, it is not "on the Web". 
Similarly, if data is not standards-interpretable as RDF, it is not 
Semantic Web data, and hence it is not what most of us would call Linked 
Data.

David Booth

Received on Tuesday, 25 June 2013 04:19:55 UTC