RDF Links are maddening; anybody wanna talk me down?
let the rant commence...
it's cool that RDF has links everywhere; i like that.
"RDF is based on the idea of identifying things using Web identifiers (called Uniform Resource Identifiers, or URIs), and describing resources in terms of simple properties and property values."
but, frankly, the way links are expressed in RDF documents is maddening!
"Resources (as in Universal Resource Identifer) are precisely that identified by URIs. Web pages and email messages are thought of as resources. RDF unfortunately uses the term for anything which can be talked about - any concept no matter how abstract"
let me show you what i mean...
see, some URIs are dereferencable:
<http://mamund.com/foaf.rdf#me> <http://xmlns.com/foaf/0.1/weblog> <http://amundsen.com/blog/> .
and some are just identifiers not meant for de-referencing:
<http://mamund.com/foaf.rdf#me> <http://xmlns.com/foaf/0.1/weblog> <http://amundsen.com/blog/> .
but some identifiers can be de-referenced:
<http://mamund.com/foaf.rdf#me> <http://xmlns.com/foaf/0.1/weblog> <http://amundsen.com/blog/> .
and some dereferencable URIs contain fragment identifiers (usually meaning only a sub-section of the de-referenced document is the target):
_:x <http://www.w3.org/ns/auth/cert#identity> <http://mamund.com/foaf.rdf#me> .
and some dereferencable URIs should be treated as navigational links:
<http://mamund.com/foaf.rdf#me> <http://xmlns.com/foaf/0.1/homepage> <http://amundsen.com/> .
and some dereferencable URIs should be treated as imbed links:
<http://mamund.com/foaf.rdf#me> <http://xmlns.com/foaf/0.1/depiction> <http://amundsen.com/foaf/mamund.jpeg> .
finally, and here's the part that really gets me:
none of the RDF serialization formats has any way to indicate within the document any of these varying attributes of the links (dereferencable, identifer, navigational, imbed-able)
and that's just for READ-ONLY link semantics. don't even get me started on the lack of native write and templated read semantics in RDF.
no wonder that, more than a decade after it's initial description and six years after the first RDF standards, we still don't have a powerful, ubiquitous RDF Browser to match that of the HTML browser which, i will point out, was available less 24 months after the first description of HTML itself.
RDF may be great for read-only linked data handled by custom one-off applications but, unless some important changes are made here, i seriously doubt that RDF is going to be "the" Web format of the future.
why do i say this?
because the Web does not run on data. it runs on hypermedia and (to paraphrase) "RDF, you're no hypermedia type".
anybody wanna talk me down?