[foaf-protocols] The dereference issue
kidehen at openlinksw.com
Thu Dec 16 23:38:00 CET 2010
On 12/16/10 4:46 PM, Nathan wrote:
> Hi All,
> Thought I'd take a few minutes to note something which may need covered
> in the WebID IG, and probably in collaboration with the RDF WG.
> RDF URI-References are simple names, logical constants, there is no
> dereferencable constraint at all, as in any URI of any scheme, is
> perfectly valid, furthermore, none of the RDF serializations add any
> hypermedia semantics to the URIs we use in RDF (bar RDFa in some places
> like @href).
> The dereference part is implied in part by URIs, and suggested by the
> Linked Data design issue, but to the best of my knowledge there's no
> specification or rec that places any constraint of deferenceability on
> the URIs used in RDF, and as far as I can tell WebID Protocol would need
> this constraint, to the point that it may even need it's own registered
> mediatype with hypermedia/link semantics and constrains on the types of
> URIs which can be used (?)
> foaf-protocols mailing list
> foaf-protocols at lists.foaf-project.org
HTTP URIs + EAV doesn't need any application specific controls re.
hypermedia. The Identifiers are Hypertext Transfer Protocol based, you
de-reference them en route to accessing structured data. Rather than
having a myriad of structs you have one EAV/SPO. A real Linked Data
aware application will process the self-describing data and deductively
deliver application behavior to its users. Of course, a little scripting
here and there for specialist patterns e.g. GUI.
H factors and RDF are just other examples of stuff that just confuses
people. One minute they grok @href (Hyper.... Reference) and then later
on they are told it's about high, low, or none existent H factors :-)
1. http://www.amundsen.com/hypermedia/hfactor/ -- well intended but
confusing matters more often than not.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the foaf-protocols