My favorites | Sign in
Project Home Downloads Wiki Issues Source
Search
for
XrdFiles  
about XRD
Updated Feb 4, 2010 by bradfitz
Comment by aredri...@gmail.com, Aug 15, 2009

Why XRD and not FOAF or another easily-extended, regular format?

Comment by tfi...@gmail.com, Aug 15, 2009

Using RDF and the FOAF vocabulary makes a lot of sense.

Comment by wnorris, Aug 16, 2009

The latest draft of XRD can be found in the OASIS svn repository here: http://tools.oasis-open.org/version-control/svn/xri/xrd/1.0/drafts/

Comment by tee...@gmail.com, Aug 17, 2009

+1,000,000 for FOAF.

Comment by project member martmart...@gmail.com, Aug 30, 2009

-1 for any kind of RDF. This specification doesn't need to get bogged down in that can of worms.

Comment by drewpca, Oct 4, 2009

FOAF is already in use for pretty much this kind of thing. http://foaf.qdos.com/ is one example of an email->user data search, and it even supports FOAF's md5(email) technique. Try "henry.story@sun.com" for an nice sample of what can be done. Or submit your own http://{user}.livejournal.com/data/foaf data if you have an LJ account, and then browse users from there.

I see no disadvantage of RDF over XRD. To the webfinger users, they're both formats that can be represented in XML that the users don't want to learn or type. But there are already lots of active FOAF generators, and the extensibility of RDF is better.

Comment by henry.st...@gmail.com, Nov 29, 2009

the reasons people use to think RDF was a can of worms were:

  • they confused syntax with semantics
  • the tools for RDF were not widely available, making it very difficult to use the syntactic tools (DOM, XPath, XSLT, ...) to work with the standard serialisation known as RDF/XML.

XML has gone donw a bit in favor now, though I think that is mistaken too, in favor of JSON and other just as mistaken moves. But nevermind.

What you should define is the Semantics of what you need. Then if you want a simple XML document as above, namespace is, write it out as you have above, and provide a GRDDL for your document to your ontology.

Comment by henry.st...@gmail.com, Nov 29, 2009

btw, there is an interesting discussion on the mailing list and an alternative proposal called fingerpoint that is more RDFish.

There is an interesting counter proposal called fingerpoint. See the discussion here: http://groups.google.com/group/webfinger/browse_thread/thread/e16791bbdf3c4606/1156e796e9b4ab46?lnk=gst&q=toby+inkster#1156e796e9b4ab46

or ... See More http://buzzword.org.uk/2009/fingerpoint/spec

Comment by Ray.Rac...@gmail.com, Feb 14, 2010

As someone who only pops in every now and then to see how all its all unfolding... am invariably surprised by the endless hamster wheel spinning.

RDF just has this nailed. Its a shame RDF's XML serialization is so ... horrible. But the Turtle serialization syntax for example is very clean, easy to read, easy to generate, easy to parse and compact.

As a bonus, the whole impedance mismatch of serialized data <-> XML Graph <-> Language representation is avoided. In other words, Turtle syntax is really easy to parse and it maps readily into any language with dictionaries and lists/arrays.

A nice exercise is to map XRD are one's current webfinger into RDF (turtle syntax) with FOAF.


Sign in to add a comment
Powered by Google Project Hosting