On 17 Jan 2005, Walter Underwood wrote:
Call it "AtomAsRDF".
That is indeed a much better name. Ok, off to change the page again...
On 17 Jan 2005, at 20:38, Sam Ruby wrote:
Henry Story wrote:Yes, very good point. I have created a new Page called AtomRDF [3], whose aim
is to track what would be needed for an Atom document to be an RDF document.
I have a number of problems with the AtomRDF document as it stands now.
For starters, the current atom:generator element has optional "uri" and "version" elements, not "atom:uri" and "atom:version" elements. Unfortunately, these are quite different.
Oh God! I just looked this up in "Xml in a Nutshell". I had not realized that
default namespaces did not apply to attributes. How awful! Is there any good reason
for this? Any clever way to fix it? Otherwise why is XML broken like this?
If there is no simple fix, I'll have to point this one out on the AtomRDF page
as something to find a solution to.
More siginificantly, the document does not address the largest issue, which is content.
Can you expand a little on this. I have put this wiki up in order to serve as a
reference point for all these problems. This is one I had not yet noticed, or not
under your description at least.
I do believe that it would be possible, and perhaps quite useful, to define an XSLT tranform from Atom/XML to RDF/XML.
Yes. The problem is that one can xslt transform something to many many different
types of rdf/xml documents.
Also the purpose of narrowing the gap between Atom and RDF down to 0 is that it would help give a very clear foundation for explaining extensibility, when combined with the AtomOWL ontology. The power of this is very similar to the power one gains
with OO programming in creating frameworks. The relationship between OO programming
and ontologies should be obvious from definition of ontology itself.
"An explicit formal specification of how to represent the objects, concepts and other entities that are assumed to exist in some area of interest and the relationships that hold among them." [1]
So yes. If we can narrow the gap, we can do Object Oriented XML development: which
is way cool. (No fear, OO xml looks like plain old xml btw)
One solution may be to have what we could call a canonical Atom format that
would present an Atom document in a clean RDF way. This format would perhaps require
that all attributes be in the atom namespace.
Henry Story
[1] http://dictionary.reference.com/search?q=ontology from the online dictionary of computing.
