Giovanni Tummarello wrote:
RDFa will not generally negate the essential separation of Name (via
URI.URN-URL) and Address (via URI.URL) since Linked Data oriented triples
will still contain de-referencable URIs :-)
if you can put the RDF and the human legible HTML version in the same
address there is absolutely no reason to have separate resources.
If you really want to make it clear that "its not an informative
resource" (its not like up to today we had any evidence of this being
practically useful or enabling so far, matter of fact there are
evidences of the contrary [1]) then just say that in the RDF
<thisuri> <isnot> <aninformativeresource> :-)
gone with content negotiation, gone with multiple URI URN URL and
distinctions among them.
I hope we can agree on the principle of keeping things absolutely as
easy as possible, as the only way to win (back..) interest from the
actual web development circles and have adoption
Giovanni
[1] http://google-code-updates.blogspot.com/2008/02/urls-are-people-too.html
Giovanni,
I am absolutely game for clarity and simplicity.
So let's work on a document, or contribute to any that may be in
development, re. injecting more RDFa into the Linked Data conversation :-)
--
Regards,
Kingsley Idehen Weblog: http://www.openlinksw.com/blog/~kidehen
President & CEO
OpenLink Software Web: http://www.openlinksw.com
