Eric Scheid wrote:
> The article goes into content as text/xhtml, and the RDF goes
> into the <entry> as extensions.
        Every Atom entry is supposed to have an alternate HTML
representation or be an alternate representation of some HTML. Given that,
then where would one find the RDF in the HTML representation of the entry?
        I believe it doesn't make sense for us to add data-carrying elements
to Atom other than atom:content or atom:summary. Atom provides a definition
of a collection of entries and it provides the entry format. Frankly, it
should stop there. The data payload should be carried in the content
element.
        This is the approach we take with "Stuctured Blogging"[1]. The idea
is that you insert the machine parseable data into the content element and
you also insert it into the body of the HTML that is an alternate of the
atom:entry. Thus, the machine readable data can be found either by reading
the feed or by reading the HTML pages that are alternates of the entries in
the feed. One of the very nice things about this approach is that it enables
data to be carried in HTML, XHTML, RSS, and Atom with complete consistency
and without having to define extensions to any of the formats. Additionally,
because Structured Blogging relies on normal XML, it means that the same
data that you embed in HTML, XHTML, RSS and Atom can also be processed
without modification by normal and widely available XML parsers, tools, etc.
By providing GRDDL transformations for Structured Blogging entries, we also
gain the benefit of being able to directly address the needs of the
RDF/OWL-semantic processing crowd. All this can be done without defining new
extension elements, without having to coordinate between extensions in RSS
and Atom and without breaking any rules of HTML and/or XHTML. 
        Data should be in the content element...

                Bob wyman

[1] http://structuredblogging.org



Reply via email to