David Powell wrote:
> I think that there is a risk of interoperability problems if we don't
> state which attributes are allowed on the xhtml:div wrapper.
>
>
> As the xhtml:div wrapper is not part of the content: is it allowed to
> contain XHTML attributes such as class and style?
>
> I assume that if these attributes are allowed, then they should be
> ignored by renderers, and the attributed xhtml:div should be treated as
> an un-attributed one.

What about:
<atom:content type="xhtml">
  <div xmlns="http://www.w3.org/1999/xhtml";>
    <p style="position: absolute; top: 0; left: 0; width: 100%; height:
100%">
      Hey, hey! I can show ads that can take the whole viewport, hiding
everything else in the containing page! (if the feed content is
shown inside another HTML page)
    </p>
  </div>
</atom:content>

Moreover, you can still add constraints on the XHTML content but you'll
never be able to constraint HTML content...

Maybe that's why NITF (or its subset XMLNews-story) defines its own
formatting markup... though NITF also adds semantic markup (location,
person, etc.)

Anyway, the -06 spec says XHTML is used in its "basic" flavor (and that's
good! <applause/>), not allowing inline styles (but still the class
attribute) nor the script element.
Though you can still not put such constraints on type="html" content, and
this moves the interop' matter elsewhere: publishers are then required to
check and filter their XHTML content (or they'll just put it in
type="html" to bypass the validation process... worse than everything
else...) before including it in an Atom feed...

Oh, and as we're talking about it: the -06 spec states "If the value of
"type" is "xhtml", the content of the Text construct MUST be a single
XHTML div element" but the RNG schema says it can contain mixed content
with "anyElement".

However, I still can't understand the _need_ for this xhtml:div wrapper...
if someone likes to have a prefix-free feed, he can use a wrapper div, as
divs can contain divs!
So please revert to "mixed content without wrapper"! (I know it's too
late...)

-- 
Thomas Broyer


Reply via email to