[uf-discuss] hReview feedback

[Benjamin Carlyle]

On Wed, 2006-01-11 at 23:14 -0800, Ryan King wrote:
> The problem of "identifying authors in (X)HTML documents" is IMNSHO,  
> solved with <address> + hcard. Sure, the scope of <address> is fuzzy  
> and an hcard may just be the person's name. Yet, I think this is a  
> case where we need to make sure to reuse stuff we already have in  
> html and microformats, rather than reinventing.

This hasn't quite been adopted verbatim into hAtom at this stage. A
class="author" is currently required, and doesn't have to be placed on
an <address> element. Atom allows for multiple associations with between
the work and people connected to it, specificall "author" and
"contributor". I believe the current momentum for hAtom is to specify
"author" only, for the time being. We'll see whether the contributor use
case pops, up and how it does so.

I currently have two problems with the bare <author> notation. Perhaps
members of this list can help me resolve them:
1. Just as the atom specifiers seem to, I feel that there may be
multiple associations between a work and various people, each of whome
might deserve an <address> element as part of their credit. To take an
example from a little googling, it appears that "DigitalSpace
Corporation" is the author of http://www.ccon.org/ but it would
presumably be reasonable to provide an address for "Contact Consortium"
themselves on their website. In this case we would have "author" and
"about" relationships. It seems presumptious to me to restrict the scope
of legal <address> uses to "author", when html does not provide this
restriction. Perhaps the argument here is that author is not commonly
used, so can be somewhat hijacked in meaning as our use of the term will
probably be the most common in a few years.
2. My understanding is that <address> is an inline element, and it is
not legal to include certain constructs within it. I currently use a
<dl> for the hCard on my own web site, so can't put it inside and
address element. Is there a work-around for this in xhtml strict?

To frame this post specifcally in terms of hAtom, do you think that an
<author> element should be sufficient to define the atom entry author?
Speculatively, how would you specify other types of relationship such as
contributor? Opacity may also come into play when we ask, "What if the
entry author differs from the page author (say, for a planet)? How do we
parse that without confusing ourselves?"

-- 
Benjamin Carlyle <benjamincarlyle at optusnet.com.au>