On 22 Nov 2004, at 11:31, Chris Croome wrote:
> On Sun 21-Nov-2004 at 11:34:55PM +0000, Andy Powell wrote:
> > I think that it is important that this group keeps up to date with
> > latest RDF/XHTML proposals being developed by the W3C. The latest
> > RDF/A notation is available at:
> >
> > http://www.formsplayer.com/notes/rdf-a.html
Eventually I have found some time to take a look at this...
> > - This notation is different to the current DC/XHTML recommendation
> > [1]
> > and it doesn't seem sensible long term to have two syntaxes that
> > achieve the same thing. Therefore, we will need to respond to
> > this work at some point in the future.
It is important to note that this draft is different from the current
W3C recommendation for XHTML itself. They are looking at XML
languages in general and take a fairly thorough look at the
possibilities. The examples imply changes to the XHTML recommendation
to allow link and meta elements in the body, add an "about" and
"property" attribute to meta and link elements for instance.
The DC focus is on the current document and its relations to other
resources, whereas this draft also has a scheme for making statements
about the secondary resources too. For instance, a citation refers to
another document and one can also make statements about the creator
of the cited resource, etc. Their aim is to introduce the full scope
of RDF into any XML document type.
> Yep, but the differences are small -- they have used : rather than .
> (which is something I remember was considered here before...) and also
> a lower case dc, and there are examples of using rel and rev
> attributes on <a> elements in addition to <link> elements -- these are
> the essential differences I think?
I think it is more problematic for DC than that. In many of the
examples given in the draft, the authors use the DC vocabulary
without once discussing the current DC in XHTML scheme [1]. I think
there are considerable strengths in the DC scheme and this group
ought to make sure its key aspects are included in the RDF/A plan
before it progresses very much farther.
In particular, the use of existing features in XHTML seem to have
been missed; the name and scheme attributes of meta are never
mentioned. The use of the head profile attribute and metadata scheme
prefix are not mentioned. The RDF/A plans may not have developed to
this extent yet, but the DC group should make sure its own scheme is
not left aside.
> Personally I think that using dc:title in XHTML rather than DC.title
> probably does make more sense...
As I recall, the DC group were asked not to use a colon separator in
anticipation of the RDF/A draft. Looking at it now, I don't see why.
If anything, DC should have used the colon to avoid subsequent
changes to its DC in XHTML recommendation. Personally, I think the
form of the separator is irrelevant but the dot has established usage
and should be retained. The coincidence with the XML namespace
separator in a non-XML context has potential to confuse newcomers. In
all likelihood W3C will win the day though.
> > - It probably isn't sensible for us to make modifications to the
> > current
> > DCMI recommendation [1] until the RDF/A syntax becomes a W3C
> > recommendation.
> >
> > - However, we may want to consider whether DCMI (i.e. this working
> > group)
> > should have some formal (or at least more structured) method of
> > input into RDF/A syntax specification as it develops.
I would urge input now to make sure the work that has gone into the
DC in XHTML scheme is not overlooked by the RDF/A group. There are
many publishers making extensive use of the DC in XHTML scheme and I
think it is this group's responsibility to conserve their efforts too.
[1] Expressing Dublin Core in HTML/XHTML meta and link elements
http://dublincore.org/documents/dcq-html/
--
<URL:http://www.codestyle.org/>
|