On 13 Aug 2003, at 22:37, Andy Powell wrote:
> > > http://www.ukoln.ac.uk/metadata/dcmi/dcq-html/
> Fair point. (It was worth a try! ;-) ). OK, how about "... please
> respond by Fri, Sept 5th if you think this should not happen"? That
> gives people a little over 3 weeks to respond.
Since you've been busy on the abstract model discussion, I hope it's
not too late to comment on this.
To set the context, I am interested in processing XHTML documents to
extract and transform metadata to other formats. One of my concerns
is compatibility with previous forms of DC encoding and enabling
software to switch between the two where necessary. For a
heterogeneous set of XHTML documents, the Web, how can an application
determine which extraction algorithm to use?
Can we specify that the use of the XHTML head element profile
attribute is determinant for the new format? That is, a change from
"should" to "must" in section 2.7?
I welcome the reversion to the dot notation, but I noticed in section
3 there is still a mention:
"... should treat both '.' (full-stop) and ':' (colon) as valid
And the examples in this section also use colons. These should only
show upper case and lower case prefix examples.