Andy, and All,
Comments on the proposed 'Expressing Qualified Dublin Core in
HTML/XHTML meta elements.
Best wishes,
Ann
----------------------------------------------------------------------
1. Terminology: resource. I think including 'human beings' as an
example of a resource may be rather contentious, and so shouldn't
really be in this document. There is no DCMI Type for human
beings - in fact they were explicitly excluded. The same may apply
to corporations. I don't wish to start an argument here on this
subject - I just think this document is the wrong place to raise the
issue.
2. Is there a good reason for changing the case of the first letter of
elements, eg DC.Date to DC.date? I realise you have dealt with
this in section 3, but I'm not sure if that is sufficient. I would guess
there is a lot of DC within HTML meta tags out there. The existing
recommendations have been in place for a long time and are linked
from the DC website via the Usage Guide. 'Encoding Dublin Core
Metadata in HTML' at http://www.ietf.org/rfc/rfc2731.txt is dated
1999. Are you proposing that this recommendation also be
changed?
There is also a knock-on effect to local recommendations - I
suspect there are a lot of these. For example, a few months ago I
wrote local recommendations for adding DC metadata to our web
pages as part of our 'make web pages accessible' exercise
(because of the UK disability legislation). I wrote these following
recommendations on the DC website, eg to use DC.Date. If these
proposed recommendations are accepted, do I then have to change
my local recomemndations? Then we would have inconsistent 'new'
and 'old' pages.
I think you need to be aware that those who are encoding DC in
HTML are generally not conversant with XML architectures, etc,
and it will be dificult to explain why the existing guidelines, which
have been in place for a long time, have been changed. It will not do
anything for the image of DC 'out there'!
I think it would be better to say something less strong like: either
DC.Date or DC.date is acceptable, but the latter is preferred.
3. In 2.1, last paragraph about CDATA. maybe it would be a good
idea to point out explicitly that &, <, > should encoded as
character entities, ie. &, etc, if included in the 'content'
string. Again this document will be read by people who don't
understand what CDATA means.
4. 2.6 Namespaces. As this is a guidelines document, maybe a
title string should be included in the namespace declarations:
<link rel="schema.DC" href="..." title="Dublin Core Metadata
Element Set, Version 1.1" />
<link rel="schema.DCTERMS" href="..." title="DCMI Elements and
DCMI Qualifiers" />
5. Namespaces. If I use the DCMI Type Vocabulary, do I also need
to
declare a schema for that? Eg:
<link rel="schema.DCMIType" href="..."
title="..." />
<meta name="DC.type" scheme="DCMIType" content="Text" />
6. Example. This needs to declare the DCTERMS schema
because it uses 'issued' and 'abstract'.
--------------------------------------------------------------------------
Mrs. Ann Apps. Senior Analyst - Research & Development, MIMAS,
University of Manchester, Oxford Road, Manchester, M13 9PL, UK
Tel: +44 (0) 161 275 6039 Fax: +44 (0) 0161 275 6040
Email: [log in to unmask] WWW: http://epub.mimas.ac.uk/ann.html
--------------------------------------------------------------------------
|