Comments on HTML Microdata, W3C Working Draft 24 June 2010

Comments on HTML Microdata, W3C Working Draft 24 June 2010
-- http://www.w3.org/TR/2010/WD-microdata-20100624/

I am happy to see DCMI Terms included in the HTML Microdata
specification and have a few constructive comments on points
of detail.

1) Use of dcterms:source

   In section 5.2 (normative) [1], dcterms:source is being used
   where dcterms:references might be more appropriate.

   The "cite" element is described as follows [2]:

        The cite element represents the title of a work...  that is
        being quoted or referenced in detail (i.e. a citation),
        or it can just be a work that is mentioned in passing.

   Point 4 of the RDF conversion algorithm [1], then, appears
   to map cites to triples using dcterms:source with,
   in the object position, an "absolute URL that results
   from resolving the value of the element's cite attribute
   relative to the element".

   If I am grasping this mechanism correctly, then dcterms:source
   would seem to be misplaced here, as dcterms:source is defined as follows:

        A related resource from which the described resource is derived.

   In other words, a triple that says:

        <the document's current address> <dcterms:source> <URL of cite attribute>

   would in effect mean:

        This document has been derived from the resource identified by 
        the URL of the cite attribute.

   which I do not think is intended.  The property dcterms:references, 
   on the other hand, is defined as [3]:

        A related resource that is referenced, cited, or
        otherwise pointed to by the described resource.

   and is "intended to be used with non-literal values"
   (i.e., URIs or blank nodes as objects), so the triple:

        <the document's current address> <dcterms:references> <URL of cite attribute>

   would in effect mean:

        This document cites the resource identified by the
        URL of the cite attribute.

   I'd need to see some more examples, with triples, to be
   sure, but it appears to me that dcterms:references may
   more closely capture the intended semantics of "cites".

2) Use of dcterms:creator

   In section 5.2.1 (non-normative), the first example expresses the author
   of the book as

        <span itemprop="http://purl.org/dc/terms/creator">Wil Wheaton</span>

   As with dcterms:references, however, dcterms:creator is intended to be
   used with non-literal values (i.e., URIs or blank nodes).

   On the other hand, dc:creator, which has no formal range (having been 
   coined before the notion of formal domains and ranges in RDF was finalized),
   would be correct, i.e.:

        <span itemprop="http://purl.org/dc/elements/1.1/creator">Wil Wheaton</span>

   The difference between the /elements/1.1/ and /terms/ namespaces is
   discussed more fully in [4].

3) Mapped prefixes for Dublin Core namespaces

   In the Examples section [5], the prefix "dc:" is mapped
   to http://purl.org/dc/terms/:

        @prefix dc: <http://purl.org/dc/terms/> .

   Of course, mappings are local and implementers are free to
   choose arbitrary alphanumeric sequences for use as prefixes.
   In a Semantic Web context, however, we more commonly find 
   prefixes mapped as follows:

        @prefix dc:                <http://purl.org/dc/elements/1.1/> .
        @prefix dcterms: (or dct:) <http://purl.org/dc/terms/> .

Best regards,
Tom Baker

[1] http://www.w3.org/TR/2010/WD-microdata-20100624/#rdf
[2] http://www.w3.org/TR/2010/WD-html5-20100624/text-level-semantics.html#the-cite-element
[3] http://dublincore.org/documents/dcmi-terms/#terms-references
[4] http://lists.w3.org/Archives/Public/public-esw-thes/2009Jun/0017.html
[5] http://www.w3.org/TR/2010/WD-microdata-20100624/#examples
[6] http://www.w3.org/TR/2010/WD-html5-20100624/rendering.html#represents

-- 
Thomas Baker <tbaker@tbaker.de>
CIO, Dublin Core Metadata Initiative

Received on Friday, 27 August 2010 19:06:53 UTC