W3C home > Mailing lists > Public > www-svg@w3.org > September 2008

Re: [1.2T-LC] datatype (5.10.1) (ISSUE-2066)

From: Dr. Olaf Hoffmann <Dr.O.Hoffmann@gmx.de>
Date: Thu, 25 Sep 2008 14:45:56 +0200
To: www-svg@w3.org
Message-Id: <200809251445.57055.Dr.O.Hoffmann@gmx.de>

Doug Schepers:
> Hi, Dr. Olaf-
>
> Dr. Olaf Hoffmann wrote (on 9/23/08 9:44 AM):
> > Doug Schepers:
> >> As with 'typeof' and the rest of these metadata attributes, 'datatype'
> >> is not intended
> >> to affect rendering or execution of the element. It is left
> >> intentionally loose to allow other formats (RDFa, Microformats) to
> >> define what the relationship is, but I agree this could use some
> >> clarification about use.  We will discuss how we can tighten this up
> >> with that other group, to be as clear as possible without being too
> >> restrictive.
> >
> > Maybe some (informative) references may help authors too to
> > find a way to use it.
> > Because if it not understandable, how this is defined and authors
> > have to use a search engine to get some results, what the purpose
> > of those attributes might be, if there is not more than a 'secret'
> > existence in the document or the DOM, there is a pretty good chance, that
> > almost noone understands the purpose and usage ;o)
> > Obviously this might not be an easy task to reference something
> > reliable, if some formats have no normative recommendation, but then
> > it might alreay help to have one or too examples of such constructions,
> > already having any effect on any viewer.
>
> Excellent suggestion.
>
> I've incorporated some of this thread in the current draft. [1]  Please
> take a look and let me know if this clarifies matters.  Any further
> suggestions that might help would also be most welcome.

As far as I understand the samples in RDFa, the 'datatype' seems
to be related to the content of the attribute 'content' or to the content
of the element itself, for example:

'<span property="cal:dtstart" content="20070916T1600-0500"
            datatype="xsd:datetime">
        September 16th at 4pm
</span>'

(o;
looking into xsd:datetime this looks like a wrong interpretation
of xsd:datetime
http://www.w3.org/TR/xmlschema-2/#dateTime

The content of the element does not fit to xsd:datetime and
in the value of the content attribute there are missing some
signs: 2007-09-16T16:00-05:00
;o)

or
'<span property="dbp:dateOfBirth" datatype="xsd:date">1879-03-14</span>'

or
'<h2 property="dc:title" datatype="rdf:XMLLiteral">
  E = mc<sup>2</sup>: The Most Urgent Problem of Our Time
</h2>'

I did not read the complete RDFa to find the paragraph, how to
avoid this plurivalence, but obviously it is not just an(y)other
attribute, maybe only 'content' or the content of the element itself.
Whereever the content comes from, datatype seems to be related
to the content, not directly to the value of an(y)other attribute.
The current SVG definition at least seems to exclude the second 
possibility. And if it is true, that it is always related to the content,
then maybe one can write:
'The 'datatype' attribute specifies a datatype for the content of the
element, 'datatype' is the attribute for, respectively for the plain 
text attribute value of the 'ontent' attribute if provided.'


'typeof' is just an association, therefore there is no need to avoid
plurivalences.

>
> > If in this case the other format is intended to specify the '(an)other
> > attribute', isn't it necessary, that authors have to provide a namespace
> > or something to indicate, which format specifies this, if there is more
> > than one? Else the result will always be somehow plurivalent and
> > arbitrary and therefore typically not understandable for anyone else but
> > the author and therefore most authors might think, it is not worth the
> > trouble to use it at all...
>
> With the informative reference, do you still think there's a problem?
>

I think, I understand this RDFa and their samples, with this reference
this looks much better and is much more convenient to find some
information, how to use it ;o)


> Please see also the wording on "metadata attributes" for more details. [2]
>
>
> [1] http://dev.w3.org/SVG/profiles/1.2T/publish/struct.html#Core.attrib
> [2]
> http://dev.w3.org/SVG/profiles/1.2T/publish/metadata.html#MetadataAttribute
>s
>
> Regards-
> -Doug Schepers
> W3C Team Contact, SVG and WebApps WGs


Olaf
Received on Thursday, 25 September 2008 12:58:35 GMT

This archive was generated by hypermail 2.3.1 : Friday, 8 March 2013 15:54:40 GMT