W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > July 2010

Re: ISSUE-24: Proposal for dealing with case-insensitive terms in the XHTML vocabulary

From: Shane McCarron <shane@aptest.com>
Date: Wed, 14 Jul 2010 10:35:05 -0500
Message-ID: <4C3DD929.7040304@aptest.com>
To: Felix Sasaki <felix.sasaki@fh-potsdam.de>
CC: Toby Inkster <tai@g5n.co.uk>, RDFa WG <public-rdfa-wg@w3.org>
That's not really what I meant.  More that the polyglot document is 
incomplete w.r.t. attribute values.  It oversimplifies the case.  I am 
sure that it will get sorted out in due course.  However, with regard to 
TERMs from a specific vocabulary - that is, values of @rel, @rev, and 
some other attributes and that match the production NCName, we are 
committed to ensuring that the requirements of RDFa and HTML5 are 
consistent with one another.

At least, that's my belief right now.  The Working Group could come up 
with some other approach as well.

On 7/14/2010 10:07 AM, Felix Sasaki wrote:
>
>
> 2010/7/14 Shane McCarron <shane@aptest.com <mailto:shane@aptest.com>>
>
>     I believe this definition is consistent with that one for values
>     without colons (non-CURIEs, or TERMs in RDFa Core 1.1).  For
>     values with colons, we disagree that values of rel are always
>     lower case since mixed case is essential for supporting our use
>     cases, as well as supporting other uses.
>
> So polyglot markup is not meant to be relevant for the CURIE use case? 
> Just trying to understand.
> Felix
>
>
>
>
>     On 7/14/2010 9:50 AM, Felix Sasaki wrote:
>>     Sorry for lurcking, a clarification question ...
>>
>>     2010/7/14 Toby Inkster <tai@g5n.co.uk <mailto:tai@g5n.co.uk>>
>>
>>         On Tue, 13 Jul 2010 17:01:52 -0500
>>         Shane McCarron <shane@aptest.com <mailto:shane@aptest.com>>
>>         wrote:
>>
>>         > In XHTML+RDFa 1.1 we should say:
>>         >
>>         >     When referencing TERMs in the vocabulary at
>>         > http://www.w3.org/1999/xhtml/vocab, TERMs must be mapped to
>>         lower
>>         > case.
>>
>>         Does this also cover the case of the empty prefix? According
>>         to my
>>         reading of XHTML+RDFa 1.0, given the following:
>>
>>                rel="Next"
>>                rel=":Next"
>>
>>         The first is mapped to lower-case; the second is treated
>>         case-sensitively. This is because the former is a term and
>>         the latter
>>         is a CURIE.
>>
>>     At
>>     http://www.w3.org/TR/2010/WD-html-polyglot-20100624/#attribute-values ,
>>     values of "rel" are described as always lower case. How does that
>>     (does it at all?) relate to what you describe above?
>>     Thanks for the clarification in advance,
>>     Felix
>>
>>
>>         I suggest we retain this distinction, as there do exist
>>         mixed-case
>>         terms defined in the XHTML vocabulary (though they're not
>>         used by RDFa
>>         currently).
>>
>>         Ivan wrote:
>>
>>         > 2. in RDFa+XHTML (and I presume in RDFa+HTML5) there is a a
>>         default
>>         > @vocab, namely http://www.w3.org/1999/xhtml/vocab# and, if
>>         this is the
>>         > one in effect, then terms are interpreted in a case
>>         insensitive way,
>>         > ie, terms must be mapped on lower case.
>>
>>         Personally I'd ditch that too -- you get too many junk
>>         results. Better
>>         to say that the default vocabulary in (X)HTML+RDFa is null.
>>
>>         Otherwise you'll end up with having to deal with rel values from
>>         Microformats, etc. The problem with that it that Microformats
>>         have
>>         their own ways of determining what the "subject" is for a
>>         rel; applying
>>         RDF's @about/@src/chaining rules doesn't work well.
>>
>>         To handle the case of rel="next", rel="prev", etc define a
>>         default
>>         *profile* for XHTML+RDFa, and have that profile define those
>>         terms
>>         case-insensitively. (The profile SHOULD be hard-coded in
>>         XHTML+RDFa
>>         parsers; authors MAY include an explicit @profile attribute
>>         referencing
>>         the default profile.)
>>
>>         This means that parsers need to internally allow profiles to
>>         define
>>         both case-sensitive and case-insensitive terms, though we may
>>         decide
>>         not that profiles other than the XHTML vocab are not
>>         privileged to
>>         declare case-insensitive terms.
>>
>>         This solution allows us to pick up "our" @rel terms but skim
>>         safely
>>         over the @rel terms used by Microformats, Javascript
>>         libraries, etc.
>>
>>         --
>>         Toby A Inkster
>>         <mailto:mail@tobyinkster.co.uk <mailto:mail@tobyinkster.co.uk>>
>>         <http://tobyinkster.co.uk <http://tobyinkster.co.uk/>>
>>
>>
>
>     -- 
>     Shane P. McCarron                          Phone: +1 763 786-8160 x120
>     Managing Director                            Fax: +1 763 786-8180
>     ApTest Minnesota                            Inet:shane@aptest.com  <mailto:shane@aptest.com>
>
>          
>
>

-- 
Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: shane@aptest.com
Received on Wednesday, 14 July 2010 15:35:51 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 04:55:07 GMT