W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > March 2007

Re: [RDFa] rdf:XMLLiteral (was RE: Missing issue on the list: identification of RDFa content)

From: Mark Birbeck <mark.birbeck@x-port.net>
Date: Sun, 18 Mar 2007 20:15:01 -0700
Message-ID: <640dd5060703182015g795baf7fn4ac1ad1df1761a2e@mail.gmail.com>
To: "Elias Torres" <elias@torrez.us>
Cc: "Dan Brickley" <danbri@danbri.org>, "Ian Davis" <iand@internetalchemy.org>, "Ben Adida" <ben@adida.net>, public-rdf-in-xhtml-tf@w3.org

Hi Elias,

> I just want to make sure we remember that this is not about losing an
> ability but choosing the default. Your benefit is ability to mark-up
> *mark-up* not text (this would just be a plain literal) w/o adding a
> datatype attribute. My benefit is having the ability to mark-up text w/o
> adding a datatype attribute. Which one happens more often? Isn't this
> how defaults should be chosen?

In RDFa at the moment, plain literals only come from the content
attribute. Everything else generates an rdf:XMLLiteral, even a simple
string of text:

  <span property="a:b">hi</span>

  <> a:b "hi"^^rdf:XMLLiteral .

So your suggestion that "your benefit" allows us to mark up text
without the need for a datatype attribute is already catered for in
the current specification, regardless of what datatype is defined as
the 'default'. And your comment that current RDFa behaviour ("my
benefit") only relates to marking up mark-up is not correct, since it
concerns simple strings of text, too.

In other words, the whole discussion about 'default' types has
significance only for some triple store that might be taking the
parsed values, and the current behaviour, as defined in the current
spec, has *no* impact on the mark-up. However, as I've said, making
plain literals the default does have consequence for the mark-up.

> BTW, I also have a problem with @content. It violates DRY and I'm not
> sure why it couldn't be done with a meta element somewhere else, if the
> content cannot be parsed deterministically and we are not maintaining a
> relationship to the original user input in the model.

First, @content is about providing an alternative value, that is
perhaps more accurate, for example:

  <span content="20070316">Last Friday</span> in Parliament ...

Second, the reason you can't always take this out to a meta element is
that there isn't always something you can latch on to, to make your
statement 'about'.



  Mark Birbeck, formsPlayer

  mark.birbeck@x-port.net | +44 (0) 20 7689 9232
  http://www.formsPlayer.com | http://internet-apps.blogspot.com

  standards. innovation.
Received on Monday, 19 March 2007 03:15:05 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:50:22 UTC