W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2002

Issue faq-html-compliance

From: Dave Beckett <dave.beckett@bristol.ac.uk>
Date: Wed, 12 Jun 2002 22:00:57 +0100
To: w3c-rdfcore-wg@w3.org
Message-ID: <10970.1023915657@tatooine.ilrt.bris.ac.uk>
This email closes ACTION 2002-06-07#5

Sean B. Palmer posted the following feedback on the www-rdf-comments
list, announced in
  Input on faq-html-compliance
  http://lists.w3.org/Archives/Public/www-rdf-comments/2002AprJun/0170.html
and a copy archived at:
  RDF in HTML: Approaches
  http://lists.w3.org/Archives/Public/www-archive/2002Jun/att-0002/01-index

In order to create a way to embed RDF in XHTML, the way that the
MathML folks did it was to create an all new modular MathML+XHTML DTD
but I don't see this as something we should do, since RDF isn't so
directed at user markup.

Sean came to a couple of tentative conclusions on ways to link and
embed; the former using the (x)html <link> element, and the latter
embedding an encoded RDF/XML blob in the <script> element with
a mime type.

The latter seems rather awful and I don't propose we use it, which
unfortunately rules out an embedded RDF/XML in XHTML, with
validation.  Of course, without validation, it works OK as long as
the attribute-form of RDF/XML is used.


Proposal:

We recommend using <link>[1][3] in the <head> of the (x)html which is an
approach that Dublin Core has been using for several years on its web
site and works ok.  In particular we should say to use the <link> type
attribute with the value of "application/rdf+xml".

With respect to link types[2][4], here the value of the rel attribute.
That may be application dependent.  Dublin Core has used and
recommended rel="meta", which is allowed since [2] notes:

  "Authors may wish to define additional link types not described in
  this specification. If they do so, they should use a profile to
  cite the conventions used to define the link types. Please see the
  profile attribute of the HEAD element for more details."

but that is optional since it is a _should_ not a _must_.
I don't think <head profile=".."> would be expected by most
applications.

However [4] doesn't mention other allowed types!

I attach an example of using this.

---

I'm open to suggestions to how much emphasis to use here on
suggesting / requiring / recommending an approach here.

Dave


[1] <link> in HTML 4.01
http://www.w3.org/TR/html401/struct/links.html#edef-LINK

[2] HTML 4.01 link types
http://www.w3.org/TR/html401/types.html#type-links

[3] XHTML Modularization - 5.19 Link Module
http://www.w3.org/TR/xhtml-modularization/abstract_modules.html#s_linkmodule

[4] XHTML Modularization - LinkTypes
http://www.w3.org/TR/xhtml-modularization/abstraction.html#dt_LinkTypes



Received on Wednesday, 12 June 2002 17:00:58 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:49:14 EDT