W3C home > Mailing lists > Public > public-xhtml2@w3.org > March 2008

Re: XHTML-Role last call working draft ready for review

From: Mark Birbeck <mark.birbeck@x-port.net>
Date: Wed, 19 Mar 2008 13:03:10 +0000
Message-ID: <a707f8300803190603u2e29d223q2c988695310d0c30@mail.gmail.com>
To: "Shane McCarron" <shane@aptest.com>
Cc: "XHTML WG" <public-xhtml2@w3.org>

Hi Shane,

>  I have uploaded a last call-ready editors draft to
>  http://www.w3.org/MarkUp/2008/ED-xhtml-role-20080318/
>  Please review this with an eye toward resolving to go to last call
>  tomorrow.

Great. Here are my comments:

Section 2.1

We say:

  A conforming XHTML Role Attribute Module document is a document that
  requires only the facilities described as mandatory in this specification and
  the facilities described as mandatory in its host language.

I think we can't phrase it like this; the problem is, how do we know
if the document conforms to the host language? Pedantic, I know. :)
But I think all we can say is:

  A document that conforms to the XHTML Role Attribute Module is a document
  that requires only the facilities described as mandatory in this
specification. Note
  that the host language may impose further mandatory requirements on the
  document, for it to be conformant with the host language.

Or something like that...

Section 2.1 Bullet 2

We need to say more than just 'use this namespace'; we also need to
say that @role must be prefixed.

Section 2.2

When we say that the content model must include @role, aren't there
still two ways it could be done? Couldn't a language use either
@xh:role or @role?

Section 3 1st Paragraph

We refer to "Role Attribute Module", but everywhere else we refer to
"XHTML Role Attribute Module".

We also say that:

  Any non-qualified value MUST be interpreted as being from the XHTML
  vocabulary at http://www.w3.org/1999/xhtml/vocab#, and MUST be taken
  from the list defined in this section.

However, that doesn't make it particularly tempting for people to
include @role in their new language. I think it would be better to do
what I did for @rel in RDFa, and say that if the value appears in the
following list, it MUST be interpreted as being in the XHTML vocab
namespace. After that, values are then deemed as being in the default
namespace as defined in CURIEs, and that _could_ be set to something
other than the vocab namespace, depending on the host.

Section 3 1st Paragraph after the example

The terms are referred to as defining "regions of a document", but I'm
not sure if that's the best way of looking at it, particularly in
relation to "definition".

banner...why not "header"? Banner is a loaded term, in relation to
advertising, etc.

complementary...no way! :) Also, the notions of "separable" and
"completely separable" are a bit confusing.

contentinfo...a bit vague, but not terrible. But what happened to "footer"?

definition...I'd like to see the use-case for this.

3.1 Extending the collection of roles

See points above about using the CURIE mechanism to allow for unprefixed values.

Appendix C

I'm not sure what this is trying to achieve. If we want to define the
@role vocabulary in terms of RDF Schema and OWL, then why would it
include WAI terms? And if we want to just illustrate how someone could
define a vocabulary, shouldn't that vocabulary include terms that are
*not* in our vocabulary (other people shouldn't have to define our
vocabulary for us).

Which makes me think that we should define terms like xh:region in our
own namespace, and then use them in place of wairole:region. In
particular, we should define xh:document as an XHTML document.



  Mark Birbeck

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

  x-port.net Ltd. is registered in England and Wales, number 03730711
  The registered office is at:

    2nd Floor
    Titchfield House
    69-85 Tabernacle Street
    EC2A 4RR
Received on Wednesday, 19 March 2008 13:03:45 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:30:29 UTC