W3C home > Mailing lists > Public > www-html@w3.org > May 2007

@role (was RE: Cleaning House)

From: John Foliot - WATS.ca <foliot@wats.ca>
Date: Mon, 7 May 2007 07:14:23 -0700
To: "'Maciej Stachowiak'" <mjs@apple.com>, "'Tina Holmboe'" <tina@greytower.co.uk>
Cc: "'Murray Maloney'" <murray@muzmo.com>, <www-html@w3.org>, <public-html@w3.org>
Message-ID: <003d01c790b2$08017480$6501a8c0@bosshog>

Maciej Stachowiak wrote:
> What is the correct way, according to you, to mark up a ship name?
> Instead of <i class="ship">HMS Pinafore</i>. I'm assuming <span
> class="ship">HMS Pinafore</span> is no better, since <span> is
> semantically null. <em>HMS Pinafore</em> is clearly wrong, since
> setting ship names in italics has nothing to do with emphasis. Using
> no markup at all is wrong, since then there is no way to achieve the
> italic visual presentation.

But this is *exactly* the problem that those arguing for better semantics
are stating.  All three above will render in italicized on screen font 99%
of the time.  However, just because you've given the text a "class" does not
ensure that you have attached meaning to the text.  All three examples above
lack semantic definition.

I suggested @role earlier as the W3C has already established a means to
attach "meaning" to whatever term is being used via RDF - a language
designed and used to define "meaning".  If the WHAT WG want to abandon @role
in favor of @class to attach semantic meanings fair enough; however it must
then also allow for the similar ability to define *what* the class means.
Bandying about "ship" and "copyright" has no real practical effect in this
*debate*, as it only scratches the surface of the discussion.

Attempting to "reserve" a series of class names might help, but it lacks the
scalability that all good systems should have; @role's ability to scale out
via RDF does not have this "problem".

Finally, there has been an insistence that those of us arguing for
more/better semantics (beyond <i> and <b>) provide "proof".  As Tina H has
stated, we've been attempting to show the shortcomings, and I would suggest
that the burden of proof for improved access and semantics using only @class
should be squarely on the shoulders of those who are suggesting that it is
sufficient.  Defend your position with real and practical "in the wild"
proof, instead of insisting that we "prove" otherwise: we can't yet, the
means does not exist and this is what we are asking for/demanding!

Received on Monday, 7 May 2007 14:15:22 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 30 April 2020 16:21:03 UTC