W3C home > Mailing lists > Public > public-html@w3.org > February 2012

Re: ISSUE-30: Native HTML vs ARIA/XML

From: Charles Pritchard <chuck@jumis.com>
Date: Mon, 13 Feb 2012 14:24:10 -0800
Message-ID: <4F398D8A.6090109@jumis.com>
To: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
CC: HTMLWG WG <public-html@w3.org>
On 2/13/2012 5:32 AM, Leif Halvard Silli wrote:
> If these words are true, why are we then trying to ARIA-fy everything
> related to HTML accessibility? In particular, why remove native
> features? How can Jonas be right when he claims that it will be simpler
> to be able to say "just use ARIA" as opposed to "use @longdesc for
> <img>  and @summary for table and …" ? (I noted that Rich Schwerdtfeger
> from the ARIA community once last year said a similar thing: that ARIA
> in the long run could take over even for @alt.)

The direction has usually been to bring things from ARIA into HTML5.

ARIA was designed because HTML a11y was falling apart in HTML4.  Authors 
started using JS to style menus and implement form elements and such.
HTML5 has adopted a lot of vocabulary first introduced in ARIA.

For the time being, it's useful to be redundant in HTML5, and to include 
ARIA along with HTML5 semantics.
Such is the case for something like range and slider, where the UA may 
not have the field implemented.

Partial example:
<input type="range" role="slider" aria-valuemin="0" min="0" />

As for ARIA itself -- it can run independently of HTML. It can be used 
by desktop applications.
CSS can also run independently of HTML. Sometimes, these days, it feels 
like it already does.

It looks like vendor-prefixed CSS and non-semantic tags (div and span), 
and ARIA markup.
In some cases, HTML5 is more of a shortcut for CSS+ARIA, a subset.

It confirms that we have a rich set of primitives. Building consensus on 
CSS vocabulary is slow. So it goes.
It can be a lot of work to support new css vocabulary.

The thread seems to be about simplifying HTML5. Getting rid of <font>, 
things like that.
Many of the votes against @longdesc in the last poll talked about why 
they see @longdesc as inferior to ARIA.

I think in many ways it is, and that those authors ought to have 
supplemented or replaced their @longdesc efforts with ARIA.
Those votes though, seem to see @longdesc as threatening. I follow more 
of a polyfill methodology.

Received on Monday, 13 February 2012 22:24:50 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:16:20 UTC