W3C home > Mailing lists > Public > public-html@w3.org > May 2008

Re: ARIA as stop-gap (was Re: Next steps for the ARIA syntax discussion)

From: Elliotte Harold <elharo@metalab.unc.edu>
Date: Fri, 30 May 2008 07:57:27 -0700
Message-ID: <484015D7.90704@metalab.unc.edu>
To: Steven Pemberton <steven.pemberton@cwi.nl>
Cc: Aaron M Leventhal <aleventh@us.ibm.com>, "L. David Baron" <dbaron@dbaron.org>, public-html@w3.org, public-xhtml2@w3.org, wai-xtech@w3.org, wai-xtech-request@w3.org, www-tag@w3.org

It seems to me the ARIA is stuck between Scylla and Charybdis. There is 
no good solution. The question is only which is least bad. I personally 
don't know the answer to that. My gut is to prefer namespaces, but there 
are some very good practical arguments against that.

It seems to me that the problem is primarily caused by attempting to 
shoehorn modular specs (ARIA) into a monolithic vocabulary (HTML). 
Similar problems have surfaced with the efforts to merge SVG and MathML 
into HTML. I increasingly don't think this will work. One of two 
plausible choices needs to be made:

1. Accept that HTML is monolithic. Merge ARIA into the core HTML spec, 
or incorporate it by reference. Make an ARIA attribute programmatically 
no different  than class or style or id. ARIA is just a part of HTML 5, 
and provides no support for other host languages.

2. Modularize HTML. Require XHTML. Provide no ARIA (or SVG, or MathML) 
support in classic HTML. If you want that, you must use well-formed XHTML.

I'm not sure the community has the will to make either of these choices; 
but if they don't ARIA, SVG, and MathML are all likely to fail. :-(

-- 
Elliotte Rusty Harold  elharo@metalab.unc.edu
Java I/O 2nd Edition Just Published!
http://www.cafeaulait.org/books/javaio2/
http://www.amazon.com/exec/obidos/ISBN=0596527500/ref=nosim/cafeaulaitA/
Received on Friday, 30 May 2008 14:58:13 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:55 UTC