W3C home > Mailing lists > Public > public-html@w3.org > February 2010

Re: ISSUE-95 hidden - Chairs Solicit Proposals

From: Maciej Stachowiak <mjs@apple.com>
Date: Mon, 01 Feb 2010 00:28:39 -0800
Cc: HTML WG <public-html@w3.org>
Message-id: <AB1787D2-CDD9-41C8-9E87-D13983E1D739@apple.com>
To: Henri Sivonen <hsivonen@iki.fi>

On Feb 1, 2010, at 12:15 AM, Henri Sivonen wrote:

> On Jan 29, 2010, at 17:49, Leif Halvard Silli wrote:
> 
>> (1) irrelevant,hidden,conceal (Karl's proposal) are mysterious names 
>> which all raise the question "irrelevant,hidden,concealed from what?". 
> 
> FWIW, the attribute used to be called "irrelevant". It was renamed to "hidden" to align with ARIA that has "aria-hidden" for the same semantic.
> 
> HTML5 "hidden" is mostly redundant with "aria-hidden". It provides three differences:
> 1) "hidden" will eventually have UA style sheet support.
> 2) "hidden" in theory applies to non-accessibility unusual modalities while "aria-hidden", in theory, only matters to accessibility API mapping.
> 3) "hidden" is part of the host language on the principle that ARIA is a transitional solution and should become syntactically obsolete as host languages gain enough features to address the use cases.

Besides these functional differences, they also have different use cases. In my experience advising teams at Apple on adding ARIA markup to content, aria-hidden is most useful when you have content that you *do* want visible in normal rendering, but not through accessibility APIs. Of course, there are likewise valid use cases for content that is semantically hidden (not currently relevant) in all renderings.

So in conclusion, I am not sure that "hidden" fully obsoletes "aria-hidden". But it does have a valuable distinct use.

Regards,
Maciej
Received on Monday, 1 February 2010 08:29:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:17:01 GMT