W3C home > Mailing lists > Public > public-html@w3.org > March 2011

Re: example spec text for longdesc

From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
Date: Tue, 22 Mar 2011 21:06:36 +0100
To: Ian Hickson <ian@hixie.ch>
Cc: Laura Carlson <laura.lee.carlson@gmail.com>, Steve Faulkner <faulkner.steve@gmail.com>, Jonas Sicking <jonas@sicking.cc>, Henri Sivonen <hsivonen@iki.fi>, HTML Accessibility Task Force <public-html-a11y@w3.org>, HTMLWG WG <public-html@w3.org>
Message-ID: <20110322210636450387.c5186106@xn--mlform-iua.no>
Ian Hickson, Tue, 22 Mar 2011 15:41:54 +0000 (UTC):
> On Tue, 22 Mar 2011, Laura Carlson wrote:
>> 
>> Ian, I would really appreciate your advice on that spec text too. What 
>> is technically wrong with it?
> 
> Well, it reintroduces longdesc, a feature which is almost universally 
> misused and will therefore do basically nothing but cause users pain, 
> something which has been repeatedly explained.

The massive misuse leads to massive distrust argument?

96% of @longdesc's errors can easily detected. [1] Hence HTML5 
conformance checkers could easily detect both those 9&% as well as 
other common misuses that Laura's CP proposes to check for.

W.r.t. the repeated explanations, then they have focused on well 
intended/accessibility inspired misunderstandings and errors. However, 
what has gotten little attention is the *conscious* misuse by image 
gallery software etc. 

One of those developers even approached you, Ian, suggesting a 
@fullsize attribute: http://addfullsize.com. He says he got a negative 
reply from you and states that he therefore uses the "fully valid" 
@longdesc instead.

Taking away @longdesc creates the impression that HTML5 has introduced 
features that can replace it. For example the idea that 
aria-describedby can replace it [1]. In case of fullsize, then taking 
@longdesc away *perhaps* will make those developers start using 
@data-fullsize? 

> That's the main problem 
> with it. I didn't examine it any closer since that's pretty much a fatal 
> problem with the text as it is.

It seems like another main problem is the very fact that @longdesc 
belongs to the <img> element and that it, for the <img> element, is a 
very common thing to be wanting to present both a small and a large 
image.

[1] http://blog.whatwg.org/the-longdesc-lottery

[1] http://www.w3.org/Bugs/Public/show_bug.cgi?id=12243#c2

-- 
leif halvard silli
Received on Tuesday, 22 March 2011 20:07:13 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:39:23 UTC