W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2012

Re: title attributes on links

From: Steve Faulkner <faulkner.steve@gmail.com>
Date: Mon, 28 May 2012 12:05:31 +0100
Message-ID: <CA+ri+Vm2AaBK=Qn9eeg9TB=VfiqwJFHpyA5uc349oOPwXkbr4A@mail.gmail.com>
To: Harry Loots <harry.loots@ieee.org>
Cc: Ramón Corominas <listas@ramoncorominas.com>, Glen Wallis <glen.wallis@gmail.com>, w3c-wai-ig@w3.org
Hi Harry, i don't see how your example helps.

The use of value=search does not provide an accessible name for the
control. the addition of label around the input does nothing as far as I
can tell.

adding value=search means users have to clear the field text before they
input it.

perhaps using placeholder would provide the visible label (in supporting
browsers)

<input type="text" id="search" title="search" placeholder="search">

regards
stevef

On 28 May 2012 11:45, Harry Loots <harry.loots@ieee.org> wrote:

> Hi Steve
> I agree that there are situations where a visual label element is not
> appropriate.
>
> However, instead of  <input type=text title=search>
> <button>search</search> (your example), why not use <label><input
> type="text" id="search" value="search"></label><button>search</button>?
>
> Regards, Harry
>
>
> On Mon, May 28, 2012 at 12:16 PM, Steve Faulkner <faulkner.steve@gmail.com
> > wrote:
>
>> Hi Harry,
>>
>> the problem from a design point of view is that in some circumstances a
>> visual label is not appropriate as it is redundant. So where design
>> dictates that a label is inappropriate, use of the title or aria-label or
>> aria-labelldby provides an accessible name.
>>
>> Dragon has access to the accessible name information via the
>> accessibility API, it is up to the user agent in this case to provide the
>> access to the control.
>>
>> I understand what labels do and as a general rule they should be used for
>> the reasons you cite, but for example in many applications where a data
>> grid is used containing both read only text and form fields (example refer
>> to: Form controls in data tables:
>> http://www.usability.com.au/resources/wcag2/) then inclusion of a label
>> for each control is a non starter.
>>
>>
>> regards
>> steveF
>>
>>
>>
>> On 28 May 2012 10:51, Harry Loots <harry.loots@ieee.org> wrote:
>>
>>> Hi Steve
>>> The problem as a I see it if title is used instead of label, is that the
>>> functionality that comes with label, i.e., being able to click on it and
>>> place focus in the associated field element, is lost. Thus a Dragon user
>>> will have to execute multiple commands to achieve what they could have
>>> achieved with a single command. Similarly, users using magnification, can
>>> click on the label, when they've read it, and place the focus in the form
>>> field, without having to scroll horizontally.
>>> This capability is lost when the title attribute is used on the physical
>>> form element.
>>> Harry
>>>
>>>
>>> On Mon, May 28, 2012 at 10:37 AM, Steve Faulkner <
>>> faulkner.steve@gmail.com> wrote:
>>>
>>>> Hi Harry,
>>>>
>>>> Providing a control with an accessible name is a level A requirement:
>>>> http://www.w3.org/TR/WCAG20/#ensure-compat
>>>> using label or title provides a programmatically associated accessible
>>>> name for a control. The information is provided by the browsers through the
>>>> accessibility API Dragon can access this infromation, if it does not expose
>>>> it to the user its a bug in Dragon, not in WCAG 2.0.
>>>>
>>>> The use of title is only recommended under specific circumstances when
>>>> a visible label would be redundant.
>>>>
>>>> example:
>>>>
>>>> <input type=text title=search> <button>search</search>
>>>>
>>>> regards
>>>> SteveF
>>>>
>>>>
>>>> On 28 May 2012 09:13, Harry Loots <harry.loots@ieee.org> wrote:
>>>>
>>>>> Ramón Corominas wrote:
>>>>>
>>>>>> Which makes me wonder why it is acceptable (and even promoted) to use
>>>>>> the title instead of a <label> for form controls. Don't we have techniques
>>>>>> to hide labels preserving their accessibility?
>>>>>>
>>>>>
>>>>> Who and why would anyone want to promote a non-accessible technique?
>>>>> 'label' is an essential navigation mechanism for people using all kinds of
>>>>> AT, as I observed, once again, while testing software with a Dragon user.
>>>>> 'label' is even more important than 'alt'!!! While I can describe the
>>>>> contents of an image in the content, or even use the 'title' attribute,
>>>>> without losing functionality, whereas without the 'label' attribute I have
>>>>> reduced functionality and reduced mobility.
>>>>>
>>>>> Failure to use 'label', if it is not already so, should be a Level A
>>>>> infringement!
>>>>> Harry
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> with regards
>>>>
>>>> Steve Faulkner
>>>> Technical Director - TPG
>>>>
>>>> www.paciellogroup.com | www.HTML5accessibility.com |
>>>> www.twitter.com/stevefaulkner
>>>> HTML5: Techniques for providing useful text alternatives -
>>>> dev.w3.org/html5/alt-techniques/
>>>> Web Accessibility Toolbar - www.paciellogroup.com/resources/wat-ie-about.html
>>>>
>>>>
>>>>
>>>
>>
>>
>> --
>> with regards
>>
>> Steve Faulkner
>> Technical Director - TPG
>>
>> www.paciellogroup.com | www.HTML5accessibility.com |
>> www.twitter.com/stevefaulkner
>> HTML5: Techniques for providing useful text alternatives -
>> dev.w3.org/html5/alt-techniques/
>> Web Accessibility Toolbar - www.paciellogroup.com/resources/wat-ie-about.html
>>
>>
>>
>


-- 
with regards

Steve Faulkner
Technical Director - TPG

www.paciellogroup.com | www.HTML5accessibility.com |
www.twitter.com/stevefaulkner
HTML5: Techniques for providing useful text alternatives -
dev.w3.org/html5/alt-techniques/
Web Accessibility Toolbar - www.paciellogroup.com/resources/wat-ie-about.html
Received on Monday, 28 May 2012 11:06:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 28 May 2012 11:06:46 GMT