E-mail List Archives
Thread: Is use of <label> and title redundant?
Number of posts in this thread: 43 (In chronological order)
From: Angela French
Date: Fri, Nov 09 2012 11:22AM
Subject: Is use of <label> and title redundant?
No previous message | Next message →
Hello,
Is the use of a <label> tag to identify association with a text field and the simultaneous use of the title attribute in the <input> tag redundant, such that it should be avoided? In other words, would a screen reader say the word twice?
Example:
<label for="searchfield" class="offscreen"> Search</label><input id="searchfield" type="text" title="Search">
Thank you,
Angela French
Internet Specialist
State Board for Community and Technical Colleges
360-704-4316
= EMAIL ADDRESS REMOVED =
http://www.checkoutacollege.com/
From: Léonie Watson
Date: Fri, Nov 09 2012 12:32PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Angela French wrote:
" Is the use of a <label> tag to identify association with a text field and
the simultaneous use of the title attribute in the <input> tag redundant,
such that it should be avoided? In other words, would a screen reader say
the word twice?"
Most screen readers don't acknowledge the title attribute by default,
although they can be configured to do so. VoiceOver is the exception I
think.
Assuming the label is visible, it's worth asking what value the title
attribute brings in this case?
Léonie.
From: Angela French
Date: Fri, Nov 09 2012 12:40PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
I've discovered that on my site there is such redundancy but the <label> is positioned off screen.
>Angela French wrote:
>" Is the use of a <label> tag to identify association with a text field and the
>simultaneous use of the title attribute in the <input> tag redundant, such
>that it should be avoided? In other words, would a screen reader say the
>word twice?"
>
>Most screen readers don't acknowledge the title attribute by default,
>although they can be configured to do so. VoiceOver is the exception I think.
>
>Assuming the label is visible, it's worth asking what value the title attribute
>brings in this case?
>
>
>Léonie.
>
From: Bourne, Sarah (ITD)
Date: Fri, Nov 09 2012 1:24PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Despite ancient "best practices" that had us put them in, we've gotten rid of almost all title attributes that repeat other elements or attributes on the Mass.Gov site. What pushed us to action was that it was hurting our search engine rankings. It seems machines don't like redundancy any more than screen reader users.
sb
Sarah E. Bourne
Director of Assistive Technology &
Mass.Gov Chief Technology Strategist
Information Technology Division
Commonwealth of Massachusetts
1 Ashburton Pl. rm 1601 Boston MA 02108
617-626-4502
= EMAIL ADDRESS REMOVED =
http://www.mass.gov/itd
From: Ryan E. Benson
Date: Fri, Nov 09 2012 4:59PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Léonie Watson wrote:
> Most screen readers don't acknowledge the title attribute by default,
> although they can be configured to do so. VoiceOver is the exception I
> think.
ZoomText (v9.5 at least) will read the title attribute instead of the
label. The same goes with links with a title attribute, it reads the
title versus linking words.
--
Ryan E. Benson
On Fri, Nov 9, 2012 at 2:32 PM, Léonie Watson < = EMAIL ADDRESS REMOVED = > wrote:
> Angela French wrote:
> " Is the use of a <label> tag to identify association with a text field and
> the simultaneous use of the title attribute in the <input> tag redundant,
> such that it should be avoided? In other words, would a screen reader say
> the word twice?"
>
> Most screen readers don't acknowledge the title attribute by default,
> although they can be configured to do so. VoiceOver is the exception I
> think.
>
> Assuming the label is visible, it's worth asking what value the title
> attribute brings in this case?
>
>
> Léonie.
>
>
From: Michael R. Burks
Date: Fri, Nov 09 2012 5:50PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
It has not been my experience that title attributes work well with Dragon
Sincerely,
Mike Burks
919-882-1884 - Fax
919-349-6661 - Office
From: Michael R. Burks
Date: Fri, Nov 09 2012 5:56PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
I would like to add that we should code to the standards, not to a specific
type of Assistive Technology.
Sincerely,
Mike Burks
From: Sailesh Panchang
Date: Sat, Nov 10 2012 5:53PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
As per WCAG2 techniques, title on form controls should be used when
visible label is not present ... i.e. when purpose of form control is
obvious generally to sighted users / as per UI design.
(If purpose is not obvious visually, then the fix is to incorporate a
visual text label for the benefit of all users).
Using title in these situations mainly to convey purpose of control to
non-sighted users is well supported by browsers and AT, both recent
and legacy ones.
Title is also available via mouseover to some sighted users who might
need them.
I do not recommend using off-screen label: it is more work and code
and not available to anyone except screen reader / magnification users
unless CSS is turned off.
Sailesh Panchang
Deque Systems
On 11/9/12, Michael R. Burks < = EMAIL ADDRESS REMOVED = > wrote:
> I would like to add that we should code to the standards, not to a specific
> type of Assistive Technology.
>
> Sincerely,
>
> Mike Burks
>
>
>
From: Sailesh Panchang
Date: Mon, Nov 12 2012 1:21PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Patrick,
Screen magnification s/w like Zoomtext expose / speak off-screen text labels.
Simply using the OS' / browser's features to magnify will not expose
off-screen labels.
Off-screen technique should be used only where necessary;
Inappropriate / over usage results in duplication and may not be
available to magnification users in some situations.
Sailesh
On 11/10/12, Patrick Dunphy < = EMAIL ADDRESS REMOVED = > wrote:
> I practice that there are usually only 2 instances where I consider
> acceptable use of title attributes. First being for Iframe usage & the
> second would be as you mentioned - form controls such as a site's global
> search.
>
> Can you please clarify something you've said? How is an off screen label
> available for magnification users? I wouldn't have thought that to be the
> case. Also curious to hear your thoughts about this being applicable to
> other off screen text usages - making text links understood out of context &
> indicating position within a section nav.
>
> Thanks.
> -PD
>
>
> On 2012-11-10, at 7:53 PM, Sailesh Panchang < = EMAIL ADDRESS REMOVED = >
> wrote:
>
>> As per WCAG2 techniques, title on form controls should be used when
>> visible label is not present ... i.e. when purpose of form control is
>> obvious generally to sighted users / as per UI design.
>> (If purpose is not obvious visually, then the fix is to incorporate a
>> visual text label for the benefit of all users).
>>
>> Using title in these situations mainly to convey purpose of control to
>> non-sighted users is well supported by browsers and AT, both recent
>> and legacy ones.
>> Title is also available via mouseover to some sighted users who might
>> need them.
>> I do not recommend using off-screen label: it is more work and code
>> and not available to anyone except screen reader / magnification users
>> unless CSS is turned off.
>> Sailesh Panchang
>> Deque Systems
>>
>>
>> On 11/9/12, Michael R. Burks < = EMAIL ADDRESS REMOVED = > wrote:
>>> I would like to add that we should code to the standards, not to a
>>> specific
>>> type of Assistive Technology.
>>>
>>> Sincerely,
>>>
>>> Mike Burks
>>>
>>>
>>>
From: Patrick Dunphy
Date: Mon, Nov 12 2012 3:06PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Thank you Sailesh for clarifying.. Zoomtext isn't something I've worked
with so I feel better having read your response. The instances where I
prefer to use the off screen technique isn't for duplication rather to give
more context when appropriate.
Thanks again - I appreciate the response.
On Nov 12, 2012 3:22 PM, "Sailesh Panchang" < = EMAIL ADDRESS REMOVED = >
wrote:
> Patrick,
> Screen magnification s/w like Zoomtext expose / speak off-screen text
> labels.
> Simply using the OS' / browser's features to magnify will not expose
> off-screen labels.
> Off-screen technique should be used only where necessary;
> Inappropriate / over usage results in duplication and may not be
> available to magnification users in some situations.
> Sailesh
>
>
> On 11/10/12, Patrick Dunphy < = EMAIL ADDRESS REMOVED = > wrote:
> > I practice that there are usually only 2 instances where I consider
> > acceptable use of title attributes. First being for Iframe usage & the
> > second would be as you mentioned - form controls such as a site's global
> > search.
> >
> > Can you please clarify something you've said? How is an off screen label
> > available for magnification users? I wouldn't have thought that to be the
> > case. Also curious to hear your thoughts about this being applicable to
> > other off screen text usages - making text links understood out of
> context &
> > indicating position within a section nav.
> >
> > Thanks.
> > -PD
> >
> >
> > On 2012-11-10, at 7:53 PM, Sailesh Panchang < = EMAIL ADDRESS REMOVED = >
> > wrote:
> >
> >> As per WCAG2 techniques, title on form controls should be used when
> >> visible label is not present ... i.e. when purpose of form control is
> >> obvious generally to sighted users / as per UI design.
> >> (If purpose is not obvious visually, then the fix is to incorporate a
> >> visual text label for the benefit of all users).
> >>
> >> Using title in these situations mainly to convey purpose of control to
> >> non-sighted users is well supported by browsers and AT, both recent
> >> and legacy ones.
> >> Title is also available via mouseover to some sighted users who might
> >> need them.
> >> I do not recommend using off-screen label: it is more work and code
> >> and not available to anyone except screen reader / magnification users
> >> unless CSS is turned off.
> >> Sailesh Panchang
> >> Deque Systems
> >>
> >>
> >> On 11/9/12, Michael R. Burks < = EMAIL ADDRESS REMOVED = > wrote:
> >>> I would like to add that we should code to the standards, not to a
> >>> specific
> >>> type of Assistive Technology.
> >>>
> >>> Sincerely,
> >>>
> >>> Mike Burks
> >>>
> >>>
> >>>
From: Angela French
Date: Tue, Nov 13 2012 9:31AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
The context of my question is a search field. One rarely sees the search field actually labeled (at least on screen). The text on the button reads "Search" and is therefore meant to convey the functionality. But of course the button follows the form field.
Angela French
: Re: [WebAIM] Is use of <label> and title redundant?
>
>As per WCAG2 techniques, title on form controls should be used when visible
>label is not present ... i.e. when purpose of form control is obvious generally
>to sighted users / as per UI design.
>(If purpose is not obvious visually, then the fix is to incorporate a visual text
>label for the benefit of all users).
>
>Using title in these situations mainly to convey purpose of control to non-
>sighted users is well supported by browsers and AT, both recent and legacy
>ones.
>Title is also available via mouseover to some sighted users who might need
>them.
>I do not recommend using off-screen label: it is more work and code and not
>available to anyone except screen reader / magnification users unless CSS is
>turned off.
>Sailesh Panchang
>Deque Systems
>
From: GILLENWATER, ZOE M
Date: Mon, Nov 19 2012 6:41AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Hi Sailesh,
One advantage for hidden labels that I've found when testing the difference between them and title attributes on form fields is that a few screen reader-browser combinations will not read the title text in normal reading mode, only when focused in the form field. The hidden label always gets read in both normal and forms reading modes. This seems non-trivial to me, as hearing label-less form fields announced would be confusing and doesn't give you the information you need to be able to decide whether you want to go into forms reading mode. Have you never seen this behavior with title text in normal reading mode?
I've even seen cases where not only is the title text not announced in normal reading mode, but the type of form field is not announced, so you have no idea anything is even there to enter forms mode for. Granted, this has only happened to me in older screen readers, but still, it's a pretty big problem. Again, this is just in normal reading mode--I've had no problems with title text in forms reading mode. But because of this, using regular old labels seems far safer in many contexts. Title text has its place too, but I'm starting to think that hidden labels should be the technique of first resort (apart from adding visible labels whenever possible, of course!).
I'd love to hear differing opinions on this! :-)
Thanks,
Zoe
Zoe Gillenwater
Web Accessibility Technical Architect
AT&T Consumer Digital Experience
o: 919-241-4083
e: = EMAIL ADDRESS REMOVED =
4625 Creekstone Dr | Durham, NC 27703
This email and any files transmitted with it are AT&T property, are confidential, and are intended solely for the use of the individual or entity to whom this email is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender and delete this message immediately from your computer. Any other uses, retention, dissemination, forwarding, printing, or copying of this email is strictly prohibited.
From: Steve Faulkner
Date: Mon, Nov 19 2012 6:59AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Hi Zoe,
When a control lacks another form of explicit labelling, the title
attribute content is mapped to the accessible name for the control. This is
consistent across plaforms and accessibility APIs, this has been the case
for a long time and is the standardized behaviour [1].
While we should take into account the bugs in AT when newer technologies
are used (example ARIA), we should be filing bugs against those AT/browsers
that do not follow the standards as trying to code to take into account
bugs in every AT that do not make use of the accessibility information
exposed in a standardized/agreed way is a losing proposition.
[1] http://dvcs.w3.org/hg/html-api-map/raw-file/tip/Overview.html#calc
regards
SteveF
On 19 November 2012 13:41, GILLENWATER, ZOE M < = EMAIL ADDRESS REMOVED = > wrote:
> Hi Sailesh,
>
> One advantage for hidden labels that I've found when testing the
> difference between them and title attributes on form fields is that a few
> screen reader-browser combinations will not read the title text in normal
> reading mode, only when focused in the form field. The hidden label always
> gets read in both normal and forms reading modes. This seems non-trivial to
> me, as hearing label-less form fields announced would be confusing and
> doesn't give you the information you need to be able to decide whether you
> want to go into forms reading mode. Have you never seen this behavior with
> title text in normal reading mode?
>
> I've even seen cases where not only is the title text not announced in
> normal reading mode, but the type of form field is not announced, so you
> have no idea anything is even there to enter forms mode for. Granted, this
> has only happened to me in older screen readers, but still, it's a pretty
> big problem. Again, this is just in normal reading mode--I've had no
> problems with title text in forms reading mode. But because of this, using
> regular old labels seems far safer in many contexts. Title text has its
> place too, but I'm starting to think that hidden labels should be the
> technique of first resort (apart from adding visible labels whenever
> possible, of course!).
>
> I'd love to hear differing opinions on this! :-)
>
> Thanks,
> Zoe
>
> Zoe Gillenwater
> Web Accessibility Technical Architect
> AT&T Consumer Digital Experience
>
> o: 919-241-4083
> e: = EMAIL ADDRESS REMOVED =
>
> 4625 Creekstone Dr | Durham, NC 27703
>
> This email and any files transmitted with it are AT&T property, are
> confidential, and are intended solely for the use of the individual or
> entity to whom this email is addressed. If you are not one of the named
> recipient(s) or otherwise have reason to believe that you have received
> this message in error, please notify the sender and delete this message
> immediately from your computer. Any other uses, retention, dissemination,
> forwarding, printing, or copying of this email is strictly prohibited.
>
>
>
From: GILLENWATER, ZOE M
Date: Mon, Nov 19 2012 7:42AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Sure, but if there's another completely valid, easy way to label an element that is not as buggy--namely, using a label!--why not use that method?
Zoe
Zoe Gillenwater
Web Accessibility Technical Architect
AT&T Consumer Digital Experience
o: 919-241-4083
e: = EMAIL ADDRESS REMOVED =
4625 Creekstone Dr | Durham, NC 27703
This email and any files transmitted with it are AT&T property, are confidential, and are intended solely for the use of the individual or entity to whom this email is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender and delete this message immediately from your computer. Any other uses, retention, dissemination, forwarding, printing, or copying of this email is strictly prohibited.
From: Steve Faulkner
Date: Mon, Nov 19 2012 7:55AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
>Sure, but if there's another completely valid, easy way to label an
element that is not as buggy--namely, using a label!--why not use that
method?
I have encountered buggy screen reader behaviour with hidden label elements
in past. in the end you pick what you think works best.
regards
steve
On 19 November 2012 14:42, GILLENWATER, ZOE M < = EMAIL ADDRESS REMOVED = > wrote:
> Sure, but if there's another completely valid, easy way to label an
> element that is not as buggy--namely, using a label!--why not use that
> method?
>
> Zoe
>
> Zoe Gillenwater
> Web Accessibility Technical Architect
> AT&T Consumer Digital Experience
>
> o: 919-241-4083
> e: = EMAIL ADDRESS REMOVED =
>
> 4625 Creekstone Dr | Durham, NC 27703
>
> This email and any files transmitted with it are AT&T property, are
> confidential, and are intended solely for the use of the individual or
> entity to whom this email is addressed. If you are not one of the named
> recipient(s) or otherwise have reason to believe that you have received
> this message in error, please notify the sender and delete this message
> immediately from your computer. Any other uses, retention, dissemination,
> forwarding, printing, or copying of this email is strictly prohibited.
>
>
>
From: Sailesh Panchang
Date: Mon, Nov 19 2012 8:09AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Zoe,
Certainly I have seen that behavior where the screen reader reads the
off-screen label as one is arrowing down the page.
But then one is reading content that is not visually available on the
page and in that mode the label is really not programmatically
associated with the form control.
The user has to use his/her understanding / judgment to make this determination.
Off-screen text that helps to convey structure / info-relationships
(e.g. headings) when no other cues are available is alright but
otherwise visual content should generally match / be in sync with
screen reader output.
In forms mode, the title clearly conveys the purpose of the control.
This technique is in line withH65 of WCAG 2 techniques.
Thanks,
Sailesh Panchang
www.deque.com
Tel 703-225-0380 ext 105
On 11/19/12, GILLENWATER, ZOE M < = EMAIL ADDRESS REMOVED = > wrote:
> Hi Sailesh,
>
> One advantage for hidden labels that I've found when testing the difference
> between them and title attributes on form fields is that a few screen
> reader-browser combinations will not read the title text in normal reading
> mode, only when focused in the form field. The hidden label always gets read
> in both normal and forms reading modes. This seems non-trivial to me, as
> hearing label-less form fields announced would be confusing and doesn't give
> you the information you need to be able to decide whether you want to go
> into forms reading mode. Have you never seen this behavior with title text
> in normal reading mode?
>
> I've even seen cases where not only is the title text not announced in
> normal reading mode, but the type of form field is not announced, so you
> have no idea anything is even there to enter forms mode for. Granted, this
> has only happened to me in older screen readers, but still, it's a pretty
> big problem. Again, this is just in normal reading mode--I've had no
> problems with title text in forms reading mode. But because of this, using
> regular old labels seems far safer in many contexts. Title text has its
> place too, but I'm starting to think that hidden labels should be the
> technique of first resort (apart from adding visible labels whenever
> possible, of course!).
>
> I'd love to hear differing opinions on this! :-)
>
> Thanks,
> Zoe
>
> Zoe Gillenwater
> Web Accessibility Technical Architect
> AT&T Consumer Digital Experience
>
> o: 919-241-4083
> e: = EMAIL ADDRESS REMOVED =
>
> 4625 Creekstone Dr | Durham, NC 27703
>
> This email and any files transmitted with it are AT&T property, are
> confidential, and are intended solely for the use of the individual or
> entity to whom this email is addressed. If you are not one of the named
> recipient(s) or otherwise have reason to believe that you have received this
> message in error, please notify the sender and delete this message
> immediately from your computer. Any other uses, retention, dissemination,
> forwarding, printing, or copying of this email is strictly prohibited.
>
>
>
From: GILLENWATER, ZOE M
Date: Mon, Nov 19 2012 8:10AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Hi Steve,
Can you share what the bugs you found are? I've had trouble with hidden labels if the hiding method had overflow:hidden on it, but removing that got rid of all my problems. Here's the hiding CSS I've tested without problems so far:
.hidden-spoken {
position: absolute !important;
clip: rect(1px 1px 1px 1px);
clip: rect(1px, 1px, 1px, 1px);
padding:0 !important;
border:0 !important;
height: 1px !important;
width: 1px !important;
}
As I said, I'd love to hear differing opinions on this topic! I posted a thread on this issue months ago because I'm genuinely trying to figure out the pros and cons to different approaches, and I still haven't made a decision as to which will be my method of first resort.
Thanks for your help,
Zoe
Zoe Gillenwater
Web Accessibility Technical Architect
AT&T Consumer Digital Experience
o: 919-241-4083
e: = EMAIL ADDRESS REMOVED =
4625 Creekstone Dr | Durham, NC 27703
This email and any files transmitted with it are AT&T property, are confidential, and are intended solely for the use of the individual or entity to whom this email is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender and delete this message immediately from your computer. Any other uses, retention, dissemination, forwarding, printing, or copying of this email is strictly prohibited.
From: Sailesh Panchang
Date: Mon, Nov 19 2012 8:18AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Steve,
I think I had previously expressed my concern about the logic stated in
HTML to Platform Accessibility APIs Implementation Guide.
I see that the algorithm accords a higher order to ARIA attributesand
not the native HTML elements / attributes.
This is contrary to what the Intro to ARIA advises : use ARIA only
where native technology features are absent or inadequate.
So why should standard INPUT type=text with a proper LABEL with for-id
association be trumped by aria-labelledby attribute?
One should use aria-labelledby in situations like a custom form
control that will not support LABEL with for-id technique.
Or, where multiple labels need to be associated with a form control
and browser/ATs do not support that well / uniformly.
I agree ARIA attributes are very useful (as stated above) to relate
visually available content on the page with other elements like form
controls on the page.
But I do not see the benefit of placing off-screen content and, say,
using aria-describedby to relate that text to a form control as its
accessible-description when the good old title accomplishes the same.
And then according aria-describedby a higher status than the title.
Anyway the off-screen text is not available to non AT-users.
The title is at least available to mouse users besides AT users.
Thanks and regards,
Sailesh Panchang
www.deque.com
Tel 703-225-0380 ext 105
On 11/19/12, Steve Faulkner < = EMAIL ADDRESS REMOVED = > wrote:
>>Sure, but if there's another completely valid, easy way to label an
> element that is not as buggy--namely, using a label!--why not use that
> method?
>
> I have encountered buggy screen reader behaviour with hidden label elements
> in past. in the end you pick what you think works best.
>
> regards
> steve
>
> On 19 November 2012 14:42, GILLENWATER, ZOE M < = EMAIL ADDRESS REMOVED = > wrote:
>
>> Sure, but if there's another completely valid, easy way to label an
>> element that is not as buggy--namely, using a label!--why not use that
>> method?
>>
>> Zoe
>>
>> Zoe Gillenwater
>> Web Accessibility Technical Architect
>> AT&T Consumer Digital Experience
>>
>> o: 919-241-4083
>> e: = EMAIL ADDRESS REMOVED =
>>
>> 4625 Creekstone Dr | Durham, NC 27703
>>
>> This email and any files transmitted with it are AT&T property, are
>> confidential, and are intended solely for the use of the individual or
>> entity to whom this email is addressed. If you are not one of the named
>> recipient(s) or otherwise have reason to believe that you have received
>> this message in error, please notify the sender and delete this message
>> immediately from your computer. Any other uses, retention,
>> dissemination,
>> forwarding, printing, or copying of this email is strictly prohibited.
>>
>>
>>
From: GILLENWATER, ZOE M
Date: Mon, Nov 19 2012 8:18AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Hi Sailesh,
I'm afraid I don't understand your feedback. Do you think it's good for a screen reader to not read a form field's title text in normal reading mode? Or do you think a screen reader shouldn't read a hidden label in normal reading mode? Or do you mean something else entirely?
Even if a label is hidden visually, it's still programmatically connected to the field, so I don't understand what you mean by this part of your response either...
Genuinely trying to figure this out and do the right thing!
Thanks,
Zoe
Zoe Gillenwater
Web Accessibility Technical Architect
AT&T Consumer Digital Experience
o: 919-241-4083
e: = EMAIL ADDRESS REMOVED =
4625 Creekstone Dr | Durham, NC 27703
This email and any files transmitted with it are AT&T property, are confidential, and are intended solely for the use of the individual or entity to whom this email is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender and delete this message immediately from your computer. Any other uses, retention, dissemination, forwarding, printing, or copying of this email is strictly prohibited.
From: Steve Faulkner
Date: Mon, Nov 19 2012 8:25AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Hi Sailesh,
>I think I had previously expressed my concern about the logic stated in
>HTML to Platform Accessibility APIs Implementation Guide.
the API guide only reflects what is implemented in browsers in this regard.
The API mappings and algorithms are distinct from the advice in ARIA about
what authors should do.
ARIA also states that ARIA roles override native roles, which is what is
implemented in browsers.
example:
<button role=checkbox> is exposed as a checkbox not a button.
whether authors are advised to do this is a different matter.
regards
Steve
On 19 November 2012 15:18, Sailesh Panchang < = EMAIL ADDRESS REMOVED = >wrote:
> Steve,
> I think I had previously expressed my concern about the logic stated in
> HTML to Platform Accessibility APIs Implementation Guide.
> I see that the algorithm accords a higher order to ARIA attributesand
> not the native HTML elements / attributes.
> This is contrary to what the Intro to ARIA advises : use ARIA only
> where native technology features are absent or inadequate.
> So why should standard INPUT type=text with a proper LABEL with for-id
> association be trumped by aria-labelledby attribute?
>
> One should use aria-labelledby in situations like a custom form
> control that will not support LABEL with for-id technique.
> Or, where multiple labels need to be associated with a form control
> and browser/ATs do not support that well / uniformly.
>
> I agree ARIA attributes are very useful (as stated above) to relate
> visually available content on the page with other elements like form
> controls on the page.
> But I do not see the benefit of placing off-screen content and, say,
> using aria-describedby to relate that text to a form control as its
> accessible-description when the good old title accomplishes the same.
> And then according aria-describedby a higher status than the title.
> Anyway the off-screen text is not available to non AT-users.
> The title is at least available to mouse users besides AT users.
> Thanks and regards,
> Sailesh Panchang
> www.deque.com
>
> Tel 703-225-0380 ext 105
>
>
>
>
>
>
> On 11/19/12, Steve Faulkner < = EMAIL ADDRESS REMOVED = > wrote:
> >>Sure, but if there's another completely valid, easy way to label an
> > element that is not as buggy--namely, using a label!--why not use that
> > method?
> >
> > I have encountered buggy screen reader behaviour with hidden label
> elements
> > in past. in the end you pick what you think works best.
> >
> > regards
> > steve
> >
> > On 19 November 2012 14:42, GILLENWATER, ZOE M < = EMAIL ADDRESS REMOVED = > wrote:
> >
> >> Sure, but if there's another completely valid, easy way to label an
> >> element that is not as buggy--namely, using a label!--why not use that
> >> method?
> >>
> >> Zoe
> >>
> >> Zoe Gillenwater
> >> Web Accessibility Technical Architect
> >> AT&T Consumer Digital Experience
> >>
> >> o: 919-241-4083
> >> e: = EMAIL ADDRESS REMOVED =
> >>
> >> 4625 Creekstone Dr | Durham, NC 27703
> >>
> >> This email and any files transmitted with it are AT&T property, are
> >> confidential, and are intended solely for the use of the individual or
> >> entity to whom this email is addressed. If you are not one of the named
> >> recipient(s) or otherwise have reason to believe that you have received
> >> this message in error, please notify the sender and delete this message
> >> immediately from your computer. Any other uses, retention,
> >> dissemination,
> >> forwarding, printing, or copying of this email is strictly prohibited.
> >>
> >>
> >>
From: Sailesh Panchang
Date: Mon, Nov 19 2012 9:39AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
>ARIA also states that ARIA roles override native roles, which is what is
>implemented in browsers.
Yes I have noted that too and a link with role=button is exposed as a button.
So then is there a disconnect in the ARIA specs? I mean is the
recommendation not in sync with the intent?
Sailesh
On 11/19/12, Steve Faulkner < = EMAIL ADDRESS REMOVED = > wrote:
> Hi Sailesh,
>
>>I think I had previously expressed my concern about the logic stated in
>>HTML to Platform Accessibility APIs Implementation Guide.
>
> the API guide only reflects what is implemented in browsers in this regard.
>
> The API mappings and algorithms are distinct from the advice in ARIA about
> what authors should do.
>
> ARIA also states that ARIA roles override native roles, which is what is
> implemented in browsers.
>
> example:
>
> <button role=checkbox> is exposed as a checkbox not a button.
>
> whether authors are advised to do this is a different matter.
>
> regards
> Steve
>
>
>
> On 19 November 2012 15:18, Sailesh Panchang
> < = EMAIL ADDRESS REMOVED = >wrote:
>
>> Steve,
>> I think I had previously expressed my concern about the logic stated in
>> HTML to Platform Accessibility APIs Implementation Guide.
>> I see that the algorithm accords a higher order to ARIA attributesand
>> not the native HTML elements / attributes.
>> This is contrary to what the Intro to ARIA advises : use ARIA only
>> where native technology features are absent or inadequate.
>> So why should standard INPUT type=text with a proper LABEL with for-id
>> association be trumped by aria-labelledby attribute?
>>
>> One should use aria-labelledby in situations like a custom form
>> control that will not support LABEL with for-id technique.
>> Or, where multiple labels need to be associated with a form control
>> and browser/ATs do not support that well / uniformly.
>>
>> I agree ARIA attributes are very useful (as stated above) to relate
>> visually available content on the page with other elements like form
>> controls on the page.
>> But I do not see the benefit of placing off-screen content and, say,
>> using aria-describedby to relate that text to a form control as its
>> accessible-description when the good old title accomplishes the same.
>> And then according aria-describedby a higher status than the title.
>> Anyway the off-screen text is not available to non AT-users.
>> The title is at least available to mouse users besides AT users.
>> Thanks and regards,
>> Sailesh Panchang
>> www.deque.com
>>
>> Tel 703-225-0380 ext 105
>>
>>
>>
>>
>>
>>
>> On 11/19/12, Steve Faulkner < = EMAIL ADDRESS REMOVED = > wrote:
>> >>Sure, but if there's another completely valid, easy way to label an
>> > element that is not as buggy--namely, using a label!--why not use that
>> > method?
>> >
>> > I have encountered buggy screen reader behaviour with hidden label
>> elements
>> > in past. in the end you pick what you think works best.
>> >
>> > regards
>> > steve
>> >
>> > On 19 November 2012 14:42, GILLENWATER, ZOE M < = EMAIL ADDRESS REMOVED = > wrote:
>> >
>> >> Sure, but if there's another completely valid, easy way to label an
>> >> element that is not as buggy--namely, using a label!--why not use that
>> >> method?
>> >>
>> >> Zoe
>> >>
>> >> Zoe Gillenwater
>> >> Web Accessibility Technical Architect
>> >> AT&T Consumer Digital Experience
>> >>
>> >> o: 919-241-4083
>> >> e: = EMAIL ADDRESS REMOVED =
>> >>
>> >> 4625 Creekstone Dr | Durham, NC 27703
>> >>
>> >> This email and any files transmitted with it are AT&T property, are
>> >> confidential, and are intended solely for the use of the individual or
>> >> entity to whom this email is addressed. If you are not one of the
>> >> named
>> >> recipient(s) or otherwise have reason to believe that you have
>> >> received
>> >> this message in error, please notify the sender and delete this
>> >> message
>> >> immediately from your computer. Any other uses, retention,
>> >> dissemination,
>> >> forwarding, printing, or copying of this email is strictly prohibited.
>> >>
>> >>
>> >>
From: Sailesh Panchang
Date: Mon, Nov 19 2012 10:12AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
On 11/19/12, GILLENWATER, ZOE M < = EMAIL ADDRESS REMOVED = > wrote:
> Hi Sailesh,
>
> I'm afraid I don't understand your feedback. Do you think it's good for a
> screen reader to not read a form field's title text in normal reading mode?
> Or do you think a screen reader shouldn't read a hidden label in normal
> reading mode? Or do you mean something else entirely?
Sailesh: It is generally in situations where the UI design limits the
ability of developers to place a visible text label (like a search
edit box or controls in a table with row/col headers that serve as
'labels') that a title works adequately.
In these situations I am not adversely impacted when the screen
reader does not read the title in browse mode.
So an off-screen label is not a game changer in these circs. It does
not add a whole lot of accessibility benefit.
It is like a data table: the screen reader does not speak row/col
headers unless one is in table-nav mode.
Most screen readers speak the alt for an image in browse mode and do
not speak the title for a form control. It has been so always maybe
for a reason.
If the visible label text is not adequate for non-sighted users,
appending off-screen text to the label is fine.
>
> Even if a label is hidden visually, it's still programmatically connected to
> the field, so I don't understand what you mean by this part of your response
> either...
Sailesh: Sure, but that association kicks-in only in forms mode and
for non-sighted users a label is no different from any other piece of
text on the page in non-forms mode.
Sailesh Panchang
www.deque.com
Tel 703-225-0380 ext 105
>
> Genuinely trying to figure this out and do the right thing!
>
> Thanks,
> Zoe
>
> Zoe Gillenwater
> Web Accessibility Technical Architect
> AT&T Consumer Digital Experience
>
> o: 919-241-4083
> e: = EMAIL ADDRESS REMOVED =
>
> 4625 Creekstone Dr | Durham, NC 27703
>
> This email and any files transmitted with it are AT&T property, are
> confidential, and are intended solely for the use of the individual or
> entity to whom this email is addressed. If you are not one of the named
> recipient(s) or otherwise have reason to believe that you have received this
> message in error, please notify the sender and delete this message
> immediately from your computer. Any other uses, retention, dissemination,
> forwarding, printing, or copying of this email is strictly prohibited.
>
>
>
From: GILLENWATER, ZOE M
Date: Mon, Nov 19 2012 11:17AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Hi Sailesh,
But if you were using visible labels you *would* hear them read in browse mode, so using title text that is not read in browse mode is a different experience from the norm, and it seems to me a worse one.
For instance, if I had three fields for month, day, and year of a birthday, and each had a visible label in front of it, I would hear in browse mode something like "Birthday. Month, edit, blank. Day, edit, blank. Year, edit, blank." It's clear to me what those fields are, which allows me to know whether or not I want to enter forms mode and fill out this form.
But if I got rid of the visible labels and just used title text, some screen reader users would now hear in browse mode something like "Birthday. Edit, blank. Edit, blank. Edit, blank." You don't think this is a worse experience than the former? Sure, I can guess at what each of those fields is, and have my guess validated when I enter forms mode and finally hear the title text, but why should the user have to do this? Why not just provide hidden labels so they hear the exact same thing they would hear had visible labels been used instead?
I still don't see what advantage using title has over a hidden label, apart from:
-- it's slightly less bytes of HTML
-- it's slightly easier to implement (if you don't already have a hiding CSS class in your CSS, but if you do, creating hidden labels is pretty much just as easy)
-- it will show on hover as a tooltip to mouse-using sighted users (which could be seen as a disadvantage by some people in some situations)
What am I missing? Are these the only reasons why you find title to be superior?
Thanks,
Zoe
Zoe Gillenwater
Web Accessibility Technical Architect
AT&T Consumer Digital Experience
o: 919-241-4083
e: = EMAIL ADDRESS REMOVED =
4625 Creekstone Dr | Durham, NC 27703
This email and any files transmitted with it are AT&T property, are confidential, and are intended solely for the use of the individual or entity to whom this email is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender and delete this message immediately from your computer. Any other uses, retention, dissemination, forwarding, printing, or copying of this email is strictly prohibited.
From: Bryan Garaventa
Date: Mon, Nov 19 2012 11:50AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
When you refer to browse mode, do you mean when using the arrow keys to
navigate up and down the page in the Virtual Buffer?
When title attributes are included on form fields, I'm hearing these
announced correctly as form field labels in addition to tabbing using
interactive mode.
This can be broken if a title attribute is included on a form field that
also includes an explicit label element with matching for and ID attributes
in some ATs though.
----- Original Message -----
From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
Sent: Monday, November 19, 2012 10:17 AM
Subject: Re: [WebAIM] Is use of <label> and title redundant?
Hi Sailesh,
But if you were using visible labels you *would* hear them read in browse
mode, so using title text that is not read in browse mode is a different
experience from the norm, and it seems to me a worse one.
For instance, if I had three fields for month, day, and year of a birthday,
and each had a visible label in front of it, I would hear in browse mode
something like "Birthday. Month, edit, blank. Day, edit, blank. Year, edit,
blank." It's clear to me what those fields are, which allows me to know
whether or not I want to enter forms mode and fill out this form.
But if I got rid of the visible labels and just used title text, some screen
reader users would now hear in browse mode something like "Birthday. Edit,
blank. Edit, blank. Edit, blank." You don't think this is a worse experience
than the former? Sure, I can guess at what each of those fields is, and have
my guess validated when I enter forms mode and finally hear the title text,
but why should the user have to do this? Why not just provide hidden labels
so they hear the exact same thing they would hear had visible labels been
used instead?
I still don't see what advantage using title has over a hidden label, apart
from:
-- it's slightly less bytes of HTML
-- it's slightly easier to implement (if you don't already have a hiding CSS
class in your CSS, but if you do, creating hidden labels is pretty much just
as easy)
-- it will show on hover as a tooltip to mouse-using sighted users (which
could be seen as a disadvantage by some people in some situations)
What am I missing? Are these the only reasons why you find title to be
superior?
Thanks,
Zoe
Zoe Gillenwater
Web Accessibility Technical Architect
AT&T Consumer Digital Experience
o: 919-241-4083
e: = EMAIL ADDRESS REMOVED =
4625 Creekstone Dr | Durham, NC 27703
This email and any files transmitted with it are AT&T property, are
confidential, and are intended solely for the use of the individual or
entity to whom this email is addressed. If you are not one of the named
recipient(s) or otherwise have reason to believe that you have received this
message in error, please notify the sender and delete this message
immediately from your computer. Any other uses, retention, dissemination,
forwarding, printing, or copying of this email is strictly prohibited.
From: Sailesh Panchang
Date: Mon, Nov 19 2012 3:15PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Zoe / Bryan,
JAWS does read title in Firefox and IE when arrowing down the page
(without activating forms mode).
NVDA does not read the title unless one is in forms mode.
Using title is simpler as I have maintained all along as compared to
off-screen label in the situations discussed.
Failing to markup visible labels as labels and use title instead will
fail SC 3.3.2.
It is not alright to simply use title there.
Sailesh
On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> When you refer to browse mode, do you mean when using the arrow keys to
> navigate up and down the page in the Virtual Buffer?
>
> When title attributes are included on form fields, I'm hearing these
> announced correctly as form field labels in addition to tabbing using
> interactive mode.
>
> This can be broken if a title attribute is included on a form field that
> also includes an explicit label element with matching for and ID attributes
>
> in some ATs though.
>
>
>
> ----- Original Message -----
> From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
> To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
> Sent: Monday, November 19, 2012 10:17 AM
> Subject: Re: [WebAIM] Is use of <label> and title redundant?
>
>
> Hi Sailesh,
>
> But if you were using visible labels you *would* hear them read in browse
> mode, so using title text that is not read in browse mode is a different
> experience from the norm, and it seems to me a worse one.
>
> For instance, if I had three fields for month, day, and year of a birthday,
>
> and each had a visible label in front of it, I would hear in browse mode
> something like "Birthday. Month, edit, blank. Day, edit, blank. Year, edit,
>
> blank." It's clear to me what those fields are, which allows me to know
> whether or not I want to enter forms mode and fill out this form.
>
> But if I got rid of the visible labels and just used title text, some screen
>
> reader users would now hear in browse mode something like "Birthday. Edit,
> blank. Edit, blank. Edit, blank." You don't think this is a worse experience
>
> than the former? Sure, I can guess at what each of those fields is, and have
>
> my guess validated when I enter forms mode and finally hear the title text,
>
> but why should the user have to do this? Why not just provide hidden labels
>
> so they hear the exact same thing they would hear had visible labels been
> used instead?
>
> I still don't see what advantage using title has over a hidden label, apart
>
> from:
> -- it's slightly less bytes of HTML
> -- it's slightly easier to implement (if you don't already have a hiding CSS
>
> class in your CSS, but if you do, creating hidden labels is pretty much just
>
> as easy)
> -- it will show on hover as a tooltip to mouse-using sighted users (which
> could be seen as a disadvantage by some people in some situations)
>
> What am I missing? Are these the only reasons why you find title to be
> superior?
>
> Thanks,
> Zoe
>
>
>
> Zoe Gillenwater
> Web Accessibility Technical Architect
> AT&T Consumer Digital Experience
>
> o: 919-241-4083
> e: = EMAIL ADDRESS REMOVED =
>
> 4625 Creekstone Dr | Durham, NC 27703
>
> This email and any files transmitted with it are AT&T property, are
> confidential, and are intended solely for the use of the individual or
> entity to whom this email is addressed. If you are not one of the named
> recipient(s) or otherwise have reason to believe that you have received this
>
> message in error, please notify the sender and delete this message
> immediately from your computer. Any other uses, retention, dissemination,
> forwarding, printing, or copying of this email is strictly prohibited.
>
>
>
From: Bryan Garaventa
Date: Mon, Nov 19 2012 3:24PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Wouldn't this be a bug in NVDA then? It doesn't make sense that, if an
explicit label is present, that it would not be announced when arrowing.
----- Original Message -----
From: "Sailesh Panchang" < = EMAIL ADDRESS REMOVED = >
To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
Sent: Monday, November 19, 2012 2:15 PM
Subject: Re: [WebAIM] Is use of <label> and title redundant?
Zoe / Bryan,
JAWS does read title in Firefox and IE when arrowing down the page
(without activating forms mode).
NVDA does not read the title unless one is in forms mode.
Using title is simpler as I have maintained all along as compared to
off-screen label in the situations discussed.
Failing to markup visible labels as labels and use title instead will
fail SC 3.3.2.
It is not alright to simply use title there.
Sailesh
On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> When you refer to browse mode, do you mean when using the arrow keys to
> navigate up and down the page in the Virtual Buffer?
>
> When title attributes are included on form fields, I'm hearing these
> announced correctly as form field labels in addition to tabbing using
> interactive mode.
>
> This can be broken if a title attribute is included on a form field that
> also includes an explicit label element with matching for and ID
> attributes
>
> in some ATs though.
>
>
>
> ----- Original Message -----
> From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
> To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
> Sent: Monday, November 19, 2012 10:17 AM
> Subject: Re: [WebAIM] Is use of <label> and title redundant?
>
>
> Hi Sailesh,
>
> But if you were using visible labels you *would* hear them read in browse
> mode, so using title text that is not read in browse mode is a different
> experience from the norm, and it seems to me a worse one.
>
> For instance, if I had three fields for month, day, and year of a
> birthday,
>
> and each had a visible label in front of it, I would hear in browse mode
> something like "Birthday. Month, edit, blank. Day, edit, blank. Year,
> edit,
>
> blank." It's clear to me what those fields are, which allows me to know
> whether or not I want to enter forms mode and fill out this form.
>
> But if I got rid of the visible labels and just used title text, some
> screen
>
> reader users would now hear in browse mode something like "Birthday. Edit,
> blank. Edit, blank. Edit, blank." You don't think this is a worse
> experience
>
> than the former? Sure, I can guess at what each of those fields is, and
> have
>
> my guess validated when I enter forms mode and finally hear the title
> text,
>
> but why should the user have to do this? Why not just provide hidden
> labels
>
> so they hear the exact same thing they would hear had visible labels been
> used instead?
>
> I still don't see what advantage using title has over a hidden label,
> apart
>
> from:
> -- it's slightly less bytes of HTML
> -- it's slightly easier to implement (if you don't already have a hiding
> CSS
>
> class in your CSS, but if you do, creating hidden labels is pretty much
> just
>
> as easy)
> -- it will show on hover as a tooltip to mouse-using sighted users (which
> could be seen as a disadvantage by some people in some situations)
>
> What am I missing? Are these the only reasons why you find title to be
> superior?
>
> Thanks,
> Zoe
>
>
>
> Zoe Gillenwater
> Web Accessibility Technical Architect
> AT&T Consumer Digital Experience
>
> o: 919-241-4083
> e: = EMAIL ADDRESS REMOVED =
>
> 4625 Creekstone Dr | Durham, NC 27703
>
> This email and any files transmitted with it are AT&T property, are
> confidential, and are intended solely for the use of the individual or
> entity to whom this email is addressed. If you are not one of the named
> recipient(s) or otherwise have reason to believe that you have received
> this
>
> message in error, please notify the sender and delete this message
> immediately from your computer. Any other uses, retention, dissemination,
> forwarding, printing, or copying of this email is strictly prohibited.
>
>
>
From: James Nurthen
Date: Mon, Nov 19 2012 11:36PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Sailesh,
I'm not sure how this fails 3.3.2. Can you explain how the following fails
3.3.2?
<div>First Name</div><input type="text" title="First Name">
I understand it is better practice to specify the label using for and id,
but I'm not seeing how this example actually fails 3.3.2
regards,
James
On Mon, Nov 19, 2012 at 2:15 PM, Sailesh Panchang <
= EMAIL ADDRESS REMOVED = > wrote:
> Zoe / Bryan,
>
> JAWS does read title in Firefox and IE when arrowing down the page
> (without activating forms mode).
> NVDA does not read the title unless one is in forms mode.
> Using title is simpler as I have maintained all along as compared to
> off-screen label in the situations discussed.
> Failing to markup visible labels as labels and use title instead will
> fail SC 3.3.2.
> It is not alright to simply use title there.
> Sailesh
>
>
> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> > When you refer to browse mode, do you mean when using the arrow keys to
> > navigate up and down the page in the Virtual Buffer?
> >
> > When title attributes are included on form fields, I'm hearing these
> > announced correctly as form field labels in addition to tabbing using
> > interactive mode.
> >
> > This can be broken if a title attribute is included on a form field that
> > also includes an explicit label element with matching for and ID
> attributes
> >
> > in some ATs though.
> >
> >
> >
> > ----- Original Message -----
> > From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
> > To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
> > Sent: Monday, November 19, 2012 10:17 AM
> > Subject: Re: [WebAIM] Is use of <label> and title redundant?
> >
> >
> > Hi Sailesh,
> >
> > But if you were using visible labels you *would* hear them read in browse
> > mode, so using title text that is not read in browse mode is a different
> > experience from the norm, and it seems to me a worse one.
> >
> > For instance, if I had three fields for month, day, and year of a
> birthday,
> >
> > and each had a visible label in front of it, I would hear in browse mode
> > something like "Birthday. Month, edit, blank. Day, edit, blank. Year,
> edit,
> >
> > blank." It's clear to me what those fields are, which allows me to know
> > whether or not I want to enter forms mode and fill out this form.
> >
> > But if I got rid of the visible labels and just used title text, some
> screen
> >
> > reader users would now hear in browse mode something like "Birthday.
> Edit,
> > blank. Edit, blank. Edit, blank." You don't think this is a worse
> experience
> >
> > than the former? Sure, I can guess at what each of those fields is, and
> have
> >
> > my guess validated when I enter forms mode and finally hear the title
> text,
> >
> > but why should the user have to do this? Why not just provide hidden
> labels
> >
> > so they hear the exact same thing they would hear had visible labels been
> > used instead?
> >
> > I still don't see what advantage using title has over a hidden label,
> apart
> >
> > from:
> > -- it's slightly less bytes of HTML
> > -- it's slightly easier to implement (if you don't already have a hiding
> CSS
> >
> > class in your CSS, but if you do, creating hidden labels is pretty much
> just
> >
> > as easy)
> > -- it will show on hover as a tooltip to mouse-using sighted users (which
> > could be seen as a disadvantage by some people in some situations)
> >
> > What am I missing? Are these the only reasons why you find title to be
> > superior?
> >
> > Thanks,
> > Zoe
> >
> >
> >
> > Zoe Gillenwater
> > Web Accessibility Technical Architect
> > AT&T Consumer Digital Experience
> >
> > o: 919-241-4083
> > e: = EMAIL ADDRESS REMOVED =
> >
> > 4625 Creekstone Dr | Durham, NC 27703
> >
> > This email and any files transmitted with it are AT&T property, are
> > confidential, and are intended solely for the use of the individual or
> > entity to whom this email is addressed. If you are not one of the named
> > recipient(s) or otherwise have reason to believe that you have received
> this
> >
> > message in error, please notify the sender and delete this message
> > immediately from your computer. Any other uses, retention, dissemination,
> > forwarding, printing, or copying of this email is strictly prohibited.
> >
> >
> >
From: GILLENWATER, ZOE M
Date: Tue, Nov 20 2012 7:11AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Sailesh,
OK, so it sounds like the only advantage you're attributing to the title attribute is its ease of use, which I do acknowledge. But as I said, with a CSS class to hide content already present in your style sheet, using a hidden label is just as easy, in my opinion. Because of that, and the issue with titles not being spoken in normal reading mode in NVDA, I remain convinced that hidden labels are a better first choice, with titles being a valid "plan B" if necessary.
I agree that if there is visible text on the page that could be used to label the field, it should be made a real <label>. Those aren't the type of situations we've been discussing.
Thanks for your insight! I really do appreciate it.
Zoe
Zoe Gillenwater
Web Accessibility Technical Architect
AT&T Consumer Digital Experience
o: 919-241-4083
e: = EMAIL ADDRESS REMOVED =
4625 Creekstone Dr | Durham, NC 27703
This email and any files transmitted with it are AT&T property, are confidential, and are intended solely for the use of the individual or entity to whom this email is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender and delete this message immediately from your computer. Any other uses, retention, dissemination, forwarding, printing, or copying of this email is strictly prohibited.
From: Sailesh Panchang
Date: Tue, Nov 20 2012 9:43AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Hello James,
Technique H65 says use title if UI does not support visible label text.
In your example there is no apparent reason for not using label
element and for-id association. Title should not be used there.
Refer to: Description for H44:[1]
"However, for Success Criterion 3.3.2, the label element must be
visible since it provides assistance to all users who need help
understanding the purpose of the field. "
There are other user groups besides screen reader users who are
benefited by label association.
Sailesh
[1] http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/H44
On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
> Sailesh,
> I'm not sure how this fails 3.3.2. Can you explain how the following fails
> 3.3.2?
>
> <div>First Name</div><input type="text" title="First Name">
>
> I understand it is better practice to specify the label using for and id,
> but I'm not seeing how this example actually fails 3.3.2
>
> regards,
> James
>
>
> On Mon, Nov 19, 2012 at 2:15 PM, Sailesh Panchang <
> = EMAIL ADDRESS REMOVED = > wrote:
>
>> Zoe / Bryan,
>>
>> JAWS does read title in Firefox and IE when arrowing down the page
>> (without activating forms mode).
>> NVDA does not read the title unless one is in forms mode.
>> Using title is simpler as I have maintained all along as compared to
>> off-screen label in the situations discussed.
>> Failing to markup visible labels as labels and use title instead will
>> fail SC 3.3.2.
>> It is not alright to simply use title there.
>> Sailesh
>>
>>
>> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
>> > When you refer to browse mode, do you mean when using the arrow keys to
>> > navigate up and down the page in the Virtual Buffer?
>> >
>> > When title attributes are included on form fields, I'm hearing these
>> > announced correctly as form field labels in addition to tabbing using
>> > interactive mode.
>> >
>> > This can be broken if a title attribute is included on a form field
>> > that
>> > also includes an explicit label element with matching for and ID
>> attributes
>> >
>> > in some ATs though.
>> >
>> >
>> >
>> > ----- Original Message -----
>> > From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
>> > To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
>> > Sent: Monday, November 19, 2012 10:17 AM
>> > Subject: Re: [WebAIM] Is use of <label> and title redundant?
>> >
>> >
>> > Hi Sailesh,
>> >
>> > But if you were using visible labels you *would* hear them read in
>> > browse
>> > mode, so using title text that is not read in browse mode is a
>> > different
>> > experience from the norm, and it seems to me a worse one.
>> >
>> > For instance, if I had three fields for month, day, and year of a
>> birthday,
>> >
>> > and each had a visible label in front of it, I would hear in browse
>> > mode
>> > something like "Birthday. Month, edit, blank. Day, edit, blank. Year,
>> edit,
>> >
>> > blank." It's clear to me what those fields are, which allows me to know
>> > whether or not I want to enter forms mode and fill out this form.
>> >
>> > But if I got rid of the visible labels and just used title text, some
>> screen
>> >
>> > reader users would now hear in browse mode something like "Birthday.
>> Edit,
>> > blank. Edit, blank. Edit, blank." You don't think this is a worse
>> experience
>> >
>> > than the former? Sure, I can guess at what each of those fields is, and
>> have
>> >
>> > my guess validated when I enter forms mode and finally hear the title
>> text,
>> >
>> > but why should the user have to do this? Why not just provide hidden
>> labels
>> >
>> > so they hear the exact same thing they would hear had visible labels
>> > been
>> > used instead?
>> >
>> > I still don't see what advantage using title has over a hidden label,
>> apart
>> >
>> > from:
>> > -- it's slightly less bytes of HTML
>> > -- it's slightly easier to implement (if you don't already have a
>> > hiding
>> CSS
>> >
>> > class in your CSS, but if you do, creating hidden labels is pretty much
>> just
>> >
>> > as easy)
>> > -- it will show on hover as a tooltip to mouse-using sighted users
>> > (which
>> > could be seen as a disadvantage by some people in some situations)
>> >
>> > What am I missing? Are these the only reasons why you find title to be
>> > superior?
>> >
>> > Thanks,
>> > Zoe
>> >
>> >
>> >
>> > Zoe Gillenwater
>> > Web Accessibility Technical Architect
>> > AT&T Consumer Digital Experience
>> >
>> > o: 919-241-4083
>> > e: = EMAIL ADDRESS REMOVED =
>> >
>> > 4625 Creekstone Dr | Durham, NC 27703
>> >
>> > This email and any files transmitted with it are AT&T property, are
>> > confidential, and are intended solely for the use of the individual or
>> > entity to whom this email is addressed. If you are not one of the named
>> > recipient(s) or otherwise have reason to believe that you have received
>> this
>> >
>> > message in error, please notify the sender and delete this message
>> > immediately from your computer. Any other uses, retention,
>> > dissemination,
>> > forwarding, printing, or copying of this email is strictly prohibited.
>> >
>> >
>> >
From: Sailesh Panchang
Date: Tue, Nov 20 2012 9:47AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Bryan,
NVDA reads LABEL element text ... it does not read title when arrowing down.
Sailesh
On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> Wouldn't this be a bug in NVDA then? It doesn't make sense that, if an
> explicit label is present, that it would not be announced when arrowing.
>
> ----- Original Message -----
> From: "Sailesh Panchang" < = EMAIL ADDRESS REMOVED = >
> To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
> Sent: Monday, November 19, 2012 2:15 PM
> Subject: Re: [WebAIM] Is use of <label> and title redundant?
>
>
> Zoe / Bryan,
>
> JAWS does read title in Firefox and IE when arrowing down the page
> (without activating forms mode).
> NVDA does not read the title unless one is in forms mode.
> Using title is simpler as I have maintained all along as compared to
> off-screen label in the situations discussed.
> Failing to markup visible labels as labels and use title instead will
> fail SC 3.3.2.
> It is not alright to simply use title there.
> Sailesh
>
>
> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
>> When you refer to browse mode, do you mean when using the arrow keys to
>> navigate up and down the page in the Virtual Buffer?
>>
>> When title attributes are included on form fields, I'm hearing these
>> announced correctly as form field labels in addition to tabbing using
>> interactive mode.
>>
>> This can be broken if a title attribute is included on a form field that
>> also includes an explicit label element with matching for and ID
>> attributes
>>
>> in some ATs though.
>>
>>
>>
>> ----- Original Message -----
>> From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
>> To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
>> Sent: Monday, November 19, 2012 10:17 AM
>> Subject: Re: [WebAIM] Is use of <label> and title redundant?
>>
>>
>> Hi Sailesh,
>>
>> But if you were using visible labels you *would* hear them read in browse
>> mode, so using title text that is not read in browse mode is a different
>> experience from the norm, and it seems to me a worse one.
>>
>> For instance, if I had three fields for month, day, and year of a
>> birthday,
>>
>> and each had a visible label in front of it, I would hear in browse mode
>> something like "Birthday. Month, edit, blank. Day, edit, blank. Year,
>> edit,
>>
>> blank." It's clear to me what those fields are, which allows me to know
>> whether or not I want to enter forms mode and fill out this form.
>>
>> But if I got rid of the visible labels and just used title text, some
>> screen
>>
>> reader users would now hear in browse mode something like "Birthday.
>> Edit,
>> blank. Edit, blank. Edit, blank." You don't think this is a worse
>> experience
>>
>> than the former? Sure, I can guess at what each of those fields is, and
>> have
>>
>> my guess validated when I enter forms mode and finally hear the title
>> text,
>>
>> but why should the user have to do this? Why not just provide hidden
>> labels
>>
>> so they hear the exact same thing they would hear had visible labels been
>> used instead?
>>
>> I still don't see what advantage using title has over a hidden label,
>> apart
>>
>> from:
>> -- it's slightly less bytes of HTML
>> -- it's slightly easier to implement (if you don't already have a hiding
>> CSS
>>
>> class in your CSS, but if you do, creating hidden labels is pretty much
>> just
>>
>> as easy)
>> -- it will show on hover as a tooltip to mouse-using sighted users (which
>> could be seen as a disadvantage by some people in some situations)
>>
>> What am I missing? Are these the only reasons why you find title to be
>> superior?
>>
>> Thanks,
>> Zoe
>>
>>
>>
>> Zoe Gillenwater
>> Web Accessibility Technical Architect
>> AT&T Consumer Digital Experience
>>
>> o: 919-241-4083
>> e: = EMAIL ADDRESS REMOVED =
>>
>> 4625 Creekstone Dr | Durham, NC 27703
>>
>> This email and any files transmitted with it are AT&T property, are
>> confidential, and are intended solely for the use of the individual or
>> entity to whom this email is addressed. If you are not one of the named
>> recipient(s) or otherwise have reason to believe that you have received
>> this
>>
>> message in error, please notify the sender and delete this message
>> immediately from your computer. Any other uses, retention, dissemination,
>> forwarding, printing, or copying of this email is strictly prohibited.
>>
>>
>>
From: James Nurthen
Date: Tue, Nov 20 2012 10:41AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Sailesh,
There is nothing in WCAG which states that you have to use any of the
prescribed techniques. The important thing is to meet the Success Criteria.
3.3.2 states
"3.3.2 Labels or Instructions: Labels or instructions are provided when
content requires user input. (Level A)".
The Understanding document provides more
"The intent of this Success Criterion is to help users avoid making
mistakes when their input is required. To help avoid mistakes it is good
user interface design to provide simple instructions and cues for entering
information. Some users with disabilities may be more likely to make
mistakes than users without disabilities or recovery from mistakes may be
more difficult, making mistake avoidance an important strategy for users
with disabilities. People with disabilities rely on well documented forms
and procedures to interact with a page. Blind users need to know exactly
what information should be entered into form fields and what the available
choices are. Simple instructions visually connected to form controls can
assist users with cognitive disabilities or those accessing a page using a
screen magnifier."
In this simple example there is a label associated visually with the field.
There is also the title (which happens in this case to be the identical to
the visible label) which is programmatically associated with the field. In
my opinion this meets 3.3.2.
I am certainly not saying that this is the optimal design as label is
certainly better, but we shouldn't rule out techniques which actually work
for users due to the fact that they are not optimal. I'd much rather that
the developers of web sites and applications spent their limited time
fixing issues which actually cause a problem rather than fix areas where
their code is not optimal but actually don't cause users any issues.
regards,
James
On Tue, Nov 20, 2012 at 8:43 AM, Sailesh Panchang <
= EMAIL ADDRESS REMOVED = > wrote:
> Hello James,
>
> Technique H65 says use title if UI does not support visible label text.
> In your example there is no apparent reason for not using label
> element and for-id association. Title should not be used there.
> Refer to: Description for H44:[1]
> "However, for Success Criterion 3.3.2, the label element must be
> visible since it provides assistance to all users who need help
> understanding the purpose of the field. "
> There are other user groups besides screen reader users who are
> benefited by label association.
>
> Sailesh
>
> [1] http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/H44
> On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
> > Sailesh,
> > I'm not sure how this fails 3.3.2. Can you explain how the following
> fails
> > 3.3.2?
> >
> > <div>First Name</div><input type="text" title="First Name">
> >
> > I understand it is better practice to specify the label using for and id,
> > but I'm not seeing how this example actually fails 3.3.2
> >
> > regards,
> > James
> >
> >
> > On Mon, Nov 19, 2012 at 2:15 PM, Sailesh Panchang <
> > = EMAIL ADDRESS REMOVED = > wrote:
> >
> >> Zoe / Bryan,
> >>
> >> JAWS does read title in Firefox and IE when arrowing down the page
> >> (without activating forms mode).
> >> NVDA does not read the title unless one is in forms mode.
> >> Using title is simpler as I have maintained all along as compared to
> >> off-screen label in the situations discussed.
> >> Failing to markup visible labels as labels and use title instead will
> >> fail SC 3.3.2.
> >> It is not alright to simply use title there.
> >> Sailesh
> >>
> >>
> >> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> >> > When you refer to browse mode, do you mean when using the arrow keys
> to
> >> > navigate up and down the page in the Virtual Buffer?
> >> >
> >> > When title attributes are included on form fields, I'm hearing these
> >> > announced correctly as form field labels in addition to tabbing using
> >> > interactive mode.
> >> >
> >> > This can be broken if a title attribute is included on a form field
> >> > that
> >> > also includes an explicit label element with matching for and ID
> >> attributes
> >> >
> >> > in some ATs though.
> >> >
> >> >
> >> >
> >> > ----- Original Message -----
> >> > From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
> >> > To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
> >> > Sent: Monday, November 19, 2012 10:17 AM
> >> > Subject: Re: [WebAIM] Is use of <label> and title redundant?
> >> >
> >> >
> >> > Hi Sailesh,
> >> >
> >> > But if you were using visible labels you *would* hear them read in
> >> > browse
> >> > mode, so using title text that is not read in browse mode is a
> >> > different
> >> > experience from the norm, and it seems to me a worse one.
> >> >
> >> > For instance, if I had three fields for month, day, and year of a
> >> birthday,
> >> >
> >> > and each had a visible label in front of it, I would hear in browse
> >> > mode
> >> > something like "Birthday. Month, edit, blank. Day, edit, blank. Year,
> >> edit,
> >> >
> >> > blank." It's clear to me what those fields are, which allows me to
> know
> >> > whether or not I want to enter forms mode and fill out this form.
> >> >
> >> > But if I got rid of the visible labels and just used title text, some
> >> screen
> >> >
> >> > reader users would now hear in browse mode something like "Birthday.
> >> Edit,
> >> > blank. Edit, blank. Edit, blank." You don't think this is a worse
> >> experience
> >> >
> >> > than the former? Sure, I can guess at what each of those fields is,
> and
> >> have
> >> >
> >> > my guess validated when I enter forms mode and finally hear the title
> >> text,
> >> >
> >> > but why should the user have to do this? Why not just provide hidden
> >> labels
> >> >
> >> > so they hear the exact same thing they would hear had visible labels
> >> > been
> >> > used instead?
> >> >
> >> > I still don't see what advantage using title has over a hidden label,
> >> apart
> >> >
> >> > from:
> >> > -- it's slightly less bytes of HTML
> >> > -- it's slightly easier to implement (if you don't already have a
> >> > hiding
> >> CSS
> >> >
> >> > class in your CSS, but if you do, creating hidden labels is pretty
> much
> >> just
> >> >
> >> > as easy)
> >> > -- it will show on hover as a tooltip to mouse-using sighted users
> >> > (which
> >> > could be seen as a disadvantage by some people in some situations)
> >> >
> >> > What am I missing? Are these the only reasons why you find title to be
> >> > superior?
> >> >
> >> > Thanks,
> >> > Zoe
> >> >
> >> >
> >> >
> >> > Zoe Gillenwater
> >> > Web Accessibility Technical Architect
> >> > AT&T Consumer Digital Experience
> >> >
> >> > o: 919-241-4083
> >> > e: = EMAIL ADDRESS REMOVED =
> >> >
> >> > 4625 Creekstone Dr | Durham, NC 27703
> >> >
> >> > This email and any files transmitted with it are AT&T property, are
> >> > confidential, and are intended solely for the use of the individual or
> >> > entity to whom this email is addressed. If you are not one of the
> named
> >> > recipient(s) or otherwise have reason to believe that you have
> received
> >> this
> >> >
> >> > message in error, please notify the sender and delete this message
> >> > immediately from your computer. Any other uses, retention,
> >> > dissemination,
> >> > forwarding, printing, or copying of this email is strictly prohibited.
> >> >
> >> >
> >> >
From: Sailesh Panchang
Date: Tue, Nov 20 2012 11:48AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
James,
>>my opinion this meets 3.3.2.
And I pointed out documentation that does not support that view.
Certainly techniques and understanding docs are not normative and are
guidance docs.
The 'How to meet' doc suggests what is appropriate in different circs.
and in fact is a guide to how to meet the SC.
Label text that is not marked up as a label is not 'associated' with
the control.
I never said title does not work for screen reader / magnifier users.
Title is simply not the recommended technique in the circumstance.
Label helps other user groups as I pointed out earlier. If you wish to
ignore them, it is your prerogative.
Sailesh Panchang
On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
> Sailesh,
>
> There is nothing in WCAG which states that you have to use any of the
> prescribed techniques. The important thing is to meet the Success Criteria.
> 3.3.2 states
> "3.3.2 Labels or Instructions: Labels or instructions are provided when
> content requires user input. (Level A)".
> The Understanding document provides more
> "The intent of this Success Criterion is to help users avoid making
> mistakes when their input is required. To help avoid mistakes it is good
> user interface design to provide simple instructions and cues for entering
> information. Some users with disabilities may be more likely to make
> mistakes than users without disabilities or recovery from mistakes may be
> more difficult, making mistake avoidance an important strategy for users
> with disabilities. People with disabilities rely on well documented forms
> and procedures to interact with a page. Blind users need to know exactly
> what information should be entered into form fields and what the available
> choices are. Simple instructions visually connected to form controls can
> assist users with cognitive disabilities or those accessing a page using a
> screen magnifier."
>
> In this simple example there is a label associated visually with the field.
> There is also the title (which happens in this case to be the identical to
> the visible label) which is programmatically associated with the field. In
> my opinion this meets 3.3.2.
>
> I am certainly not saying that this is the optimal design as label is
> certainly better, but we shouldn't rule out techniques which actually work
> for users due to the fact that they are not optimal. I'd much rather that
> the developers of web sites and applications spent their limited time
> fixing issues which actually cause a problem rather than fix areas where
> their code is not optimal but actually don't cause users any issues.
>
> regards,
> James
>
>
> On Tue, Nov 20, 2012 at 8:43 AM, Sailesh Panchang <
> = EMAIL ADDRESS REMOVED = > wrote:
>
>> Hello James,
>>
>> Technique H65 says use title if UI does not support visible label text.
>> In your example there is no apparent reason for not using label
>> element and for-id association. Title should not be used there.
>> Refer to: Description for H44:[1]
>> "However, for Success Criterion 3.3.2, the label element must be
>> visible since it provides assistance to all users who need help
>> understanding the purpose of the field. "
>> There are other user groups besides screen reader users who are
>> benefited by label association.
>>
>> Sailesh
>>
>> [1] http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/H44
>> On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
>> > Sailesh,
>> > I'm not sure how this fails 3.3.2. Can you explain how the following
>> fails
>> > 3.3.2?
>> >
>> > <div>First Name</div><input type="text" title="First Name">
>> >
>> > I understand it is better practice to specify the label using for and
>> > id,
>> > but I'm not seeing how this example actually fails 3.3.2
>> >
>> > regards,
>> > James
>> >
>> >
>> > On Mon, Nov 19, 2012 at 2:15 PM, Sailesh Panchang <
>> > = EMAIL ADDRESS REMOVED = > wrote:
>> >
>> >> Zoe / Bryan,
>> >>
>> >> JAWS does read title in Firefox and IE when arrowing down the page
>> >> (without activating forms mode).
>> >> NVDA does not read the title unless one is in forms mode.
>> >> Using title is simpler as I have maintained all along as compared to
>> >> off-screen label in the situations discussed.
>> >> Failing to markup visible labels as labels and use title instead will
>> >> fail SC 3.3.2.
>> >> It is not alright to simply use title there.
>> >> Sailesh
>> >>
>> >>
>> >> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
>> >> > When you refer to browse mode, do you mean when using the arrow keys
>> to
>> >> > navigate up and down the page in the Virtual Buffer?
>> >> >
>> >> > When title attributes are included on form fields, I'm hearing these
>> >> > announced correctly as form field labels in addition to tabbing
>> >> > using
>> >> > interactive mode.
>> >> >
>> >> > This can be broken if a title attribute is included on a form field
>> >> > that
>> >> > also includes an explicit label element with matching for and ID
>> >> attributes
>> >> >
>> >> > in some ATs though.
>> >> >
>> >> >
>> >> >
>> >> > ----- Original Message -----
>> >> > From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
>> >> > To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
>> >> > Sent: Monday, November 19, 2012 10:17 AM
>> >> > Subject: Re: [WebAIM] Is use of <label> and title redundant?
>> >> >
>> >> >
>> >> > Hi Sailesh,
>> >> >
>> >> > But if you were using visible labels you *would* hear them read in
>> >> > browse
>> >> > mode, so using title text that is not read in browse mode is a
>> >> > different
>> >> > experience from the norm, and it seems to me a worse one.
>> >> >
>> >> > For instance, if I had three fields for month, day, and year of a
>> >> birthday,
>> >> >
>> >> > and each had a visible label in front of it, I would hear in browse
>> >> > mode
>> >> > something like "Birthday. Month, edit, blank. Day, edit, blank.
>> >> > Year,
>> >> edit,
>> >> >
>> >> > blank." It's clear to me what those fields are, which allows me to
>> know
>> >> > whether or not I want to enter forms mode and fill out this form.
>> >> >
>> >> > But if I got rid of the visible labels and just used title text,
>> >> > some
>> >> screen
>> >> >
>> >> > reader users would now hear in browse mode something like "Birthday.
>> >> Edit,
>> >> > blank. Edit, blank. Edit, blank." You don't think this is a worse
>> >> experience
>> >> >
>> >> > than the former? Sure, I can guess at what each of those fields is,
>> and
>> >> have
>> >> >
>> >> > my guess validated when I enter forms mode and finally hear the
>> >> > title
>> >> text,
>> >> >
>> >> > but why should the user have to do this? Why not just provide hidden
>> >> labels
>> >> >
>> >> > so they hear the exact same thing they would hear had visible labels
>> >> > been
>> >> > used instead?
>> >> >
>> >> > I still don't see what advantage using title has over a hidden
>> >> > label,
>> >> apart
>> >> >
>> >> > from:
>> >> > -- it's slightly less bytes of HTML
>> >> > -- it's slightly easier to implement (if you don't already have a
>> >> > hiding
>> >> CSS
>> >> >
>> >> > class in your CSS, but if you do, creating hidden labels is pretty
>> much
>> >> just
>> >> >
>> >> > as easy)
>> >> > -- it will show on hover as a tooltip to mouse-using sighted users
>> >> > (which
>> >> > could be seen as a disadvantage by some people in some situations)
>> >> >
>> >> > What am I missing? Are these the only reasons why you find title to
>> >> > be
>> >> > superior?
>> >> >
>> >> > Thanks,
>> >> > Zoe
>> >> >
>> >> >
>> >> >
>> >> > Zoe Gillenwater
>> >> > Web Accessibility Technical Architect
>> >> > AT&T Consumer Digital Experience
>> >> >
>> >> > o: 919-241-4083
>> >> > e: = EMAIL ADDRESS REMOVED =
>> >> >
>> >> > 4625 Creekstone Dr | Durham, NC 27703
>> >> >
>> >> > This email and any files transmitted with it are AT&T property, are
>> >> > confidential, and are intended solely for the use of the individual
>> >> > or
>> >> > entity to whom this email is addressed. If you are not one of the
>> named
>> >> > recipient(s) or otherwise have reason to believe that you have
>> received
>> >> this
>> >> >
>> >> > message in error, please notify the sender and delete this message
>> >> > immediately from your computer. Any other uses, retention,
>> >> > dissemination,
>> >> > forwarding, printing, or copying of this email is strictly
>> >> > prohibited.
>> >> >
>> >> >
>> >> >
From: Patrick H. Lauke
Date: Tue, Nov 20 2012 11:58AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
On 20/11/2012 18:48, Sailesh Panchang wrote:
> James,
>>> my opinion this meets 3.3.2.
> And I pointed out documentation that does not support that view.
> Certainly techniques and understanding docs are not normative and are
> guidance docs.
> The 'How to meet' doc suggests what is appropriate in different circs.
> and in fact is a guide to how to meet the SC.
And it, as well, points out that "Note that all techniques are
informative". the "How to Meet" simply collates the SCs and the current
list of techniques.
> Label text that is not marked up as a label is not 'associated' with
> the control.
> I never said title does not work for screen reader / magnifier users.
> Title is simply not the recommended technique in the circumstance.
It's still one of the sufficient techniques listed
http://www.w3.org/WAI/WCAG20/quickref/#qr-minimize-error-cues
"H65: Using the title attribute to identify form controls when the label
element cannot be used "
Though yes, it's stated at the end of that part
"The techniques at the end of the above list should be considered "last
resort" and only used when the other techniques cannot be applied to the
page. The earlier techniques are preferred because they increase
accessibility to a wider user group."
P
--
Patrick H. Lauke
From: Sailesh Panchang
Date: Tue, Nov 20 2012 12:08PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Patrick,
> Though yes, it's stated at the end of that part
Please refer to the description for H65.
And I can only reiterate:
Title is simply not the recommended technique in the circumstance.
Not marking up visible text as a label and inserting a title attribute
is a failure of SC 3.3.2.
Label helps other user groups as I pointed out earlier. If you wish to
ignore them, it is your prerogative.
Sailesh Panchang
On 11/20/12, Patrick H. Lauke < = EMAIL ADDRESS REMOVED = > wrote:
> On 20/11/2012 18:48, Sailesh Panchang wrote:
>> James,
>>>> my opinion this meets 3.3.2.
>> And I pointed out documentation that does not support that view.
>> Certainly techniques and understanding docs are not normative and are
>> guidance docs.
>> The 'How to meet' doc suggests what is appropriate in different circs.
>> and in fact is a guide to how to meet the SC.
>
> And it, as well, points out that "Note that all techniques are
> informative". the "How to Meet" simply collates the SCs and the current
> list of techniques.
>
>> Label text that is not marked up as a label is not 'associated' with
>> the control.
>> I never said title does not work for screen reader / magnifier users.
>> Title is simply not the recommended technique in the circumstance.
>
> It's still one of the sufficient techniques listed
> http://www.w3.org/WAI/WCAG20/quickref/#qr-minimize-error-cues
>
> "H65: Using the title attribute to identify form controls when the label
> element cannot be used "
>
> Though yes, it's stated at the end of that part
>
> "The techniques at the end of the above list should be considered "last
> resort" and only used when the other techniques cannot be applied to the
> page. The earlier techniques are preferred because they increase
> accessibility to a wider user group."
>
> P
> --
> Patrick H. Lauke
> > re·dux (adj.): brought back; returned. used postpositively
> [latin : re-, re- + dux, leader; see duke.]
>
> www.splintered.co.uk | www.photographia.co.uk
> http://redux.deviantart.com | http://flickr.com/photos/redux/
> > twitter: @patrick_h_lauke | skype: patrick_h_lauke
> > > > >
From: Patrick H. Lauke
Date: Tue, Nov 20 2012 12:31PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
On 20/11/2012 19:08, Sailesh Panchang wrote:
> Patrick,
>> Though yes, it's stated at the end of that part
> Please refer to the description for H65.
> And I can only reiterate:
> Title is simply not the recommended technique in the circumstance.
Maybe splitting hairs here, but:
yes, the description for H65 states
"The objective of this technique is to use the title attribute to label
form controls when the visual design cannot accommodate the label (for
example, if there is no text on the screen that can be identified as a
label) or where it might be confusing to display a label"
But, does this imply that if there IS in fact text on the screen that
can be identified as a label, and the technique is used anyway, that
it's automatically a failure of 3.3.2? I'd argue that no, it doesn't.
It's certainly a stupid approach (if the text is already visible, just
mark it up as a label...simple), but even stupid code can still fulfill
the requirements of the SC.
If title is sufficient in situations when there is NO visible text, how
can you surmise that it's all of a sudden NOT sufficient when there IS text?
> Not marking up visible text as a label and inserting a title attribute
> is a failure of SC 3.3.2.
I think you're inferring this failure, unless I'm missing this being
explicitly mentioned anywhere.
P
--
Patrick H. Lauke
From: James Nurthen
Date: Tue, Nov 20 2012 12:33PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Sailesh,
The fact is that nothing in 3.3.2 requires programmatic association of
fields and labels. 3.3.2 deals with providing labels or instructions for
fields, not their association.
3.3.2 can be met by (for example) applying G131 (Providing Descriptive
Labels) and G162 (Positioning labels to maximize predictability of
relationships). so even the following:
<span>First Name</span><input type="text"> would meet 3.3.2
Programmatic association of fields and labels is covered by 1.3.1 in WCAG2.
regards,
James
On Tue, Nov 20, 2012 at 10:48 AM, Sailesh Panchang <
= EMAIL ADDRESS REMOVED = > wrote:
> James,
> >>my opinion this meets 3.3.2.
> And I pointed out documentation that does not support that view.
> Certainly techniques and understanding docs are not normative and are
> guidance docs.
> The 'How to meet' doc suggests what is appropriate in different circs.
> and in fact is a guide to how to meet the SC.
> Label text that is not marked up as a label is not 'associated' with
> the control.
> I never said title does not work for screen reader / magnifier users.
> Title is simply not the recommended technique in the circumstance.
> Label helps other user groups as I pointed out earlier. If you wish to
> ignore them, it is your prerogative.
> Sailesh Panchang
>
>
>
> On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
> > Sailesh,
> >
> > There is nothing in WCAG which states that you have to use any of the
> > prescribed techniques. The important thing is to meet the Success
> Criteria.
> > 3.3.2 states
> > "3.3.2 Labels or Instructions: Labels or instructions are provided when
> > content requires user input. (Level A)".
> > The Understanding document provides more
> > "The intent of this Success Criterion is to help users avoid making
> > mistakes when their input is required. To help avoid mistakes it is good
> > user interface design to provide simple instructions and cues for
> entering
> > information. Some users with disabilities may be more likely to make
> > mistakes than users without disabilities or recovery from mistakes may be
> > more difficult, making mistake avoidance an important strategy for users
> > with disabilities. People with disabilities rely on well documented forms
> > and procedures to interact with a page. Blind users need to know exactly
> > what information should be entered into form fields and what the
> available
> > choices are. Simple instructions visually connected to form controls can
> > assist users with cognitive disabilities or those accessing a page using
> a
> > screen magnifier."
> >
> > In this simple example there is a label associated visually with the
> field.
> > There is also the title (which happens in this case to be the identical
> to
> > the visible label) which is programmatically associated with the field.
> In
> > my opinion this meets 3.3.2.
> >
> > I am certainly not saying that this is the optimal design as label is
> > certainly better, but we shouldn't rule out techniques which actually
> work
> > for users due to the fact that they are not optimal. I'd much rather that
> > the developers of web sites and applications spent their limited time
> > fixing issues which actually cause a problem rather than fix areas where
> > their code is not optimal but actually don't cause users any issues.
> >
> > regards,
> > James
> >
> >
> > On Tue, Nov 20, 2012 at 8:43 AM, Sailesh Panchang <
> > = EMAIL ADDRESS REMOVED = > wrote:
> >
> >> Hello James,
> >>
> >> Technique H65 says use title if UI does not support visible label text.
> >> In your example there is no apparent reason for not using label
> >> element and for-id association. Title should not be used there.
> >> Refer to: Description for H44:[1]
> >> "However, for Success Criterion 3.3.2, the label element must be
> >> visible since it provides assistance to all users who need help
> >> understanding the purpose of the field. "
> >> There are other user groups besides screen reader users who are
> >> benefited by label association.
> >>
> >> Sailesh
> >>
> >> [1] http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/H44
> >> On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
> >> > Sailesh,
> >> > I'm not sure how this fails 3.3.2. Can you explain how the following
> >> fails
> >> > 3.3.2?
> >> >
> >> > <div>First Name</div><input type="text" title="First Name">
> >> >
> >> > I understand it is better practice to specify the label using for and
> >> > id,
> >> > but I'm not seeing how this example actually fails 3.3.2
> >> >
> >> > regards,
> >> > James
> >> >
> >> >
> >> > On Mon, Nov 19, 2012 at 2:15 PM, Sailesh Panchang <
> >> > = EMAIL ADDRESS REMOVED = > wrote:
> >> >
> >> >> Zoe / Bryan,
> >> >>
> >> >> JAWS does read title in Firefox and IE when arrowing down the page
> >> >> (without activating forms mode).
> >> >> NVDA does not read the title unless one is in forms mode.
> >> >> Using title is simpler as I have maintained all along as compared to
> >> >> off-screen label in the situations discussed.
> >> >> Failing to markup visible labels as labels and use title instead
> will
> >> >> fail SC 3.3.2.
> >> >> It is not alright to simply use title there.
> >> >> Sailesh
> >> >>
> >> >>
> >> >> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> >> >> > When you refer to browse mode, do you mean when using the arrow
> keys
> >> to
> >> >> > navigate up and down the page in the Virtual Buffer?
> >> >> >
> >> >> > When title attributes are included on form fields, I'm hearing
> these
> >> >> > announced correctly as form field labels in addition to tabbing
> >> >> > using
> >> >> > interactive mode.
> >> >> >
> >> >> > This can be broken if a title attribute is included on a form field
> >> >> > that
> >> >> > also includes an explicit label element with matching for and ID
> >> >> attributes
> >> >> >
> >> >> > in some ATs though.
> >> >> >
> >> >> >
> >> >> >
> >> >> > ----- Original Message -----
> >> >> > From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
> >> >> > To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
> >> >> > Sent: Monday, November 19, 2012 10:17 AM
> >> >> > Subject: Re: [WebAIM] Is use of <label> and title redundant?
> >> >> >
> >> >> >
> >> >> > Hi Sailesh,
> >> >> >
> >> >> > But if you were using visible labels you *would* hear them read in
> >> >> > browse
> >> >> > mode, so using title text that is not read in browse mode is a
> >> >> > different
> >> >> > experience from the norm, and it seems to me a worse one.
> >> >> >
> >> >> > For instance, if I had three fields for month, day, and year of a
> >> >> birthday,
> >> >> >
> >> >> > and each had a visible label in front of it, I would hear in browse
> >> >> > mode
> >> >> > something like "Birthday. Month, edit, blank. Day, edit, blank.
> >> >> > Year,
> >> >> edit,
> >> >> >
> >> >> > blank." It's clear to me what those fields are, which allows me to
> >> know
> >> >> > whether or not I want to enter forms mode and fill out this form.
> >> >> >
> >> >> > But if I got rid of the visible labels and just used title text,
> >> >> > some
> >> >> screen
> >> >> >
> >> >> > reader users would now hear in browse mode something like
> "Birthday.
> >> >> Edit,
> >> >> > blank. Edit, blank. Edit, blank." You don't think this is a worse
> >> >> experience
> >> >> >
> >> >> > than the former? Sure, I can guess at what each of those fields is,
> >> and
> >> >> have
> >> >> >
> >> >> > my guess validated when I enter forms mode and finally hear the
> >> >> > title
> >> >> text,
> >> >> >
> >> >> > but why should the user have to do this? Why not just provide
> hidden
> >> >> labels
> >> >> >
> >> >> > so they hear the exact same thing they would hear had visible
> labels
> >> >> > been
> >> >> > used instead?
> >> >> >
> >> >> > I still don't see what advantage using title has over a hidden
> >> >> > label,
> >> >> apart
> >> >> >
> >> >> > from:
> >> >> > -- it's slightly less bytes of HTML
> >> >> > -- it's slightly easier to implement (if you don't already have a
> >> >> > hiding
> >> >> CSS
> >> >> >
> >> >> > class in your CSS, but if you do, creating hidden labels is pretty
> >> much
> >> >> just
> >> >> >
> >> >> > as easy)
> >> >> > -- it will show on hover as a tooltip to mouse-using sighted users
> >> >> > (which
> >> >> > could be seen as a disadvantage by some people in some situations)
> >> >> >
> >> >> > What am I missing? Are these the only reasons why you find title to
> >> >> > be
> >> >> > superior?
> >> >> >
> >> >> > Thanks,
> >> >> > Zoe
> >> >> >
> >> >> >
> >> >> >
> >> >> > Zoe Gillenwater
> >> >> > Web Accessibility Technical Architect
> >> >> > AT&T Consumer Digital Experience
> >> >> >
> >> >> > o: 919-241-4083
> >> >> > e: = EMAIL ADDRESS REMOVED =
> >> >> >
> >> >> > 4625 Creekstone Dr | Durham, NC 27703
> >> >> >
> >> >> > This email and any files transmitted with it are AT&T property, are
> >> >> > confidential, and are intended solely for the use of the individual
> >> >> > or
> >> >> > entity to whom this email is addressed. If you are not one of the
> >> named
> >> >> > recipient(s) or otherwise have reason to believe that you have
> >> received
> >> >> this
> >> >> >
> >> >> > message in error, please notify the sender and delete this message
> >> >> > immediately from your computer. Any other uses, retention,
> >> >> > dissemination,
> >> >> > forwarding, printing, or copying of this email is strictly
> >> >> > prohibited.
> >> >> >
> >> >> >
> >> >> >
From: Sailesh Panchang
Date: Tue, Nov 20 2012 12:55PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
James,
Sure, I know that.
And Patrick,
Please refer to the first email to James which has the reasoning. It
is the description of H44:
http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/H44
Thanks and regards,
Sailesh
On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
> Sailesh,
> The fact is that nothing in 3.3.2 requires programmatic association of
> fields and labels. 3.3.2 deals with providing labels or instructions for
> fields, not their association.
> 3.3.2 can be met by (for example) applying G131 (Providing Descriptive
> Labels) and G162 (Positioning labels to maximize predictability of
> relationships). so even the following:
> <span>First Name</span><input type="text"> would meet 3.3.2
>
> Programmatic association of fields and labels is covered by 1.3.1 in WCAG2.
>
> regards,
> James
>
>
> On Tue, Nov 20, 2012 at 10:48 AM, Sailesh Panchang <
> = EMAIL ADDRESS REMOVED = > wrote:
>
>> James,
>> >>my opinion this meets 3.3.2.
>> And I pointed out documentation that does not support that view.
>> Certainly techniques and understanding docs are not normative and are
>> guidance docs.
>> The 'How to meet' doc suggests what is appropriate in different circs.
>> and in fact is a guide to how to meet the SC.
>> Label text that is not marked up as a label is not 'associated' with
>> the control.
>> I never said title does not work for screen reader / magnifier users.
>> Title is simply not the recommended technique in the circumstance.
>> Label helps other user groups as I pointed out earlier. If you wish to
>> ignore them, it is your prerogative.
>> Sailesh Panchang
>>
>>
>>
>> On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
>> > Sailesh,
>> >
>> > There is nothing in WCAG which states that you have to use any of the
>> > prescribed techniques. The important thing is to meet the Success
>> Criteria.
>> > 3.3.2 states
>> > "3.3.2 Labels or Instructions: Labels or instructions are provided when
>> > content requires user input. (Level A)".
>> > The Understanding document provides more
>> > "The intent of this Success Criterion is to help users avoid making
>> > mistakes when their input is required. To help avoid mistakes it is
>> > good
>> > user interface design to provide simple instructions and cues for
>> entering
>> > information. Some users with disabilities may be more likely to make
>> > mistakes than users without disabilities or recovery from mistakes may
>> > be
>> > more difficult, making mistake avoidance an important strategy for
>> > users
>> > with disabilities. People with disabilities rely on well documented
>> > forms
>> > and procedures to interact with a page. Blind users need to know
>> > exactly
>> > what information should be entered into form fields and what the
>> available
>> > choices are. Simple instructions visually connected to form controls
>> > can
>> > assist users with cognitive disabilities or those accessing a page
>> > using
>> a
>> > screen magnifier."
>> >
>> > In this simple example there is a label associated visually with the
>> field.
>> > There is also the title (which happens in this case to be the identical
>> to
>> > the visible label) which is programmatically associated with the field.
>> In
>> > my opinion this meets 3.3.2.
>> >
>> > I am certainly not saying that this is the optimal design as label is
>> > certainly better, but we shouldn't rule out techniques which actually
>> work
>> > for users due to the fact that they are not optimal. I'd much rather
>> > that
>> > the developers of web sites and applications spent their limited time
>> > fixing issues which actually cause a problem rather than fix areas
>> > where
>> > their code is not optimal but actually don't cause users any issues.
>> >
>> > regards,
>> > James
>> >
>> >
>> > On Tue, Nov 20, 2012 at 8:43 AM, Sailesh Panchang <
>> > = EMAIL ADDRESS REMOVED = > wrote:
>> >
>> >> Hello James,
>> >>
>> >> Technique H65 says use title if UI does not support visible label
>> >> text.
>> >> In your example there is no apparent reason for not using label
>> >> element and for-id association. Title should not be used there.
>> >> Refer to: Description for H44:[1]
>> >> "However, for Success Criterion 3.3.2, the label element must be
>> >> visible since it provides assistance to all users who need help
>> >> understanding the purpose of the field. "
>> >> There are other user groups besides screen reader users who are
>> >> benefited by label association.
>> >>
>> >> Sailesh
>> >>
>> >> [1] http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/H44
>> >> On 11/20/12, James Nurthen < = EMAIL ADDRESS REMOVED = > wrote:
>> >> > Sailesh,
>> >> > I'm not sure how this fails 3.3.2. Can you explain how the following
>> >> fails
>> >> > 3.3.2?
>> >> >
>> >> > <div>First Name</div><input type="text" title="First Name">
>> >> >
>> >> > I understand it is better practice to specify the label using for
>> >> > and
>> >> > id,
>> >> > but I'm not seeing how this example actually fails 3.3.2
>> >> >
>> >> > regards,
>> >> > James
>> >> >
>> >> >
>> >> > On Mon, Nov 19, 2012 at 2:15 PM, Sailesh Panchang <
>> >> > = EMAIL ADDRESS REMOVED = > wrote:
>> >> >
>> >> >> Zoe / Bryan,
>> >> >>
>> >> >> JAWS does read title in Firefox and IE when arrowing down the page
>> >> >> (without activating forms mode).
>> >> >> NVDA does not read the title unless one is in forms mode.
>> >> >> Using title is simpler as I have maintained all along as compared
>> >> >> to
>> >> >> off-screen label in the situations discussed.
>> >> >> Failing to markup visible labels as labels and use title instead
>> will
>> >> >> fail SC 3.3.2.
>> >> >> It is not alright to simply use title there.
>> >> >> Sailesh
>> >> >>
>> >> >>
>> >> >> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
>> >> >> > When you refer to browse mode, do you mean when using the arrow
>> keys
>> >> to
>> >> >> > navigate up and down the page in the Virtual Buffer?
>> >> >> >
>> >> >> > When title attributes are included on form fields, I'm hearing
>> these
>> >> >> > announced correctly as form field labels in addition to tabbing
>> >> >> > using
>> >> >> > interactive mode.
>> >> >> >
>> >> >> > This can be broken if a title attribute is included on a form
>> >> >> > field
>> >> >> > that
>> >> >> > also includes an explicit label element with matching for and ID
>> >> >> attributes
>> >> >> >
>> >> >> > in some ATs though.
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> > ----- Original Message -----
>> >> >> > From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
>> >> >> > To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
>> >> >> > Sent: Monday, November 19, 2012 10:17 AM
>> >> >> > Subject: Re: [WebAIM] Is use of <label> and title redundant?
>> >> >> >
>> >> >> >
>> >> >> > Hi Sailesh,
>> >> >> >
>> >> >> > But if you were using visible labels you *would* hear them read
>> >> >> > in
>> >> >> > browse
>> >> >> > mode, so using title text that is not read in browse mode is a
>> >> >> > different
>> >> >> > experience from the norm, and it seems to me a worse one.
>> >> >> >
>> >> >> > For instance, if I had three fields for month, day, and year of a
>> >> >> birthday,
>> >> >> >
>> >> >> > and each had a visible label in front of it, I would hear in
>> >> >> > browse
>> >> >> > mode
>> >> >> > something like "Birthday. Month, edit, blank. Day, edit, blank.
>> >> >> > Year,
>> >> >> edit,
>> >> >> >
>> >> >> > blank." It's clear to me what those fields are, which allows me
>> >> >> > to
>> >> know
>> >> >> > whether or not I want to enter forms mode and fill out this form.
>> >> >> >
>> >> >> > But if I got rid of the visible labels and just used title text,
>> >> >> > some
>> >> >> screen
>> >> >> >
>> >> >> > reader users would now hear in browse mode something like
>> "Birthday.
>> >> >> Edit,
>> >> >> > blank. Edit, blank. Edit, blank." You don't think this is a worse
>> >> >> experience
>> >> >> >
>> >> >> > than the former? Sure, I can guess at what each of those fields
>> >> >> > is,
>> >> and
>> >> >> have
>> >> >> >
>> >> >> > my guess validated when I enter forms mode and finally hear the
>> >> >> > title
>> >> >> text,
>> >> >> >
>> >> >> > but why should the user have to do this? Why not just provide
>> hidden
>> >> >> labels
>> >> >> >
>> >> >> > so they hear the exact same thing they would hear had visible
>> labels
>> >> >> > been
>> >> >> > used instead?
>> >> >> >
>> >> >> > I still don't see what advantage using title has over a hidden
>> >> >> > label,
>> >> >> apart
>> >> >> >
>> >> >> > from:
>> >> >> > -- it's slightly less bytes of HTML
>> >> >> > -- it's slightly easier to implement (if you don't already have a
>> >> >> > hiding
>> >> >> CSS
>> >> >> >
>> >> >> > class in your CSS, but if you do, creating hidden labels is
>> >> >> > pretty
>> >> much
>> >> >> just
>> >> >> >
>> >> >> > as easy)
>> >> >> > -- it will show on hover as a tooltip to mouse-using sighted
>> >> >> > users
>> >> >> > (which
>> >> >> > could be seen as a disadvantage by some people in some
>> >> >> > situations)
>> >> >> >
>> >> >> > What am I missing? Are these the only reasons why you find title
>> >> >> > to
>> >> >> > be
>> >> >> > superior?
>> >> >> >
>> >> >> > Thanks,
>> >> >> > Zoe
>> >> >> >
>> >> >> >
>> >> >> >
>> >> >> > Zoe Gillenwater
>> >> >> > Web Accessibility Technical Architect
>> >> >> > AT&T Consumer Digital Experience
>> >> >> >
>> >> >> > o: 919-241-4083
>> >> >> > e: = EMAIL ADDRESS REMOVED =
>> >> >> >
>> >> >> > 4625 Creekstone Dr | Durham, NC 27703
>> >> >> >
>> >> >> > This email and any files transmitted with it are AT&T property,
>> >> >> > are
>> >> >> > confidential, and are intended solely for the use of the
>> >> >> > individual
>> >> >> > or
>> >> >> > entity to whom this email is addressed. If you are not one of the
>> >> named
>> >> >> > recipient(s) or otherwise have reason to believe that you have
>> >> received
>> >> >> this
>> >> >> >
>> >> >> > message in error, please notify the sender and delete this
>> >> >> > message
>> >> >> > immediately from your computer. Any other uses, retention,
>> >> >> > dissemination,
>> >> >> > forwarding, printing, or copying of this email is strictly
>> >> >> > prohibited.
>> >> >> >
>> >> >> >
>> >> >> >
From: Patrick H. Lauke
Date: Tue, Nov 20 2012 1:05PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
On 20/11/2012 19:55, Sailesh Panchang wrote:
> James,
> Sure, I know that.
> And Patrick,
> Please refer to the first email to James which has the reasoning. It
> is the description of H44:
> http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/H44
Yes, it says that, but unless I'm wildly off, this does not mean what
you seem to imply. Using James' last example code, with added title
attribute
<span>First Name</span> <input type="text" title="First Name">
This clearly fails 3.3.2 *if you only relied on H44 to satisfy 3.3.2*,
but you're not, because while H44 fails with regards to satisfying 3.3.2
(since there IS no label), the SC is actually satisfied by H65 and its
use of title.
P
--
Patrick H. Lauke
From: Patrick H. Lauke
Date: Tue, Nov 20 2012 1:07PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
On 20/11/2012 20:05, Patrick H. Lauke wrote:
> Yes, it says that
Sorry, to clarify, by "that" I meant the bit quoted by Sailesh in his
email to James, namely:
"However, for Success Criterion 3.3.2, the label element must be visible
since it provides assistance to all users who need help understanding
the purpose of the field."
http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/H44#H44-description
--
Patrick H. Lauke
From: Ryan E. Benson
Date: Tue, Nov 20 2012 6:21PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
I am reading this thread, and the following line by Sailesh confused me:
JAWS does read title in Firefox and IE when arrowing down the page
(without activating forms mode).
Are you using a version of JAWS before 10? If you aren't, I would say there
is a bug in JAWS or your install. According to
http://www.freedomscientific.com/training/surfs-up/Forms.htm, forms mode
should activate:
Auto forms mode was introduced in JAWS 10. Now when you press *TAB* or *
SHIFT+TAB* on a Web page and land in a form field or other form control,
forms mode is automatically on. It is confirmed with a sound that is
played. You no longer have to remember to switch forms mode on manually.
Forms mode is also turned on automatically if you use the *ARROW Keys* and
move into a form control such as an edit box. Again, a sound is played.
Additionally, if someone clicks into an edit box with a *MOUSE Click*,
forms mode comes on automatically.
--
Ryan E. Benson
On Mon, Nov 19, 2012 at 5:15 PM, Sailesh Panchang <
= EMAIL ADDRESS REMOVED = > wrote:
> Zoe / Bryan,
>
> JAWS does read title in Firefox and IE when arrowing down the page
> (without activating forms mode).
> NVDA does not read the title unless one is in forms mode.
> Using title is simpler as I have maintained all along as compared to
> off-screen label in the situations discussed.
> Failing to markup visible labels as labels and use title instead will
> fail SC 3.3.2.
> It is not alright to simply use title there.
> Sailesh
>
>
> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> > When you refer to browse mode, do you mean when using the arrow keys to
> > navigate up and down the page in the Virtual Buffer?
> >
> > When title attributes are included on form fields, I'm hearing these
> > announced correctly as form field labels in addition to tabbing using
> > interactive mode.
> >
> > This can be broken if a title attribute is included on a form field that
> > also includes an explicit label element with matching for and ID
> attributes
> >
> > in some ATs though.
> >
> >
> >
> > ----- Original Message -----
> > From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
> > To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
> > Sent: Monday, November 19, 2012 10:17 AM
> > Subject: Re: [WebAIM] Is use of <label> and title redundant?
> >
> >
> > Hi Sailesh,
> >
> > But if you were using visible labels you *would* hear them read in browse
> > mode, so using title text that is not read in browse mode is a different
> > experience from the norm, and it seems to me a worse one.
> >
> > For instance, if I had three fields for month, day, and year of a
> birthday,
> >
> > and each had a visible label in front of it, I would hear in browse mode
> > something like "Birthday. Month, edit, blank. Day, edit, blank. Year,
> edit,
> >
> > blank." It's clear to me what those fields are, which allows me to know
> > whether or not I want to enter forms mode and fill out this form.
> >
> > But if I got rid of the visible labels and just used title text, some
> screen
> >
> > reader users would now hear in browse mode something like "Birthday.
> Edit,
> > blank. Edit, blank. Edit, blank." You don't think this is a worse
> experience
> >
> > than the former? Sure, I can guess at what each of those fields is, and
> have
> >
> > my guess validated when I enter forms mode and finally hear the title
> text,
> >
> > but why should the user have to do this? Why not just provide hidden
> labels
> >
> > so they hear the exact same thing they would hear had visible labels been
> > used instead?
> >
> > I still don't see what advantage using title has over a hidden label,
> apart
> >
> > from:
> > -- it's slightly less bytes of HTML
> > -- it's slightly easier to implement (if you don't already have a hiding
> CSS
> >
> > class in your CSS, but if you do, creating hidden labels is pretty much
> just
> >
> > as easy)
> > -- it will show on hover as a tooltip to mouse-using sighted users (which
> > could be seen as a disadvantage by some people in some situations)
> >
> > What am I missing? Are these the only reasons why you find title to be
> > superior?
> >
> > Thanks,
> > Zoe
> >
> >
> >
> > Zoe Gillenwater
> > Web Accessibility Technical Architect
> > AT&T Consumer Digital Experience
> >
> > o: 919-241-4083
> > e: = EMAIL ADDRESS REMOVED =
> >
> > 4625 Creekstone Dr | Durham, NC 27703
> >
> > This email and any files transmitted with it are AT&T property, are
> > confidential, and are intended solely for the use of the individual or
> > entity to whom this email is addressed. If you are not one of the named
> > recipient(s) or otherwise have reason to believe that you have received
> this
> >
> > message in error, please notify the sender and delete this message
> > immediately from your computer. Any other uses, retention, dissemination,
> > forwarding, printing, or copying of this email is strictly prohibited.
> >
> >
> >
From: Bim Egan
Date: Wed, Nov 21 2012 2:25AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
Hi Ryan,
You're right, you've described the default behaviour of JAWS 10+, but it's
also a user option. Users can change the setting so that the JAWS9 and
previous rule applies, so forms mode isn't engaged until the space bar is
pressed.
It's common for advanced JAWS users to work in the old forms mode,
especially for testing, but other users are also likely to prefer it ... I
know people who don't start to fill in a form untill they have arrowed right
through it, to be sure they can complete it before commencing. Auto forms
mode is likely to be switched off for all sorts of reasons.
Cheers,
Bim
Original Message -----
From: "Ryan E. Benson" < = EMAIL ADDRESS REMOVED = >
To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
Sent: Wednesday, November 21, 2012 1:21 AM
Subject: Re: [WebAIM] Is use of <label> and title redundant?
I am reading this thread, and the following line by Sailesh confused me:
JAWS does read title in Firefox and IE when arrowing down the page
(without activating forms mode).
Are you using a version of JAWS before 10? If you aren't, I would say there
is a bug in JAWS or your install. According to
http://www.freedomscientific.com/training/surfs-up/Forms.htm, forms mode
should activate:
Auto forms mode was introduced in JAWS 10. Now when you press *TAB* or *
SHIFT+TAB* on a Web page and land in a form field or other form control,
forms mode is automatically on. It is confirmed with a sound that is
played. You no longer have to remember to switch forms mode on manually.
Forms mode is also turned on automatically if you use the *ARROW Keys* and
move into a form control such as an edit box. Again, a sound is played.
Additionally, if someone clicks into an edit box with a *MOUSE Click*,
forms mode comes on automatically.
--
Ryan E. Benson
On Mon, Nov 19, 2012 at 5:15 PM, Sailesh Panchang <
= EMAIL ADDRESS REMOVED = > wrote:
> Zoe / Bryan,
>
> JAWS does read title in Firefox and IE when arrowing down the page
> (without activating forms mode).
> NVDA does not read the title unless one is in forms mode.
> Using title is simpler as I have maintained all along as compared to
> off-screen label in the situations discussed.
> Failing to markup visible labels as labels and use title instead will
> fail SC 3.3.2.
> It is not alright to simply use title there.
> Sailesh
>
>
> On 11/19/12, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> > When you refer to browse mode, do you mean when using the arrow keys to
> > navigate up and down the page in the Virtual Buffer?
> >
> > When title attributes are included on form fields, I'm hearing these
> > announced correctly as form field labels in addition to tabbing using
> > interactive mode.
> >
> > This can be broken if a title attribute is included on a form field that
> > also includes an explicit label element with matching for and ID
> attributes
> >
> > in some ATs though.
> >
> >
> >
> > ----- Original Message -----
> > From: "GILLENWATER, ZOE M" < = EMAIL ADDRESS REMOVED = >
> > To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
> > Sent: Monday, November 19, 2012 10:17 AM
> > Subject: Re: [WebAIM] Is use of <label> and title redundant?
> >
> >
> > Hi Sailesh,
> >
> > But if you were using visible labels you *would* hear them read in
> > browse
> > mode, so using title text that is not read in browse mode is a different
> > experience from the norm, and it seems to me a worse one.
> >
> > For instance, if I had three fields for month, day, and year of a
> birthday,
> >
> > and each had a visible label in front of it, I would hear in browse mode
> > something like "Birthday. Month, edit, blank. Day, edit, blank. Year,
> edit,
> >
> > blank." It's clear to me what those fields are, which allows me to know
> > whether or not I want to enter forms mode and fill out this form.
> >
> > But if I got rid of the visible labels and just used title text, some
> screen
> >
> > reader users would now hear in browse mode something like "Birthday.
> Edit,
> > blank. Edit, blank. Edit, blank." You don't think this is a worse
> experience
> >
> > than the former? Sure, I can guess at what each of those fields is, and
> have
> >
> > my guess validated when I enter forms mode and finally hear the title
> text,
> >
> > but why should the user have to do this? Why not just provide hidden
> labels
> >
> > so they hear the exact same thing they would hear had visible labels
> > been
> > used instead?
> >
> > I still don't see what advantage using title has over a hidden label,
> apart
> >
> > from:
> > -- it's slightly less bytes of HTML
> > -- it's slightly easier to implement (if you don't already have a hiding
> CSS
> >
> > class in your CSS, but if you do, creating hidden labels is pretty much
> just
> >
> > as easy)
> > -- it will show on hover as a tooltip to mouse-using sighted users
> > (which
> > could be seen as a disadvantage by some people in some situations)
> >
> > What am I missing? Are these the only reasons why you find title to be
> > superior?
> >
> > Thanks,
> > Zoe
> >
> >
> >
> > Zoe Gillenwater
> > Web Accessibility Technical Architect
> > AT&T Consumer Digital Experience
> >
> > o: 919-241-4083
> > e: = EMAIL ADDRESS REMOVED =
> >
> > 4625 Creekstone Dr | Durham, NC 27703
> >
> > This email and any files transmitted with it are AT&T property, are
> > confidential, and are intended solely for the use of the individual or
> > entity to whom this email is addressed. If you are not one of the named
> > recipient(s) or otherwise have reason to believe that you have received
> this
> >
> > message in error, please notify the sender and delete this message
> > immediately from your computer. Any other uses, retention,
> > dissemination,
> > forwarding, printing, or copying of this email is strictly prohibited.
> >
> >
> >
From: Mike Moore
Date: Thu, Nov 22 2012 9:34AM
Subject: Re: Is use of <label> and title redundant?
← Previous message | Next message →
We have found a lot of issues with auto forms mode in our environment JFW 13, IE 9, and Win 7. Freedom does not seem to have fully resolved the issue of switching in and out of forms mode. We currently recommend that our staff who use JAWS (about 100 folks) continue to run with JAWS in manual forms mode.
From: Subhash Chhetri
Date: Thu, Nov 22 2012 9:23PM
Subject: Re: Is use of <label> and title redundant?
← Previous message | No next message
Hi Ryan,
Tab is not an only way to navigate around the form fields.
Some user prefer viewing the form fields before filling up. using down arrow
key keeping auto forms mode off is really helpful for this.
Best regards,
Subhash Chhetri