WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: A small doubt on alt text

for

From: Pratik Patel
Date: Feb 19, 2013 3:06AM

Hello all,

I realize that this has been a great discussion. And the solutions provided
to the original problem appeared to be accurate and most part. However I
would like to raise one question and a bit of a concern looking at the
original example, I wonder if there's a need for alternative text at all.
The question that I would ask when creating this kind of link is why is the
blind person receiving this additional information? What use does the link
text actually serve? And what use does the alternative text serve? Is it
even necessary to provide the additional information to the blind person
when this information is not given to the sighted person? Does the context
surrounding the link itself provide this additional information to the
sighted person that a blind person is not getting? While it is important for
us to comply with WCAG Criteria, it is also important for us to keep in mind
that these criteria should not be taken so literally. The mindset of
compliance often appears to take over real usage patterns. Rather than
asking fundamental questions about the content, the focus is so much on the
WCAG Criteria that the real usage shifts away from the user. I think there's
always a value to asking basic questions over fancy techniques.

Sorry for the mini rant.

Regards,

Pratik

Pratik Patel
Founder and CEO, EZFire
T: 718-928-5529
M: 718-249-7019
E: <EMAIL REMOVED> (or <EMAIL REMOVED> )
Follow me on Twitter: @ppatel
Follow me on LinkedIn: http://www.linkedin.com/pub/pratik-patel/9/985/882
Skype: Patel.pratik


-----Original Message-----
From: <EMAIL REMOVED>
[mailto: <EMAIL REMOVED> ] On Behalf Of Joe Chidzik
Sent: Tuesday, February 19, 2013 3:41 AM
To: WebAIM Discussion List
Subject: Re: [WebAIM] A small doubt on alt text

> Thanks for a great summary of the situation gentlemen, nicely put.
> In case there is a CSS tooltip text outside of the <a> wouldn't it make
the most
> sense to recommend that text be referred to with aria-describedby inside
the link,
> but then the text itself be hidden from screen reader users (aria-hidden),
so as to
> avoid double speaking it?
> Of course I do need to test what would happen if a text with
aria-hidden="true" is
> referenced by aria-describedby.

[Joe Chidzik]
Out of curiosity, I tried the following example with JAWS IE9.

<a href="page2.html" aria-describedby="link-page2">Jacobs Camp </a><span
id="link-page2" class="screenreader-text">HS Jacobs - a UAHC camp in> Utica,
MS</span>
<a href="page2.html">Jacobs Camp <span id="link-page2"
class="screenreader-text">HS Jacobs - a UAHC camp in> Utica, MS</span></a>

The first link uses aria-describedby to attach some (hidden by CSS)
descriptive text to the link, the second example uses CSS to hide the
descriptive text off-screen, but leaves it within the link element. Whilst
both examples read out almost identically when tabbing through the page,
they display differently within the JAWS link list. These appear in the
links list as:

-"Jacobs camp"
-" Jacobs Camp HS Jacobs - A UAHC camp in Utica, MS"

The second example appearts more useful in this sense, as the descriptive
information is available within the links list. Perhaps it would make sense
for JAWS to pick up and include text added via aria-describedby when
displaying links in the links list, or form elements in the JAWS forms list.


Joe