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

Re: Using Heading to Replace Skip Links

From: Ramón Corominas <listas@ramoncorominas.com>
Date: Sat, 12 May 2012 14:16:19 +0200
Message-ID: <4FAE5493.50203@ramoncorominas.com>
CC: W3C WAI-IG <w3c-wai-ig@w3.org>
Hi all,

Russ said

 > Like this example which is activated on focus (becomes visible when 
"tabbed")? http://maxdesign.com.au/jobs/example-skip/03.htm

We recommended the same technique for the European Commission website. 
You can see it in action here:

Adam said:

 > For users dependent on a screen reader, every new page involves 
discovery. The almost innate relevance filtering available for visual 
users is a time-consuming and strategic process for screen reader users 
which is why navigational elements and meaningful structure is so 

Of course, but I think you are assuming that every blind user does 
exactly the same to discover the structure of a web page, and that's not 
the case. Some users use skip links, others simply ignore them, and 
others don't like them (especially when there are more than a single 
"skip to content").

Headings are used by far more screen reader users than skip links. You 
can find data about this in the different surveys that WebAIM has 
carried out.

Although the 3 surveys conducted by WebAIM asked different questions, 
but all of them are consistent in the relative order of the two 
mechanisms, and headings were always first in the navigational 
preferences in the three available surveys.

- Survey #1 (January 2009) [1]: if we sum the results for the answers 
"whenever possible", "often" and "sometimes", skip links sum up to 66%, 
while headings reach 90%.

- Survey #2 (Oct 2009): the lack of headings was mentioned as #8 of the 
"problematic issues" [2], while the lack of skip links was mentioned as #11.

- Survey #3 (Dec 2010): headings are the #1 method for finding 
information in a page (57,2%) [3] (it must be said that there was not a 
specific answer about skip links, but "navigating through the links was 
3rd with only a 12,8%). In a different question, users said that their 
use of skip links had decreased (60,6%) with respect to the first survey.

Of course, this doesn't mean that skip links are a bad idea, more than 
60% of blind users find them useful. But, if headings are properly used, 
saying that "they are not enough" is simply not true.

 > One of the most irritating issues using navigation links with a 
screen reader can be the placement of destination anchors.

Yes, but even if they are properly placed, behaviour is different with 
different browsers. As far as I can remember, some browsers didn't like 
pointing to an <a id="content" /> and prefer <a name="content" /> (or 
viceversa), or the keyboard focus is not moved, which can be very confusing.

 > so, I broadly concur with Vivienne’s sentiment that “Frankly, I think 
it should be a requirement as we're wanting to make things better for 
people to get to the content, not more difficult.“

Yes, but this is more a usability issue. If used properly, headings do 
provide a navigation mechanism for blind users, and I guess that sighted 
users can simply ignore menus and start reading from the big title of 
the main content.

 > 1.	Skip links must be the first elements on a page that accept focus

This means that skip links should probably be "orfan" elements. No 
heading before skip links. Should they belong to a section? To the 
header? Are they apart from the global structure of the document?

 > 2.	Skip links must be always visible

This can be inacceptable from the designer's point of view. In addition, 
I guess that some users with cognitive disabilities or understanding 
problems could feel confused by a link that says "skip to content". Does 
this mean that the content is not here? Where is the content, then? 
(Note: "skip" is not an easy-to-translate word in some languages; for 
example, in Spanish the most similar word would be "saltar" -jump-; 
jumping is not exactly skipping).

 > 3.	Skip links must always include link text that clearly identifies 
their purpose (I have seen on pages recently <li>Skip to: ><a 
href=”#content”>Content</a></li>. Quite apart from the very common and 
equally irritating overuse of list elements to position content, I don’t 
believe the link purpose is clear.)

Yes, the link purpose should be clear. But what you call "irritating 
overuse of list elements" is IMHO a requirement for SC 1.3.1, and most 
users find it useful more than irritating.

 > 4.	Skip to links must destinate immediately prior to relevant content


 > 5.	Avoid using the same name and id attribute values for destination 
anchors as this can result in a similar target vagueness with some 
screen readers in some situations

Why? Can you please explain this?

Kind regards,

[1] WebAIM's Survey of preferences of screen reader users #1 (Jan 2009). 
"Skip" Links

[2] WebAIM's Survey #2 (Oct 2009). Most problematic issues

[3] WebAIM's Survey #3 (Dec 2010). Finding information
Received on Saturday, 12 May 2012 12:16:52 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:44 UTC