E-mail List Archives
Thread: Duplicate links...
Number of posts in this thread: 14 (In chronological order)
From: Stephanie Sullivan
Date: Wed, Jun 07 2006 9:40AM
Subject: Duplicate links...
No previous message | Next message →
What is the proper protocol for getting around the "duplicate link failure"?
Let's say you've got a CSS styled menu... And at the bottom of your page,
you've got footer links that go to the same places... The Cynthia Says WCAG
report fails it...
Is there a recommended method of getting around this? Is it really confusing
to a user to have these same links accessible at both the top and bottom of
the page? Or is this a case of "I know best," and you just ignore the
failure?
Thanks... Stephanie
From: John Foliot
Date: Wed, Jun 07 2006 10:20AM
Subject: RE: Duplicate links...
← Previous message | Next message →
Stephanie Sullivan wrote:
>
> Is there a recommended method of getting around this? Is it really
> confusing to a user to have these same links accessible at both the
> top and bottom of the page? Or is this a case of "I know best," and
> you just ignore the failure?
Well...
On one hand, there *is* the issue of duplicate links - very often screen
reader users will bring up a list of links (in JAWS it's "INSERT + F7") to
replicate a "page scanning" type of review - one reason why link text must
also make sense when taken out of context.
However, depending on your page layout and the amount of content, having
those duplicate links at the top and bottom *may* improve
usability/accessibility for other users. So your stance of "I know best"
does have some merit.
Perhaps using the title attribute on the anchor tags could aid somewhat for
those users who do INSERT + F7; perhaps something along the lines of: <a
href="URL" title="Duplicate link to foobar">Foobar</a>. Please note however
that this presumes that the end user's verbosity settings (etc.) are such
that the title attribute is captured/conveyed.
As far as the WCAG is concerned, always remember that they are Guidelines,
and not standards, and so are subject to interpretation. You are 95% there
if you understand the issues and then make an informed decision. At the end
of the day, you have to ask yourself "does the upside for one group of users
outweigh the down side for another?" In this particular scenario, I would
suspect that you are safe in ignoring the "failure" being outputted by the
testing tool. (But, as always, this is just an opinion)
HTH
JF
From: Stephanie Sullivan
Date: Wed, Jun 07 2006 4:40PM
Subject: Re: Duplicate links...
← Previous message | Next message →
On 6/7/06 12:08 PM, "John Foliot" < = EMAIL ADDRESS REMOVED = > wrote:
> As far as the WCAG is concerned, always remember that they are Guidelines,
> and not standards, and so are subject to interpretation. You are 95% there
> if you understand the issues and then make an informed decision. At the end
> of the day, you have to ask yourself "does the upside for one group of users
> outweigh the down side for another?" In this particular scenario, I would
> suspect that you are safe in ignoring the "failure" being outputted by the
> testing tool. (But, as always, this is just an opinion)
Thanks John... That's been my thinking as well... :)
Stephanie
From: Patrick H. Lauke
Date: Wed, Jun 07 2006 5:10PM
Subject: Re: Duplicate links...
← Previous message | Next message →
John Foliot wrote:
> On one hand, there *is* the issue of duplicate links - very often screen
> reader users will bring up a list of links (in JAWS it's "INSERT + F7") to
> replicate a "page scanning" type of review - one reason why link text must
> also make sense when taken out of context.
[...]
> As far as the WCAG is concerned, always remember that they are Guidelines,
> and not standards, and so are subject to interpretation.
But, as I mentioned already, Cynthia shouldn't (and in my modest
testing, rightly doesn't) fail pages with duplicate links. It's only
when you have the same link text pointing to different URLs that you get
a failure. So, there isn't actually a validation issue here, unless
Stephanie's links are in fact pointing to different places?
Patrick
--
Patrick H. Lauke
___________
re
From: Stephanie Sullivan
Date: Wed, Jun 07 2006 6:10PM
Subject: Re: Duplicate links...
← Previous message | Next message →
On 6/7/06 6:59 PM, "Patrick H. Lauke" < = EMAIL ADDRESS REMOVED = > wrote:
> But, as I mentioned already, Cynthia shouldn't (and in my modest
> testing, rightly doesn't) fail pages with duplicate links. It's only
> when you have the same link text pointing to different URLs that you get
> a failure. So, there isn't actually a validation issue here, unless
> Stephanie's links are in fact pointing to different places?
They aren't my links... :) It's someone I was helping on a list... And they
SWEAR they're the same (on cursory look, they appear to be -- except one..
And they've tried them with trailing slashes, without trailing slashes,
etc...)
The only thing I see is the opposite... He has DIFFERENT link text to one
URL that is the same...
Top:
<a href="/legal/" class="p7PMtrg" id="n7">Legal</a>
Bottom:
<a href="/legal" title="terms of use for this website">Terms of Use</a>
(and this one has the trailing slash on one and none on the other... I think
he was testing it)...
Mysterious. :)
From: Moore, Michael
Date: Thu, Jun 08 2006 7:40AM
Subject: RE: Duplicate links...
← Previous message | Next message →
Actually there are a couple of instances when the automated checkers
will fail things for "duplicate" links.
1. Same link text different destinations. This can be a false failure if
for instance on link drops to an in-page anchor at the start of the
content and the other leads to the top of the page, otherwise this is a
real problem for all users.
2. Different link text same destination. This is frequently a problem
when we try to link to a page included in navigation with somewhat
different link text inside of the content of a page. This example is
more of a judgment call, (imho) - the different text can be helpful for
some users in gaining understanding of where they should go, however
when I bring up a link list in JAWS, I think that there are two separate
destinations. Best case - the link text in the navigation is so clear
that there is no reason to use different wording in the body of the
content.
mike
-----Original Message-----
From: = EMAIL ADDRESS REMOVED =
[mailto: = EMAIL ADDRESS REMOVED = ] On Behalf Of Patrick H.
Lauke
Sent: Wednesday, June 07, 2006 6:00 PM
To: WebAIM Discussion List
Subject: Re: [WebAIM] Duplicate links...
John Foliot wrote:
> On one hand, there *is* the issue of duplicate links - very often
> screen reader users will bring up a list of links (in JAWS it's
> "INSERT + F7") to replicate a "page scanning" type of review - one
> reason why link text must also make sense when taken out of context.
[...]
> As far as the WCAG is concerned, always remember that they are
> Guidelines, and not standards, and so are subject to interpretation.
But, as I mentioned already, Cynthia shouldn't (and in my modest
testing, rightly doesn't) fail pages with duplicate links. It's only
when you have the same link text pointing to different URLs that you get
a failure. So, there isn't actually a validation issue here, unless
Stephanie's links are in fact pointing to different places?
Patrick
--
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
___________
Web Standards Project (WaSP) Accessibility Task Force
http://webstandards.org/
___________
Address list
messages to = EMAIL ADDRESS REMOVED =
From: Andrew Kirkpatrick
Date: Thu, Jun 08 2006 8:10AM
Subject: RE: Duplicate links...
← Previous message | Next message →
> 1. Same link text different destinations. This can be a false
> failure if for instance on link drops to an in-page anchor at
> the start of the content and the other leads to the top of
> the page, otherwise this is a real problem for all users.
This is, I believe, what Cynthia says. I see this most frequently for
"read more" links that might follow short news headlines and that there
might be lots of on a single page. There is an open question about how
much you can rely on context for a link that the WCAG group has
discussed.
> 2. Different link text same destination. This is frequently
> a problem when we try to link to a page included in
> navigation with somewhat different link text inside of the
> content of a page. This example is more of a judgment call,
> (imho) - the different text can be helpful for some users in
> gaining understanding of where they should go, however when I
> bring up a link list in JAWS, I think that there are two
> separate destinations. Best case - the link text in the
> navigation is so clear that there is no reason to use
> different wording in the body of the content.
I don't know that this is a problem. It seems like you could easily
have a page with links to the same place but that usability indicates
different users are inclined to follow the links when presented
differently - some people like a link that says "shopping cart" at the
top, others perfer the "view cart and checkout" link on the side. As
long as the links are appropriately labeled, this shouldn't be an issue.
AWK
From: Patrick Lauke
Date: Thu, Jun 08 2006 9:10AM
Subject: RE: Duplicate links...
← Previous message | Next message →
> Moore, Michael
> 2. Different link text same destination.
But this does not contravene WCAG, does it? I haven't come across
any validators that flag this as a validation fail.
Patrick
________________________________
Patrick H. Lauke
Web Editor / University of Salford
http://www.salford.ac.uk
________________________________
Web Standards Project (WaSP) Accessibility Task Force
http://webstandards.org/
________________________________
From: Alastair Campbell
Date: Thu, Jun 08 2006 9:40AM
Subject: RE: Duplicate links...
← Previous message | Next message →
Micheal Moore wrote:
> > 2. Different link text same destination.
Patrick wrote:
> But this does not contravene WCAG, does it? I haven't come across
> any validators that flag this as a validation fail.
I'm writing:
I believe Bobby checks this, and I thought Cynthia says does but I could
be mistaken.
The guideline says "clearly identify", the techniques doc has "If more
than one link on a page shares the same link text, all those links
should point to the same resource."
By implication, that means if they don't share the same text, they
should point to different resources?
The first paragraph in the techniques implies that to, but isn't
explicit:
http://www.w3.org/TR/WCAG10-HTML-TECHS/#link-text
It's a bit of a pain, the typical case (as someone has mentioned) is
things like having "_Contact_" in the main navigation precluding you
from having "_contact_us_" in the content.
In an audit situation I would let this go as people would understand,
but if you are just going by automated checks you would probably change
it to "_contact_ us".
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: Andrew Kirkpatrick
Date: Thu, Jun 08 2006 10:10AM
Subject: RE: Duplicate links...
← Previous message | Next message →
> The guideline says "clearly identify", the techniques doc has
> "If more than one link on a page shares the same link text,
> all those links should point to the same resource."
>
> By implication, that means if they don't share the same text,
> they should point to different resources?
Oh I don't think that it implies that at all. I don't think that there
is a valid logical proof to be made from that reversal.
> The first paragraph in the techniques implies that to, but isn't
> explicit:
> http://www.w3.org/TR/WCAG10-HTML-TECHS/#link-text
I didn't see anything that alluded to different link locations. What
are you seeing?
AWK
From: Stephanie Sullivan
Date: Thu, Jun 08 2006 11:40AM
Subject: Re: Duplicate links...
← Previous message | Next message →
On 6/8/06 11:05 AM, "Patrick Lauke" < = EMAIL ADDRESS REMOVED = > wrote:
>
>> 2. Different link text same destination.
>
> But this does not contravene WCAG, does it? I haven't come across
> any validators that flag this as a validation fail.
Since I started this convo, let me update you on what the issue was...
It was indeed the link I showed him with the differing link text pointing to
the same URL. This is completely illogical to me as there are many cases
where you would have valid reasons for having differing link text...
I must agree with Andrew, I really don't think the converse is implied in
the guideline ...
As an FYI, this user was validating his work at Cynthia says...
Stephanie Sullivan
Community MX Partner :: http://www.communitymx.com/author.cfm?cid=1008
Adobe Community Expert :: http://tinyurl.com/6huw3
Co-Author .: "Macromedia Dreamweaver MX 2004 Magic" :. New Riders
Dreamweaver Task Force for WaSP
"The competitor to be feared is one who never bothers about you at all, but
goes on making his own business better all the time." - Henry Ford
From: Moore, Michael
Date: Thu, Jun 08 2006 12:00PM
Subject: RE: Duplicate links...
← Previous message | Next message →
Stephanie Sullivan:
"It was indeed the link I showed him with the differing link text
pointing to the same URL. This is completely illogical to me as there
are many cases where you would have valid reasons for having differing
link text...
I must agree with Andrew, I really don't think the converse is implied
in the guideline ...
As an FYI, this user was validating his work at Cynthia says..."
Mike:
If memory serves, Bobby (WebXM) Cynthia Says and Lift will all flag this
situation as an error. This is why blind adherence to the results of
automated testing tools is never advisable. It is much more important
to understand the reasoning behind the guidelines and the implications
of our coding decisions. I stand by my assertion that this situation is
a judgment call. Different text can be helpful to some users but does
raise a minor concern in a screen reader link list.
Mike
From: Alastair Campbell
Date: Fri, Jun 09 2006 2:50AM
Subject: RE: Duplicate links...
← Previous message | Next message →
Moore, Michael wrote:
> If memory serves, Bobby (WebXM) Cynthia Says and Lift will
> all flag this
> situation as an error. This is why blind adherence to the results of
> automated testing tools is never advisable. It is much more important
> to understand the reasoning behind the guidelines and the implications
> of our coding decisions. I stand by my assertion that this
> situation is a judgment call. Different text can be helpful to some
users but
> does raise a minor concern in a screen reader link list.
Totally agreed.
If I were writing an automated rule set, I can understand reading it one
way (same link text should go to same resource), and assuming that it
should be the other way around to - it's not explicit.
But to be clear on what I think the criteria should be: would a person
understand where they are going?
In simple cases like "contact" vs "contact us", I wouldn't be concerned,
but as you said, it's a judgement call.
Kind regards,
-Alastair
From: Jukka K. Korpela
Date: Fri, Jun 09 2006 4:20AM
Subject: Re: Duplicate links...
← Previous message | No next message
On Wed, 7 Jun 2006, Stephanie Sullivan wrote:
> The only thing I see is the opposite... He has DIFFERENT link text to one
> URL that is the same...
>
> Top:
> <a href="/legal/" class="p7PMtrg" id="n7">Legal</a>
>
> Bottom:
> <a href="/legal" title="terms of use for this website">Terms of Use</a>
They do _not_ use the same URL. The relative URLs "/legal/" and "/legal"
are not identical, and when resolved to absolute URLs, the results are
not identical either. Remember: URLs are just strings that may be used to
retrieve a resource.
The URLs "/legal/" and "/legal" _may_ refer to the same resource, even in
the strict sense (as opposite to the fact that exactly the same content
may appear as copies as distinct resources). For example, a server may
respond to a request for "/legal" with a temporary or permanent
redirection to "/legal/" (which itself might be redirected to
"/legal/index.html" for example). The URLs are still distinct; they are
different names for a resource, at least in a particular situation.
On quite similar grounds, "/legal/" and "/laillinen/" or "/foobar/zap/zip"
or "foo" _may_ all refer to the same resource.
The point is that by looking at the HTML source alone you cannot decide
whether "/legal/" and "legal" refer to the same resource, still less
whether they refer to the same content (e.g., two resources with content
copied verbatim from one to another). A checker could decide on the "same
resource" issue only by resolving the relative URLs and sending actual
HTTP requests and analyzing the headers of the responses. In order to
decide on the "same content" issue, there is really no way - the checker
could send GET requests and compare the actual content it receives (doing
redirection as needed), but this would as such be just a snapshot. The
pages could be identical now and completely different the next second.
As a human reader who looks at the source code, you are of course entitled
into the suspicion that there might be links to the same resource with two
different URLs and two different link texts. I would actually bet at least
1 : 5 for it.
> (and this one has the trailing slash on one and none on the other... I think
> he was testing it)...
Maybe. Maybe it was just a minor inconsistency. Generally, it is best to
avoid such usage, but there is nothing formally wrong with it, and no
explicit accessibility recommendation against it either. Users seldom get
confused, since although they might see two different URLs on the status
when they mouse over links, they will see the ultimate resultant URLs
(after any redirections) on the address (location) line, and most people
look at it much more than the status line.
Of course, it's a completely different question whether a page could
contain "duplicate" links that refer to the same document or the same
content with different or same URLs or different or same link texts, for
some values of "same" and "different". I feel dizzy now and won't comment
this topic now except by saying that I once thought it is nice to have
several links to the same page on one page, in order to make it easier to
people to find that important page, but now I think the idea was wrong.
--
Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/