E-mail List Archives
Thread: skip-nav tabindex setfocus etc.
Number of posts in this thread: 37 (In chronological order)
From: Hoffman, Allen
Date: Tue, Mar 14 2006 1:50PM
Subject: skip-nav tabindex setfocus etc.
No previous message | Next message →
I'd like to hear opinions on the following:
Providing the skip-nav capacity as described in Section 508 1194.22(o)
is most often implemented as an invisible pixel with an alt attribute.
This works fine for screen reader users, but for low-vision users of
magnification software isn't really of much help. I am thinking of
first recommending that skip-nav links be visible, second that for
non-scrolling pages, why not just use setfocus and put focus where it
should go, or use tabindex=1 to jump to the content--saves keystrokes.
thoughts?
Allen Hoffman
From: Christian Heilmann
Date: Tue, Mar 14 2006 2:10PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> I'd like to hear opinions on the following:
>
> Providing the skip-nav capacity as described in Section 508 1194.22(o) is
> most often implemented as an invisible pixel with an alt attribute. This
> works fine for screen reader users, but for low-vision users of
> magnification software isn't really of much help. I am thinking of first
> recommending that skip-nav links be visible, second that for non-scrolling
> pages, why not just use setfocus and put focus where it should go, or use
> tabindex=1 to jump to the content--saves keystrokes.
100% ACK on keeping skip links visible and real text links. This also
helps sighted users that have to use a keyboard.
Not too sure on setfocus and tabindex, as they are actually tools to
help the navigation of forms and not links.
--
Chris Heilmann
Blog: http://www.wait-till-i.com
Writing: http://icant.co.uk/
Binaries: http://www.onlinetools.org/
From: Jared Smith
Date: Tue, Mar 14 2006 2:20PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Hoffman, Allen wrote:
> I am thinking of first recommending that skip-nav links be visible
Excellent!
> for non-scrolling pages, why not just use setfocus and put
> focus where it should go, or use tabindex=1 to jump to the
> content--saves keystrokes.
But what if they don't want to jump right to the content, but instead want
to tab to your navigation. If someone enters directly at your page, they
may not even be aware that there is a navigation or head section on the
page. tabindex is almost always a bad solution. Also, how do you know that
page won't scroll? It likely will for those using screen enlargers - an
audience that can benefit greatly from skip navigation links.
I would think that the skip-nav link would be sufficient.
Jared Smith
WebAIM.org
From: Patrick H. Lauke
Date: Tue, Mar 14 2006 6:20PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Hoffman, Allen wrote:
> Providing the skip-nav capacity as described in Section 508 1194.22(o)
> is most often implemented as an invisible pixel with an alt attribute.
> This works fine for screen reader users, but for low-vision users of
> magnification software isn't really of much help. I am thinking of
> first recommending that skip-nav links be visible,
Just on this first part, how about a hybrid approach where the skip link
is initially hidden and only appears (obviously in a logical, visible
place where a user with magnification would not miss it) when
tabbed/focussed to it via CSS?
I first messed around with the idea here
http://www.splintered.co.uk/experiments/21/ and have since implemented
it on a few sites (most recently added it to http://www.salford.ac.uk
during our first round of updates to the templates).
--
Patrick H. Lauke
___________
re
From: Jared Smith
Date: Tue, Mar 14 2006 6:30PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Patrick H. Lauke wrote:
> Just on this first part, how about a hybrid approach where the skip link
> is initially hidden and only appears (obviously in a logical, visible
> place where a user with magnification would not miss it) when
> tabbed/focussed to it via CSS?
The WebAIM site (which is currently undergoing a major renovation) uses
this approach. Just be sure that the CSS does not hide/display it in a way
that screen readers will not read it (e.g., don't use display:none or
visibility:hidden). Changing the position, adding a background color, or
similar will provide this functionality without interfering much with site
design or layout for the majority of users that do not need, use, or even
notice this feature.
Jared Smith
WebAIM.org
From: Patrick H. Lauke
Date: Tue, Mar 14 2006 6:40PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Jared Smith wrote:
> Just be sure that the CSS does not hide/display it in a
> way that screen readers will not read it (e.g., don't use display:none
> or visibility:hidden). Changing the position, adding a background color,
> or similar will provide this functionality without interfering much with
> site design or layout for the majority of users that do not need, use,
> or even notice this feature.
Yup, on the Salford site I basically have this markup directly after the
logo in the header
<div><a href="#content" id="skipper">Skip to the content</a></div>
(disregarding the controversial accesskey I also included in the live site)
and the following style rules
#skipper {
position: absolute;
top: 0; left: 0;
margin: 0; padding: 0;
text-indent: -600em;
color: #000;
}
#skipper:focus, #skipper:hover, #skipper:active {
display: block;
width: 100%;
text-indent: 1em;
background: #ced318;
font-weight: bold;
}
P
--
Patrick H. Lauke
___________
re
From: Christian Heilmann
Date: Wed, Mar 15 2006 12:40AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> Patrick H. Lauke wrote:
> > Just on this first part, how about a hybrid approach where the skip link
> > is initially hidden and only appears (obviously in a logical, visible
> > place where a user with magnification would not miss it) when
> > tabbed/focussed to it via CSS?
>
> The WebAIM site (which is currently undergoing a major renovation) uses
> this approach. Just be sure that the CSS does not hide/display it in a way
> that screen readers will not read it (e.g., don't use display:none or
> visibility:hidden). Changing the position, adding a background color, or
> similar will provide this functionality without interfering much with site
> design or layout for the majority of users that do not need, use, or even
> notice this feature.
The new WaSP templates will have this as well, however, I personally
think a visible skip link shows buy-in to web accessibility a lot more
than any "WCAG compliant and Bobby Approved" banners would ever do.
We do use visible "back to top" links, why not "skip to navigation"?
The ones on my blog (http://www.wait-till-i.com/) got hardly noticed
by anyone who wanted to just read the content but got good feedback
from people who really needs skip links.
--
Chris Heilmann
Blog: http://www.wait-till-i.com
Writing: http://icant.co.uk/
Binaries: http://www.onlinetools.org/
From: Tim Beadle
Date: Wed, Mar 15 2006 2:40AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
On 15/03/06, Christian Heilmann < = EMAIL ADDRESS REMOVED = > wrote:
> The new WaSP templates will have this as well, however, I personally
> think a visible skip link shows buy-in to web accessibility a lot more
> than any "WCAG compliant and Bobby Approved" banners would ever do.
>
> We do use visible "back to top" links, why not "skip to navigation"?
>
> The ones on my blog (http://www.wait-till-i.com/) got hardly noticed
> by anyone who wanted to just read the content but got good feedback
> from people who really needs skip links.
+1 on that idea. I did the same on http://optics.org/ but haven't had
any feedback from people who appreciate the inclusion, just queries
from people wondering what the skip link is!
Actually, if anyone would like to have a look at the above site and
point out any serious access issues (I hope there are few or none but
you never know...), it'd be much appreciated.
Regards,
Tim
From: Alastair Campbell
Date: Wed, Mar 15 2006 3:00AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Damn, I wrote the below before I scrolled up in my webAIM folder, it
seems to becoming standard practice (which is great).
> I am thinking of first
> recommending that skip-nav links be visible, second that for
> non-scrolling pages, why not just use setfocus and put focus
> where it should go, or use tabindex=1 to jump to the
> content--saves keystrokes.
We usually take a slightly different approach that I first noted on Tom
Gilder's blog tom.me.uk in 2003 (not there anymore), although we
implemented it differently.
The skip-link isn't visible until you tab to it. This means that those
using the keyboard find it, but otherwise it isn't visible (and it isn't
hidden from screen readers either). See www.particlephysics.ac.uk for an
example.
With regards to tabindex, don't confuse people! Having a different order
between tabbing and arrowing (usual reading method) is very confusing,
and should only be used for specific forms or applications.
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: Gez Lemon
Date: Wed, Mar 15 2006 4:00AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Hi Alastair,
On 15/03/06, Alastair Campbell < = EMAIL ADDRESS REMOVED = > wrote:
> The skip-link isn't visible until you tab to it. This means that those
> using the keyboard find it, but otherwise it isn't visible (and it isn't
> hidden from screen readers either). See www.particlephysics.ac.uk for an
> example.
Have you done any user-testing to confirm that assertion? For screen
reader users it would be obvious, but it requires those not using
screen reader but rely on keyboard navigation to pay attention to the
screen as they navigate through the document. There is no reason for
anyone to assume they should pay special attention to the screen on
the off-chance that some navigation aid will suddenly present itself,
as it isn't expected behaviour.
I tend to agree with Christian that for a skip link to be useful, it
should be visible, as the behaviour is then predictable.
Best regards,
Gez
--
_____________________________
Supplement your vitamins
http://juicystudio.com
From: Alastair Campbell
Date: Wed, Mar 15 2006 4:30AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Gez Lemon wrote:
> Have you done any user-testing to confirm that assertion? ...
> There is no reason for anyone to assume they should pay special
> attention to the screen on the off-chance that some navigation
> aid will suddenly present itself, as it isn't expected behaviour.
A good point, and unfortunately I've found that people who rely on
keyboard navigation but don't use a screen reader are hard to come by
for user tests, so it isn't a direct observation.
However, just thinking about it from a logical interaction point of
view, if you rely on a keyboard to navigate, but without something like
a screen reader, what are the options apart from tab?
If you are using tab to navigate the page, how do you know a link is
selected if you aren't looking at the screen?
It's entirely possible I've missed a context of use, e.g. a common
technologies for keyboard oriented access?
If not though, I'm not sure who it would disadvantage.
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: Patrick Lauke
Date: Wed, Mar 15 2006 4:40AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> Alastair Campbell
> Gez Lemon wrote:
> > Have you done any user-testing to confirm that assertion? ...
> > There is no reason for anyone to assume they should pay special
> > attention to the screen on the off-chance that some navigation
> > aid will suddenly present itself, as it isn't expected behaviour.
>
> A good point, and unfortunately I've found that people who rely on
> keyboard navigation but don't use a screen reader are hard to come by
> for user tests, so it isn't a direct observation.
>
> However, just thinking about it from a logical interaction point of
> view, if you rely on a keyboard to navigate, but without
> something like
> a screen reader, what are the options apart from tab?
I think the http://www.particlephysics.ac.uk/ example was a slightly
bad one, as the skip links are very easy to miss if you're not expecting
them. That's why I'd say that the links need to appear in a position
where the user would direct their attention anyway, and be very obvious
(background colour etc).
Of course, having an always visible skip link is the most ideal solution,
but it often is at odds with visual design (and personally I still
find it to be a clunky solution to make up for the fact that user
agents don't normally allow to easily skip things like entire lists etc).
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: Christian Heilmann
Date: Wed, Mar 15 2006 4:50AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> However, just thinking about it from a logical interaction point of
> view, if you rely on a keyboard to navigate, but without something like
> a screen reader, what are the options apart from tab?
Voice recognition?
From: Christian Heilmann
Date: Wed, Mar 15 2006 5:00AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> > > However, just thinking about it from a logical interaction point of
> > > view, if you rely on a keyboard to navigate, but without
> > something like
> > > a screen reader, what are the options apart from tab?
> >
> > Voice recognition?
>
> I've not had personal experience with that (my voice isn't very clear,
> couldn't get it to work within my patience limits), how do you navigate
> within a browser page?
You can say "show links" and get labels next to each link. If you hide
the skip link you'll never reach that one.
Ok, maybe a moot point as it'll highlight the other links :-)
From: Helen A
Date: Wed, Mar 15 2006 5:10AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
At 11:30 15/03/2006, Christian wrote:
> > However, just thinking about it from a logical interaction point of
> > view, if you rely on a keyboard to navigate, but without something like
> > a screen reader, what are the options apart from tab?
>
>Voice recognition?
Or scrolling right and/or down?
From: Patrick Lauke
Date: Wed, Mar 15 2006 5:20AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> Christian Heilmann
> > However, just thinking about it from a logical interaction point of
> > view, if you rely on a keyboard to navigate, but without
> something like
> > a screen reader, what are the options apart from tab?
>
> Voice recognition?
But how does this affect the current discussion re people not looking
at the screen? Even if I had voice recognition, issuing a command like
"next link", I'd assume that the user would look at the screen to see
which link now has focus? (sorry if this is a na
From: Alastair Campbell
Date: Wed, Mar 15 2006 5:30AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> Ok, maybe a moot point as it'll highlight the other links :-)
I would say so, it's really for people on linear access devices (e.g.
switches, head wands, keyboards) that I'd be cautious about, but I have
no reason to think they wouldn't use tab or something that acts like
tab.
I'd also add my agreement for Patrick's comment:
> having an always visible skip link is the most ideal solution,
> but it often is at odds with visual design (and personally I still
> find it to be a clunky solution to make up for the fact that user
> agents don't normally allow to easily skip things like entire lists
etc).
I seem to remember WCAG v2 having something about being able to
programmatically tell that there is a group of links, so why can't user
agents do that and provide a skip method?
On the other hand, since we have main navigation at the bottom, we do
tend to provide skip *to* navigation instead.
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: Alastair Campbell
Date: Wed, Mar 15 2006 5:40AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> > However, just thinking about it from a logical interaction point of
> > view, if you rely on a keyboard to navigate, but without
> something like
> > a screen reader, what are the options apart from tab?
>
> Voice recognition?
I've not had personal experience with that (my voice isn't very clear,
couldn't get it to work within my patience limits), how do you navigate
within a browser page?
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: L
Date: Wed, Mar 15 2006 5:50AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Christian Heilman wrote:
"> However, just thinking about it from a logical interaction point of
> view, if you rely on a keyboard to navigate, but without something
> like a screen reader, what are the options apart from tab?
Voice recognition?"
Not always. People with some motor disabilities, particularly Cerebral Palsy, don't have the quality of muscle control to speak clearly enough for voice recognition systems to understand.
Regards,
L
From: Gez Lemon
Date: Wed, Mar 15 2006 6:00AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Hi Alastair,
On 15/03/06, Alastair Campbell < = EMAIL ADDRESS REMOVED = > wrote:
> It's entirely possible I've missed a context of use, e.g. a common
> technologies for keyboard oriented access?
>
> If not though, I'm not sure who it would disadvantage.
My original point was more about comprehension. The first part is
comprehending that an unexpected behaviour was in fact a feature for
the visitor to navigate the document, and secondly, the visitor needs
to comprehend that something had changed.
For example, someone who relies on keyboard navigation might look at a
screen to see a huge navigation list. They look down at the keyboard
to locate the tab key, and start to press it continuously. By the time
they look back up at the screen, they're busily tabbing through the
navigation unaware that they had bypassed a feature was meant to help
them.
Another example might be someone looking at the screen as they're
tabbing through the content, only to notice that the screen flickers
as they're tabbing through (as it displays and quickly removes the
skip link). Would they comprehend in the flicker that a feature just
popped up that could help them?
Best regards,
Gez
--
_____________________________
Supplement your vitamins
http://juicystudio.com
From: Tim Harshbarger
Date: Wed, Mar 15 2006 6:10AM
Subject: RE: skip-nav tabindex setfocus etc.RE: [WebAIM] skip-navtabindex setfocus etc.
← Previous message | Next message →
Hi,
My limited observations are that people, who use a keyboard or devices
simulating keyboard input to navigate the user interface, tend to tab
quickly to their goal and don't seem to notice a link that seems to
appear for a split second. It seems as though the user is focusing on a
specific link or form control. The user knows it is X number of tabs to
that link or form control--although they might not know the exact
number, just that it's a lot of tabbing.
I am uncertain exactly how to explain it.
Its not like:
Tab, "oh, there is a skip link I can use."
Its more like:
Tab, tab, tab, tab, tab, tab, tab, tab, tab, tab, tab..."gee, something
funny happened after the first tab." Then they may either tab merrily
on their way to their original goal or start tabbing back to the skip
link to check it out.
I am also not certain if skip links are as useful to someone using a
screen reader or screen magnifier as they are to other people using the
keyboard or devices simulating keyboard input. In the case of people
using a screen reader, my thought is that good semantic markup may be
more important. In the case of people using screen magnification, those
people I am familiar with seem to use the mouse more. They might use
the tab key once they are in a form or maybe even navigating the
browser's menubar. However, it seems they tend to use the mouse more
when just looking around a page or trying to find a specific link in a
page full of links.
Of course, this is all based on my personal observations, so I am
uncertain if it indicates approaches most people utilize or just those I
happened to observe.
Tim
From: Alastair Campbell
Date: Wed, Mar 15 2006 7:10AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Tim Harshbarger wrote:
> Tab, tab, tab, tab, tab, tab, tab, tab, tab, tab, tab..."gee,
> something
> funny happened after the first tab." Then they may either tab merrily
> on their way to their original goal or start tabbing back to the skip
> link to check it out.
I'm curious as to what would happen if they realise that they are going
through the content section first, then the right side, then the left
hand navigation (as they would on our site). That might be unexpected
to!
> I am also not certain if skip links are as useful to someone using a
> screen reader or screen magnifier as they are to other people
> using the keyboard or devices simulating keyboard input.
Difficult to compare, but skip links are useful to people using screen
readers, it's very much linear access.
> In the case of people using screen magnification, those
> people I am familiar with seem to use the mouse more. They might use
> the tab key once they are in a form or maybe even navigating the
> browser's menubar.
That matches my observations, I've done a fair bit of usability testing
with people on screen readers and magnifiers (both for clients and
research). The first being entirely keyboard dependant, the second being
very mouse oriented. (I've also noticed that screen reader users seem
more aware of skip links in general, screen magnifier users being
generally unaware of what "skip to menu" might mean).
Going back to the skip links, it's a difficult one to persuade a client
about design-wise, although to be fair I haven't tried very hard,
relying on the pop-out on tab approach.
In the fast-tab scenario above, I guess it comes down to either:
- The skip link has to be first and visible, or
- The pop-out has to be very obvious.
I am very curious as to what effect code-source order has on people who
use the keyboard for navigation and can see the page.
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: Kynn Bartlett
Date: Wed, Mar 15 2006 7:20AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
On 3/15/06, Gez Lemon < = EMAIL ADDRESS REMOVED = > wrote:
>
> Have you done any user-testing to confirm that assertion? For screen
> reader users it would be obvious, but it requires those not using
> screen reader but rely on keyboard navigation to pay attention to the
> screen as they navigate through the document. There is no reason for
> anyone to assume they should pay special attention to the screen on
> the off-chance that some navigation aid will suddenly present itself,
> as it isn't expected behaviour.
>
> I tend to agree with Christian that for a skip link to be useful, it
> should be visible, as the behaviour is then predictable.
This is my concern too. I don't get the desire to have skip-navigation links
and yet hide them like some kind of easter egg. They're not predictable
interaction that way.
Also, have you tested these on mobile devices? When you're viewing the web
on a Treo palm/cell phone, skip navigation really /is/ useful. Especially
test any blog that puts the navigation bar first in source order, e.g.
http://atrios.blogspot.com/ -- it could definitely use a "skip navigation"
link at the beginning.
--Kynn
From: Martin Pistorius
Date: Wed, Mar 15 2006 7:30AM
Subject: Re: skip-nav tabindex setfocus etc.RE: [WebAIM]skip-navtabindex setfocus etc.
← Previous message | Next message →
Hi All,
As a person with a disability who either uses a head mouse or the keyboard
to navigate, I'd say first and foremost make sure your semantic structure
is correct, because if it is generally tabindex it's not needed.
I'd only use tab orders on site with ads because you don't want to have to
tab through six links to some other products site before you get to the
navigation..
As for Skip to content links, my personal opinion is they are useful but
then I feel they must always be visible, not only when tabbed to.
Of course it goes with out saying, make sure the site is keyboard
accessibility, there is nothing more frustrating being not being able to
navigate a site.
Anyway that's my two cents worth
Martin
----- Original Message -----
From: "Tim Harshbarger" < = EMAIL ADDRESS REMOVED = >
To: "WebAIM Discussion List" < = EMAIL ADDRESS REMOVED = >
Sent: Wednesday, March 15, 2006 3:02 PM
Subject: RE: [WebAIM] skip-nav tabindex setfocus etc.RE: [WebAIM]
skip-navtabindex setfocus etc.
Hi,
My limited observations are that people, who use a keyboard or devices
simulating keyboard input to navigate the user interface, tend to tab
quickly to their goal and don't seem to notice a link that seems to
appear for a split second. It seems as though the user is focusing on a
specific link or form control. The user knows it is X number of tabs to
that link or form control--although they might not know the exact
number, just that it's a lot of tabbing.
I am uncertain exactly how to explain it.
Its not like:
Tab, "oh, there is a skip link I can use."
Its more like:
Tab, tab, tab, tab, tab, tab, tab, tab, tab, tab, tab..."gee, something
funny happened after the first tab." Then they may either tab merrily
on their way to their original goal or start tabbing back to the skip
link to check it out.
I am also not certain if skip links are as useful to someone using a
screen reader or screen magnifier as they are to other people using the
keyboard or devices simulating keyboard input. In the case of people
using a screen reader, my thought is that good semantic markup may be
more important. In the case of people using screen magnification, those
people I am familiar with seem to use the mouse more. They might use
the tab key once they are in a form or maybe even navigating the
browser's menubar. However, it seems they tend to use the mouse more
when just looking around a page or trying to find a specific link in a
page full of links.
Of course, this is all based on my personal observations, so I am
uncertain if it indicates approaches most people utilize or just those I
happened to observe.
Tim
From: Alastair Campbell
Date: Wed, Mar 15 2006 7:40AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Kynn Bartlett wrote:
> Also, have you tested these on mobile devices? When you're
> viewing the web on a Treo palm/cell phone, skip navigation
> really /is/ useful.
That will depend on their CSS support. If they have limited support
(e.g. don't understand @import or positioning), then yes, they should
get the skip link at the top, as do screen readers.
Difficult to tell what would happen if they have mixed support for
things, have you got one to try?
> I don't get the desire to have skip-navigation
> links and yet hide them like some kind of easter egg. They're not
> predictable interaction that way.
There are several different user groups for whom this functionality is a
benefit, and a group or two that doesn't like them there (often the
client!).
My desire it to provide the function to those that need it, preferably
without having to set site preferences.
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: Kynn Bartlett
Date: Wed, Mar 15 2006 8:00AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
On 3/15/06, Alastair Campbell < = EMAIL ADDRESS REMOVED = > wrote:
>
> Kynn Bartlett wrote:
> > Also, have you tested these on mobile devices? When you're
> > viewing the web on a Treo palm/cell phone, skip navigation
> > really /is/ useful.
>
> That will depend on their CSS support. If they have limited support
> (e.g. don't understand @import or positioning), then yes, they should
> get the skip link at the top, as do screen readers.
So you're depending on broken CSS implementations?
It's important to remember that mobile devices with touchscreens often don't
have a "hover" or "focus" mode of interaction, or if they do, they're so
rarely used and hard to access that they're not worth it...
> There are several different user groups for whom this functionality is a
> benefit, and a group or two that doesn't like them there (often the
> client!).
> My desire it to provide the function to those that need it, preferably
> without having to set site preferences.
Sigh. Another case in which more widely deployed Edapta technology would
have been real, real nice. </lament>
--Kynn
From: Alastair Campbell
Date: Wed, Mar 15 2006 8:50AM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Kynn wrote:
> So you're depending on broken CSS implementations?
Not really, I'm depending on predictable implementations. E.g. my phone
doesn't support CSS beyond colours, it applies no layout at all, so the
skip links appear fine.
Mobile Opera has two modes (on my version) one which applies layout and
one that doesn't. In the layout mode, I think (although I could be
wrong) that it brings up the skip link when you press down. I don't have
mobile Opera anymore, so perhaps that's wishful thinking on my part, but
I do know that it applied the layouts well on the sites I tested,
mitigating the need for the skip link anyway (it acts more like a
mouse).
There is a middle ground of devices that try to support CSS but get
things wrong, those are the ones that will get these type of skip links
wrong. However, I would say the onus there is on the device developers
(do any of them support handheld media types apart from Opera?).
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: Al Sparber
Date: Wed, Mar 15 2006 11:30AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
From: "Christian Heilmann"
The new WaSP templates will have this as well, however, I personally
think a visible skip link shows buy-in to web accessibility a lot more
than any "WCAG compliant and Bobby Approved" banners would ever do.
------------------------------------
Is it about stickers and approvals or about accessibility?
------------------------------------
We do use visible "back to top" links, why not "skip to navigation"?
------------------------------------
Back to top links are debatable, but have been overused into an
almost-convention. We have over 30,000 customers and each time we've
developed a product (or deployed on our site) visible skip links, many
people have expressed confusion. We've also had many reports of
clients of customers being confused.
-----------------------------------
The ones on my blog (http://www.wait-till-i.com/) got hardly noticed
by anyone who wanted to just read the content but got good feedback
from people who really needs skip links.
-----------------------------------
Personally, I feel that skip links that appear on focus are always
seen by the people who would need to use them. Your mileage, as
always, will vary.
--
Al Sparber
PVII
http://www.projectseven.com
"Designing with CSS is sometimes like barreling down a crumbling
mountain road at 90 miles per hour secure in the knowledge that
repairs are scheduled for next Tuesday".
From: Christian Heilmann
Date: Wed, Mar 15 2006 11:40AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
> The new WaSP templates will have this as well, however, I personally
> think a visible skip link shows buy-in to web accessibility a lot more
> than any "WCAG compliant and Bobby Approved" banners would ever do.
>
> Is it about stickers and approvals or about accessibility?
Ask clients like councils. I've spent many a time telling them that
putting lots of banners on the site does not make it accessible.
Clients that are legally bound (or think they are more legally bound
than others) are very much for showing off that they are doing
something in terms of accessibility. That in most of the cases this is
lip-service or sometimes even harmful is the other side of the story,
but at least this is a hook you have to get them to embrace diversity
and "ambient findability".
Lame flamebait, by the way.
> ------------------------------------
> We do use visible "back to top" links, why not "skip to navigation"?
> ------------------------------------
> Back to top links are debatable, but have been overused into an
> almost-convention. We have over 30,000 customers and each time we've
> developed a product (or deployed on our site) visible skip links, many
> people have expressed confusion. We've also had many reports of
> clients of customers being confused.
So if more visible skip links were to be used all around the web
they'd become an almost-convention, too. Chicken and Egg.
From: Patrick H. Lauke
Date: Wed, Mar 15 2006 11:50AM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Christian Heilmann wrote:
> So if more visible skip links were to be used all around the web
> they'd become an almost-convention, too. Chicken and Egg.
But I for one don't want that egg to hatch. Using skip links seems to me
like a very hacky way of compensating for current user agents' inability
to provide a user friendly way to skip block level elements etc natively
(e.g. "jump over this whole list, jump to the next heading, etc").
IMHO, of course...
P
--
Patrick H. Lauke
___________
re
From: Mark D. Urban
Date: Wed, Mar 15 2006 12:10PM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
I'm actually of the opinion that Allen is right on his first suggestion -
Skip Navs should be visible.
The second is an interesting thought, although I think that a user of a
screen mag may easy get lost if the focus jumps without warning.
Regards,
-Mark D. Urban
919-395-8513 (cell)
Chair, North Carolina Governor's Advocacy Council for Persons with
Disabilities
Keep up with the latest in worldwide accessibility at (
<http://www.icdri.org/>; http://www.icdri.org/)
_____
From: Hoffman, Allen [mailto: = EMAIL ADDRESS REMOVED = ]
Sent: Tuesday, March 14, 2006 3:42 PM
To: = EMAIL ADDRESS REMOVED =
Subject: [WebAIM] skip-nav tabindex setfocus etc.
I'd like to hear opinions on the following:
Providing the skip-nav capacity as described in Section 508 1194.22(o) is
most often implemented as an invisible pixel with an alt attribute. This
works fine for screen reader users, but for low-vision users of
magnification software isn't really of much help. I am thinking of first
recommending that skip-nav links be visible, second that for non-scrolling
pages, why not just use setfocus and put focus where it should go, or use
tabindex=1 to jump to the content--saves keystrokes.
thoughts?
Allen Hoffman
From: Gez Lemon
Date: Wed, Mar 15 2006 12:20PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
On 15/03/06, Al Sparber < = EMAIL ADDRESS REMOVED = > wrote:
> Back to top links are debatable, but have been overused into an
> almost-convention. We have over 30,000 customers and each time we've
> developed a product (or deployed on our site) visible skip links, many
> people have expressed confusion. We've also had many reports of
> clients of customers being confused.
The confusion is probably more about the link phrase than the concept
of a skip link. Phrases like "skip navigation", "jump to main
content", "jump to navigation", etc, are confusing because they're
not typical of link phrases as they contain verbs that describe an
action rather than a target. A link phrase of "main content",
"navigation menu", or anything else that describes the target rather
than the action is less likely to confuse people.
> Personally, I feel that skip links that appear on focus are always
> seen by the people who would need to use them. Your mileage, as
> always, will vary.
Feelings are great, but empirical evidence to support your hypothesis
would be a lot more useful.
I agree with Patrick that this is really a user-agent issue - skip
links are a hack to make up for shortfalls in user-agents. Having said
that, if you're going to provide skip links, I don't see the point in
hiding them, as the very people that could benefit from them are
likely to miss them. Hiding skip links is merely paying lip-service to
a recognised a problem; better than nothing, but barely adequate.
Best regards,
Gez
--
_____________________________
Supplement your vitamins
http://juicystudio.com
From: Alastair Campbell
Date: Wed, Mar 15 2006 1:20PM
Subject: RE: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Unusually, I disagree with Gez on a couple of points:
> A link phrase of "main content",
> "navigation menu", or anything else that describes the target rather
> than the action is less likely to confuse people.
In some instances, yes, but it could depend on a few other factors,
including source order.
Focusing on visual & keyboard access: If you have a typical layout (e.g.
header, 3 columns including left navigation), with the navigation at the
bottom of the source code; is having a link "navigation menu" just above
the navigation (visually) not going to be confusing?
In this context, I would have thought the model for it would be more of
an action? Visually you are moving the focus rather than going
somewhere, as you aren't even scrolling significantly.
I'm struggling to think of what is actually useful in this situation,
any skip link is either going to appear useless, or actually be useless!
(i.e. it looks like the navigation is next so the skip link looks
useless, or the skip link is to the content which is next anyway.)
> I agree with Patrick that this is really a user-agent issue - skip
> links are a hack to make up for shortfalls in user-agents.
Agreed.
> Having said
> that, if you're going to provide skip links, I don't see the point in
> hiding them, as the very people that could benefit from them are
> likely to miss them.
(I'm having flash backs to an accesskey discussion!)
There are two broad groups who can benefit from skip links, people using
screen readers or linear (e.g. text or mobile) browsers, and those who
use a keyboard (or equivalent) and can see the screen (including
layout).
The first group I tend to cater for by putting the main navigation at
the bottom of the source order, and providing skip links to it. (Content
first is generally nicer to deal with linear access). They are not
affected by this discussion. The second group is then harder to deal
with, as the order of page elements is not as expected.
Usually I have worked down to an answer by this point of an email, but
the best I can think of at the moment is to move the skip link from the
left (above or near the navigation) to the somewhere on the right at the
top of the content area.
A hidden link might also work better if it were just above a
consistently placed link at the top of the content area (such as a
breadcrumb trail), rather than in the header.
Kind regards,
-Alastair
--
Alastair Campbell | Director of User Experience
Nomensa Email Disclaimer:
http://www.nomensa.com/email-disclaimer.html
From: Gez Lemon
Date: Wed, Mar 15 2006 2:00PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Hi Alastair,
On 15/03/06, Alastair Campbell < = EMAIL ADDRESS REMOVED = > wrote:
> Unusually, I disagree with Gez on a couple of points:
We're not a million miles in apart in our views.
> > A link phrase of "main content",
> > "navigation menu", or anything else that describes the target rather
> > than the action is less likely to confuse people.
>
> In some instances, yes, but it could depend on a few other factors,
> including source order.
>
> Focusing on visual & keyboard access: If you have a typical layout (e.g.
> header, 3 columns including left navigation), with the navigation at the
> bottom of the source code; is having a link "navigation menu" just above
> the navigation (visually) not going to be confusing?
Yes, and it would also be confusing to keyboard navigate a website
with such a layout. If the skip link was in a column separate from the
navigation, but first in the source order, then the link phrase
wouldn't be confusing.
> In this context, I would have thought the model for it would be more of
> an action? Visually you are moving the focus rather than going
> somewhere, as you aren't even scrolling significantly.
That's where the confusion arises, which is partly due to the hacky
nature of skip links. Skip links are actions, but users don't expect
to see links marked up as actions. Interface controls have a
well-defined meaning, and people subconsciously expect user-interface
controls to be marked up (or used, in a non-web environment)
correctly. In any GUI environment, actions are typically provided
through command buttons (or some other user interface control). On the
web, hyperlinks usually take the user to a document, or a document
fragment. When hyperlinks behave differently to the user's
expectations (which is becoming more wide-spread due to badly written
web applications), it's inevitable that some degree of confusion will
arise. This is why I suggested that link phrases should describe the
target rather than the action, as that's what people subconsciously
think when they see a link phrase.
> I'm struggling to think of what is actually useful in this situation,
> any skip link is either going to appear useless, or actually be useless!
> (i.e. it looks like the navigation is next so the skip link looks
> useless, or the skip link is to the content which is next anyway.)
Agreed - in this particular scenario, I think the web
designer/developer should put some more thought into how it's going to
work.
Best regards,
Gez
--
_____________________________
Supplement your vitamins
http://juicystudio.com
From: Al Sparber
Date: Wed, Mar 15 2006 3:20PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Christian Heilmann wrote:
>> The new WaSP templates will have this as well, however, I
>> personally
>> think a visible skip link shows buy-in to web accessibility a lot
>> more than any "WCAG compliant and Bobby Approved" banners would
>> ever
>> do.
>>
>> Is it about stickers and approvals or about accessibility?
>
> Ask clients like councils. I've spent many a time telling them that
> putting lots of banners on the site does not make it accessible.
> Clients that are legally bound (or think they are more legally bound
> than others) are very much for showing off that they are doing
> something in terms of accessibility. That in most of the cases this
> is
> lip-service or sometimes even harmful is the other side of the
> story,
> but at least this is a hook you have to get them to embrace
> diversity
> and "ambient findability".
> Lame flamebait, by the way.
Why is that? I was merely responding to your statement and asked a
legitimate, but largely rhetorical question. I certainly did not
intend to bait you - although you seem to have found a hook on your
own :-)
In any event, I am gathering that this issue has a lot of opinions but
no definitive answers. Some good points were raised, but no supporting
evidence or data to draw a good conclusion from. Onward and upward.
Thanks for the banter.
Al Sparber
PVII
http://www.projectseven.com
"Designing with CSS is sometimes like barreling down a crumbling
mountain road at 90 miles per hour secure in the knowledge that
repairs are scheduled for next Tuesday".
From: Al Sparber
Date: Wed, Mar 15 2006 3:30PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | Next message →
Gez Lemon wrote:
> The confusion is probably more about the link phrase than the
> concept
> of a skip link. Phrases like "skip navigation", "jump to main
> content", "jump to navigation", etc, are confusing because they're
> not typical of link phrases as they contain verbs that describe an
> action rather than a target. A link phrase of "main content",
> "navigation menu", or anything else that describes the target rather
> than the action is less likely to confuse people.
Partly true, no doubt.
>> Personally, I feel that skip links that appear on focus are always
>> seen by the people who would need to use them. Your mileage, as
>> always, will vary.
>
> Feelings are great, but empirical evidence to support your
> hypothesis
> would be a lot more useful.
I wish I had some, too. I can only go by what customers tell us and,
sadly, accessibility is not given the weight it should by many folks.
> I agree with Patrick that this is really a user-agent issue - skip
> links are a hack to make up for shortfalls in user-agents. Having
> said
> that, if you're going to provide skip links, I don't see the point
> in
> hiding them, as the very people that could benefit from them are
> likely to miss them. Hiding skip links is merely paying lip-service
> to
> a recognised a problem; better than nothing, but barely adequate.
Do you have any supporting data you could point me to? Specifically,
do keyboard or mouseless surfers tend to not look at the screen as
they navigate or do they tend to press the tab key in such rapid
succession that they would miss a focus link?
Thanks.
--
Al Sparber
PVII
http://www.projectseven.com
"Designing with CSS is sometimes like barreling down a crumbling
mountain road at 90 miles per hour secure in the knowledge that
repairs
are scheduled for next Tuesday".
From: Gez Lemon
Date: Wed, Mar 15 2006 4:00PM
Subject: Re: skip-nav tabindex setfocus etc.
← Previous message | No next message
Hi Al,
On 15/03/06, Al Sparber < = EMAIL ADDRESS REMOVED = > wrote:
> > I agree with Patrick that this is really a user-agent issue - skip
> > links are a hack to make up for shortfalls in user-agents. Having
> > said
> > that, if you're going to provide skip links, I don't see the point
> > in
> > hiding them, as the very people that could benefit from them are
> > likely to miss them. Hiding skip links is merely paying lip-service
> > to
> > a recognised a problem; better than nothing, but barely adequate.
>
> Do you have any supporting data you could point me to? Specifically,
> do keyboard or mouseless surfers tend to not look at the screen as
> they navigate or do they tend to press the tab key in such rapid
> succession that they would miss a focus link?
Good call; I wish I did have supporting data, but it's purely
observational. I've seen people that have been focusing on the part of
the screen they want to arrive at, and only paying partial attention
to the links they're tabbing through and have tabbed straight past a
skip link which would have put them closer to where they wanted to be.
Without usability studies, we're all second-guessing based on our own
experiences and observations.
It would be a great to get some data on hidden and permanently
revealed skip-links, preferable using eye-tracking technology, to get
a better understanding of how people who navigate mainly with the
keyboard interact both consciously and subconsciously with web
content.
Best regards,
Gez
--
_____________________________
Supplement your vitamins
http://juicystudio.com