WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: external links & new windows

for

From: Terrill Bennett
Date: Nov 27, 2010 11:39AM

If one is on a link, and wants a new tab or window, simply pressing
Ctrl+Enter - and it's done. It's a browser-thing, no Assitive
technology required. But THEY opened the new tab or window, so they
know it's there.

But, here's what W3C has to say:
http://www.w3.org/TR/WCAG20-TECHS/G200.html

Note W3C doesn't say you CAN'T open links to new windows, but does
say "...it is better not to open new windows and tabs since they can
be disorienting for people, especially people who have difficulty
perceiving visual content." Imagine, for example, people using a
screen magnifier who may not notice.

Jacob Nielsen's #2 in 1999, and #9 in 2007:
http://www.useit.com/alertbox/990530.html
http://www.useit.com/alertbox/9605.html

Dive Into Accessibility says it's not a great idea:
http://diveintoaccessibility.org/day_16_not_opening_new_windows.html

Jim Thatcher says it's disorienting:
http://www.jimthatcher.com/knowbility/65t.htm

I'm sure there's more. but this should get you started.

-- terrill --

At 10:03 AM 11/27/2010, you wrote:
>Hello,
>please excuse my very late response (I had a bicycle accident and have
>been becoming sensitized to what it's like to have no hands... though
>only temporarily.)
>
>Thank you all very much for your thoughts and findings - I'm almost
>cured of new-window-itis.
>A client of mine, however, is now requiring that I document our change
>in linking policy - i.e., indicate founded sources that corroborate that
>opening external links in same window is, indeed, more acccessible...
>I have been thinking of setting up a test suite and doing some good user
>testing on ambiguous questions such as this, but still feel
>not-quite-competent to do so, since I perceive user testing as a very
>complex thing.
>Thus, I'd be very grateful for pointers!
>
>
> >>>> On 11/9/2010 Jared Smith wrote:
> >>>>
> > On Tue, Nov 9, 2010 at 4:42 PM, Chris Hoffman wrote:
> >
> >
> >> One is for links that go to other sites, and the other is for
> links to non-HTML resources (PDFs, video files, etc.)
> >>
> >
> > Perhaps hidden in this argument is the suggestion that users don't
> > overly care if it's a link to another site.
> >
>Well yes, maybe it IS also a bit of a "tick" on the developer end to
>exaggerate the importance of whether a link is external or not.
>
>But if all our well-worded link texts, titles, "(pdf)"-labelling and
>whatnot aim at demystifying where a link is going to take us,
>my reasoning is that indicating that a link leads to an external site IS
>a relevant bit of information for users as well (after all, user will
>encounter a different site structure, layout, ... there).
>
>
> >> With Jared's comments about SEO dilution in mind, is there a
> good solution for these cases?
> >>
> >
> > A good, though not perfect, solution is to write the content to the
> > link using JavaScript after page load. The WebAIM site currently uses
> > hidden text and a visible background icon for external links. We will
> > probably dynamically add this text and icon using JavaScript in a
> > future site release. You don't even have to give the link a class name
> > - just detect whether it links to someplace other than your own domain
> > and then add the class name and text if appropriate.
> >
>The only issue I have with this is that with javascripting turned off
>the "external" information is also lost (= your "though not perfect"?)
>Of course, IF one chooses to peruse the link target adress & compare to
>the present site's, one can still see one will be going off-site.
>But who does that??? :)
>
>Nathalie
>
>
>