WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Question about links opening in new tabs/windows

for

From: Karl Groves
Date: Dec 3, 2013 12:15PM

This comes from a very old blog post of mine but is still very relevant:


Regardless of whether they're used for supplementary content or external
links, the practice [of opening new windows] raises additional usability
problems.

One of my colleagues writes:

I have been doing some usability testing for a local hospital and we have
been fortunate to have participants with no familiarity with the web all
the way to self-described experts. We tested pages where external links
opened in a new window and pages where external links opened in the same
window. Nobody had trouble navigating when links opened in the same window.
Many people had trouble navigating when links opened in a new window.

- Most everyone used the back button to navigate the site and not the
site navigation links.
- Many people didn't notice when a new window opened, even when the page
explicitly indicated that links would open in a new window.
- Many people were unable to get back to the main site when a new window
opened because the back button did not work.

I have long thought that people get disoriented when following links and
that I, as the designer, had to build in ways to help people keep their
bearings. So I've done the javascript popups and the target="whatever". But
from watching people work with the web I am starting to think that people
don't really notice when they go from one site to another, and that they
also don't really care. They are questing for information and don't much
care where they get it. The damage done by trying to be helpful and impose
"context" is much greater than that done by leaving people to make their
own way.

These observations mirrored my own. During a recent usability test, I
noticed that new windows had caused disorientation in all users who had
experienced them. Primarily, this manifested itself in the form of a
"broken" back button. When users experienced these new windows and wished
to go back to where they came from, they were unable to. In fact, ALL
participants needed to be informed by the test facilitator that a new
window had opened and that in order to get back to the site, they needed to
close the new window. This is the last thing a facilitator wants to do.
Interrupting a test participant derails the testing process, and a
facilitator will often wait quite a long time to see if the participant can
figure out the problem for themselves. However, the new windows were such a
problem that the facilitator needed to interrupt the test so that we could
continue.
Any new window causes confusion and frustration, resulting in lost users

New windows opened via the "target" attribute causes confusion for the
user, breaks the back button, and harms your users' perception of your
site. For those who think that opening a new window will help keep the user
from "losing the [original] site", the real effect is exactly the opposite.

In a recent usability test I observed, about half the participants actually
got lost. With the new window completely covering the original site, they
did not notice that they had two (or more) windows open. In an attempt to
return to the original site, they repeatedly hit the back button. When that
didn't work, they totally closed all browser windows and completely started
over. Despite the fact that some sites insist on opening new windows "so
the visitor doesn't lose our site" when in actuality, the new windows
appear to actually cause people to lose the site rather than keep them.
"Band-Aids" only cover a gaping wound

During discussions of opening new windows with a colleague of mine, the
argument was posed that seemed to make sense: The primary problem with
these new windows wasn't just the fact that the new window existed. The
biggest issue was the fact that the window was full size, therefore
covering the old window completely. This, he said, is what caused the
users' confusion. The solution my colleague proposed was that the new
window should be created so that it is noticeably smaller than full screen
- say, 700 x 500. On the surface, this seems like an excellent solution.

Then, during another recent test, I noticed that users who encountered
these smaller-than-full-sized windows would click the button in the upper
right-hand corner in order to maximize the window. Because the site being
tested opened new windows in multiple screens, some participants wound up
with 3 or even 4 windows of the site open to full screen. It almost gave
the impression that the new windows were cloning themselves. This practice
of opening smaller windows is certainly is not a "fix", in my opinion.
Indeed, there's just no cure for the problems created by new windows except
to simply avoid them.



On Tue, Dec 3, 2013 at 1:58 PM, Greg Gamble < <EMAIL REMOVED> > wrote:

> The reason I open a new tab for an offsite link is pretty simple.
> Consider this: You're on a site with a high level of complexity, you
> click a link to a similar site, but not the original. Now you're on
> someone else's site and if your following an article you could even click
> out to another site. Now ... where is that original page I was on?? Back
> buttons don't work on every site and there may have been some redirects in
> there that you're not aware of, so now you need to open the history page to
> see if you can find that original page.
>
> Easier to open a separate tab and maintain the original site ... and yes
> it's 2013, but not everyone knows how to use a browser beyond clicking a
> link or using a back button.
>
> JMO :-)
>
>
> Greg
>
>
>