WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: line length and myth of the fold

for

From: Jukka K. Korpela
Date: Apr 18, 2008 1:30PM

Karl Groves wrote:

> As mentioned in the post below, here's just a handful more resources
> on line-length and usability:
>
> http://psychology.wichita.edu/surl/usabilitynews/42/text_length.htm

Quickly checked that one. It says that no difference in reading speed
was observed (note that only short texts were considered) and that
people expressed subjective preference for shorter lines. Note that
testing only three lengths is problematic - the _optimum_ is hardly any
of them. The study is far too limited for any serious conclusions,
except perhaps mildly to the direction _opposite_ to your agenda.

If you wish to "debunk the myth", you need to do better than that. For
example, to present a reputable meta-study (over the many studies on
this topic) or a very large study.

Just picking up a few studies with conclusions you like (and not even
very carefully, as the above example shows) should not convince anyone.

There are two real problems with line length. One is excessively short
length, often appearing on portal-type sites or on pages that somehow
imitate newspaper format, with many narrow columns. The other one is
unlimited length, which you get when you author very simply without
trying anything particular.

Excessively short line length is a symptom of poor overall design. It
cannot be solved in isolation but only by refraining from such design.
But line length considerations can be relevant arguments in favor of a
design change.

Unlimited line length might be seen as a non-problem, since users can
normally control the window width. It might be a nuisance factor, if
people visit both newspaper-style pages and simple-style pages, but
that's basically a usability problem rather than accessibility. And it's
not obvious at all what should be done, technically. Personally, I'm
inclined into something like

@media screen {
body { max-width: 40em; }
}

(knowing that it doesn't work on IE 6 and older ).


Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/