Usability: Serif and Sans-Serif font faces?

[Geoff Hart <ghart -at- videotron -dot- ca>]

Ned Bedinger wondered: <<What is the usability issue that dictates the 
use of serif and sans-serif font faces?>>

Not a trivial question, since many, many other typographic factors and 
their many interactions greatly outweigh the serif/sans issue. Assuming 
that you've chosen a serif and a sans serif font that are both designed 
to be legible rather than decorative and that you've manipulated all 
other typographic factors (line length, leading, tracking, etc.) to 
support the goal of legibility:

- In print, the serifs on the fonts increase the visual distinctiveness 
of the character shapes, and thus increase their legibility compared 
with sans serif type. This is often an issue of the quality of the 
design itself, and for faces such as Garamond, there are so many 
different variations of the design from different type designers that 
it's hard to generalize even for a specific font.

- Onscreen, serifs are harder to render properly, since many subtle 
features of the type are smaller than the pixels that must be used to 
display them. So a serif font may be fuzzier as a result of 
anti-aliasing or other display artifacts. "Slab" serifs display better 
because they tend to be set at right angles (i.e., follow a line of 
pixels) and tend to be closer to the pixel size. (That's a very broad 
generalisation, thus misleading.)

This information is based on a range of cognitive psych studies that 
I've read over the years, dating back to studies of old IBM CRT fonts 
in the early 1980s in _Human Factors_ magazine and more recent 
"broad-based" studies such as Colin Wheildon's.

None of these studies has been definitive, because the authors either 
controlled so many typographic factors that the results apply only to 
that combination of factors, or left so many factors uncontrolled that 
the comparison of serif vs. sans was overwhelmed by variations in the 
other factors. Moreover, even studies that have shown _statistically_ 
significant differences have generally failed to show _practically_ 
significant differences; Eva Brumberg's recent articles in _Technical 
Communication_ are interesting in this respect.

In short, if you choose legible fonts and use them well, the 
differences between serif and sans serif aren't meaningful in practice.

<<I first got the word (that sans-serif was more readable) from Mac 
users, who seemed to value aesthetics more than PC users.>>

This may have been true at one point, since Windows took more than a 
decade to become a practical tool for design, whereas the Mac was 
design-friendly right from the start. But I doubt the generalization is 
still valid; nowadays, you'll find roughly as many designers using 
Windows as use the Mac, and the Mac folk have raised the bar high 
enough that Windows users now expect comparable esthetics.

<<In my experience, you would neither have to look far to find users 
that want all body text in sans-serif, nor would you have to look far 
to find users who would read serif body text with nary a whimper.>>

Precisely. I personally find serif body text much easier to read, but I 
can read a well-chosen, properly formatted sans serif font well enough 
to get by. The difference is largely a matter of experience, not 
inherent legibility. I read so much serif text and so little sans that 
it's only natural to expect I'd be faster with serif. I imagine Nordic 
and French readers, who tend to prefer sans serif text, read 
significantly slower than North American readers despite their font 
choice, and to me, that's the compelling point in this whole debate.

--Geoff Hart   ghart -at- videotron -dot- ca
(try geoffhart -at- mac -dot- com if you don't get a reply)


^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

SEE THE ALL NEW ROBOHELP X5 IN ACTION: RoboHelp X5 is a giant leap forward
in Help authoring technology, featuring Word 2003 support, Content 
Management, Multi-Author support, PDF and XML support and much more! http://www.macromedia.com/go/techwrldemo