[WEB4LIB] Re: sanserif font for all browsers
Thomas Dowling
tdowling at ohiolink.edu
Mon Aug 27 12:53:45 EDT 2001
> Joan,
>
> 1. I'll argue that there's no such thing.
>
> "Browser compatibility" isn't the issue. Rather, it's whether people
have
> the typeface installed on their local system.
That's why the preferred method for specifying fonts--CSS--has a fall-back
position: the generic font families. Making sure that your list of fonts
ends with "sans-serif" lets the browser pick what it (or better yet, you)
have decided is the best available sans.
[Caveat: a CSS bug in IE5+ can lead to illegible font selections if you
only suggest the default families, so you should probably always have
Arial or Times New Roman as a last-ditch choice before the generic
family.]
>
> Helvetica/Arial is, as far as I know, the only "universally installed"
sans
> typeface. (Lucida Sans may be common but certainly not universal.)
They're certainly not universal, but Verdana and Trebuchet are both part
of Microsoft's core web fonts, both available for free for Macs and PCs,
and both designed from the ground up for high readability on comparatively
low-resolution devices like monitors.
BTW, in my experience Arial is better hinted than Helvetica--true
Helvetica on Windows is abysmally jaggy on screen--so if you're suggesting
both, I'd put Arial first.
>
> 2. This is pure argument, but...Why do you need to impose your choice of
> typeface? Why not let the user have whatever typeface he/she prefers?
I'd buy this if browsers either: still allowed separate font choices for
different HTML elements (see Opera today or Mosaic long ago); or
simplified the process of creating user stylesheets. As it is, a single
setting of Font X for all elements can often make for visual monotony.
>
> I'd particularly argue that point if there's any significant amount of
> text, given the inferior readability of sans.
>
With all due respect to Walt (which is a lot), I don't believe that serif
fonts have been proven to be more readable *on screen* than a good sans,
if that's even really been proven for other media. The number of dots
available to a character on screen is about 1/500th the number available
on a top quality printer (100dpi vs 2400dpi), and at that low resolution,
graceful serifs degrade into distracting blobs.
Thomas Dowling
OhioLINK - Ohio Library and Information Network
tdowling at ohiolink.edu
More information about the Web4lib
mailing list