W3C home > Mailing lists > Public > www-style@w3.org > May 2013

[css3-fonts] Arabic and generic families

From: fantasai <fantasai.lists@inkedblade.net>
Date: Tue, 14 May 2013 06:07:33 -0700
Message-ID: <51923715.5020806@inkedblade.net>
To: "www-style@w3.org" <www-style@w3.org>
http://www.w3.org/TR/2013/WD-css3-fonts-20130212/#generic-font-families

   # Some scripts, such as Arabic, are almost always cursive.

This section seems to handle Arabic rather poorly. It's not useful
to classify all Arabic fonts as cursive, just because the script
has a cursive nature. There are differences in fonts, just as there
are in Chinese fonts, that are analogous to the serif/sans-serif/
monospace categories. And Arabic *does* have "printed" and "cursive"
forms. (I can read "printed" handwriting, but I can't read "cursive".)

A 'serif' Arabic font would be one with variations in stroke thickness,
often such as you'd get with an inked-reed pen. A 'sans-serif' one
would be one without such variation. A 'cursive' one would be one that
has a cursive, handwritten nature. And of course monospace is monospaced.

To pull out some samples,
   http://image.linotype.com/nonlatinfonts/arabicfonts/1210/arabicfonts_overview.gif
The first font should classify as 'serif', the second, fourth, and
fifth as 'sans-serif'. The third font seems to have a handwritten-
printed feel, so maybe would classify as 'cursive' for the same
eason Comic Sans seems to be, despite not actually being a cursive
design; I'm less sure on that one.

See Behdad's comments on exactly this issue from 2010:
   http://lists.w3.org/Archives/Public/www-style/2010Aug/0525.html
And John Hudson:
   http://lists.w3.org/Archives/Public/www-style/2010Aug/0527.html
Though I'm not sure I agree that DecoType Naskh would be 'serif'
rather than 'cursive'--it is very calligraphic, more like Zapfino
than Times Roman.

DecoType Ruq'ah is true cursive, compare:
   http://lspwww.epfl.ch/conferences/ridt98/decotype.html

~fantasai
Received on Tuesday, 14 May 2013 13:08:01 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:30 UTC