W3C home > Mailing lists > Public > www-font@w3.org > October to December 1999

Re: position of baseline relative to em square

From: Erik van der Poel <erik@netscape.com>
Date: Tue, 30 Nov 1999 20:52:29 -0800
Message-ID: <3844A98D.BAB64645@netscape.com>
To: Nick Nussbaum <nickn@seanet.com>
CC: www-font@w3.org, Todd Fahrner <fahrner@pobox.com>
Nick Nussbaum wrote:
> The baseline is a function of the script. Han and Math center character
> center the baseline and the em square would be correct. Latin fonts use an
> assymmetric baseline. Since some fonts support multiple scripts, the safest
> thing would be to provide a font property with a value for each script. This
> is why there's a BASE table in Truetype Open, although default values can be
> calculated for older formats based on the script.

I realize that the baseline depends on the script, but I was hoping to
avoid discussion of the various scripts until later, since they add
quite a bit of complexity. The discussion over on www-style is
complicated enough, without even going into international issues. I was
hoping to take one step at a time. Thanks for the reply/reminder,

> If you had to choose one, it would be better to use Em Square rather than
> Bounding box. Bounding box means that two fonts of the same typeface  with
> different character sets might  align differently because one had Uppercase
> accents.

Sorry, I should have elaborated on the "inline box" thing. In CSS, each
line is in its own line box, and each line box can contain multiple
inline boxes, which correspond to inline elements (like <b>, <i>,
<font>, <tt>, etc). The text is supposed to be centered inside its own
inline box, but the inline boxes in a line box are aligned according to
the vertical-align property, whose default value is "baseline".

So the 2 fonts in your example would still have their baselines aligned.
Note that aligning the inline boxes vertically can and does cause the
line box to grow taller.

Received on Tuesday, 30 November 1999 23:55:42 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:37:30 UTC