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

Re: position of baseline relative to em square

From: David Lemon <typenerd@slip.net>
Date: Wed, 1 Dec 1999 08:54:29 -0800
Message-Id: <l03130301b46afd95de0f@[209.209.14.236]>
To: <www-font@w3.org>
Cc: <erik@netscape.com>
At 9:18 PM -0800 11/30/99, Nick Nussbaum wrote:
> Shouldn't it include extra leading ?
> I can't check right now because I'm home offline but is the extra leading
> considered part of the
> the baseline to baseline?

There are two different functions here. The first is the historic one, in
which the em square is a function of the type size, and the designer fits
the font into the em square at a proportion appropriate to its use. This is
the reason that most historical designs are somewhat shorter then the em
height from ascender top to descender bottom. The second is a recommended
additional leading value (reflecting the fact that most text has something
around 20% extra leading added in the setting). But this external leading
is not part of the em square. The em height is equivalent to
baseline-to-baseline distance when set with no added leading ("solid" in
metal terminology).

What I referred to earlier is that Adobe specifies the typographic ascent
and descent such that their absolute values sum to the em height (thus
getting the em's position relative to the baseline into the font
information). We believe this is in line with the original intent of the
TrueType typographic ascender and descender values, and in the next version
of the OpenType specification we hope to explicitly recommend this practice
for all OpenType fonts. Since these values have been rather loosely
specified to date, other fonts may only approximate this effect. As Greg
Hitchcock suggested, one could take the sum of the ascent and descent
values, determine the difference from the em value (zero in Adobe fonts),
and add half that difference to each to get a pretty good placement of the
em.

- David Lemon
Received on Wednesday, 1 December 1999 11:56:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 11 June 2011 00:14:00 GMT