2012-09-10 17:06, Jonathan N. Little wrote:
>> file:///C:/Users/Jukka/Documents/CSS3/font-size-adjust.html
>
> Gotcha! Better to point to an example online
D'oh! Happens to me every millennium! And I wasn't even trying to point
to my test page (its address had just crept into my clipboard) but to
the CSS3 Fonts draft on font-size-adjust:
http://www.w3.org/TR/css3-fonts/#font-size-adjust-prop
In addition to the practical considerations, which make the property
questionable, or maybe worse than useless, there's a theoretical note to
be made. The property revolves around the "aspect value", or "aspect
ratio" in more common parlance, i.e. the x-height of the font divided by
the font size. This means that we consider the relative height of
lowercase letters without ascenders or descenders. But that's not all!
What about those ascenders or descenders, about uppercase letters, and
about diacritic marks, which are frequent in many languages (Finnish,
French, Vietnamese for example)? They also affect the "real", visual
size of a font, as compared with the font-size property.
This reminds me of my favorite example about Verdana and line height:
set the font to Verdana, do not set line-height at all (i.e. let
browsers use their default for Verdana), and test with some text
containing loads of "g" and "Å". You will see how the "g" brutally
attacks the ring of "Å" on a line below it.
Of course, many other combinations of a letter with a descender and an
uppercase letter with a diacritic mark create similar effects, if you
default line-height when using Verdana. It's just so that "Å" is one of
my favorite letters, and a very common letter in my third-best language
Swedish, and a typographer's nightmare. It is virtually impossible to
design "Å" properly, but a font like Verdana has failed in a
particularly serious manner. It uses such a large part of the font
height for uppercase letters that any diacritic above them is bound to
cause damage, to the letter.
--
Yucca,
http://www.cs.tut.fi/~jkorpela/