Re: Font-face Descriptors for Matching
On Aug 5, 2:50pm, Clive Bruton wrote:
> Is this part of the CSS1 spec, or am I on the wrong list?
It is part of CSS. It isn't part of CSS1. You are on the right list
to discuss it.
> After reading "http://www.w3.org/TR/WD-font" I have some comments:
> Items 4.5.2-8 seem derived from the font descriptors of Adobe Acrobat/ATM
That was one place that uses a similar set. There are others, including HP's
font synthesis system whose name temporarily escapes me.
> (I therefore wasn't surprised to see Adobe staff on the authors list).
> Can I assume that these descriptors are to be interpreted by ATM, using
> the Adobe Sans/Serif Multiple Masters as their base?
They could be. They are not tied to a particular technology. ATM is not
required. Applications that used ATM to do the synthesis would be a
good idea, though.
> If so I'd suggest that the full range of available Descriptors is used,
> rather than this subset. Some of the missing descriptors include flags
> for italic (rather than "slope"),
Not all italic fonts slope. Not all sloped fonts are italic. The WebFonts
'slope' descriptor describes the slope angle, regardless of whether the
font is italic, cursiv, oblique, or just happens to slope a bit. There is
also a 'font-style' descriptor which indicates whether a font is "normal",
italic, or oblique.
We don't have much support for script fonts as a specific type, true.
They can be indicated by setting the first number in the panose-1
descriptor to 3
> sans or serif,
For Latin Text and Display (first digit of panose-1 descriptor is 2)
fonts,the presence and type of the serifs is indicated with the second
digit of the panose-1 descriptor.
> non-roman fonts,
A large category ;-) and a single flag for "non-roman" would be of little
use. There is a flag to indicate a Latin script:
and similarly to indicate that a font has (some) glyphs to cover Greek,
Cyrillic, Devanagari, Klingon, or whatever.
> small caps,
The font-variant descriptor does that
> all caps or
All-caps fonts are not specifically provided for.
> symbol fonts.
For symbol fonts which contain glyphs used for rendering Unicode
characters, the unicode-range descriptor can be used. Other pi fonts
can be allocated to the private use area. Note: it is an error to apply
symbol fonts to Latin characters.
> May I also suggest that provision is made for the synthesis to be
> generated from fonts other than the common Adobe Multiple Masters as
> supplied with ATM/Acrobat. Should anyone be generous enough to donate
> such to the global font pool.
Synthesis from any Multiple Masters you have handy (or any other mutable
font technology such as OpenType or TrueType GX) is neither precluded
nor explicitly provided for in the spec, which is independent of the font
Thanks for the feedback.
Chris Lilley, W3C [ http://www.w3.org/ ]
Graphics and Fonts Guy The World Wide Web Consortium
http://www.w3.org/people/chris/ INRIA, Projet W3C
firstname.lastname@example.org 2004 Rt des Lucioles / BP 93
+33 (0)4 93 65 79 87 06902 Sophia Antipolis Cedex, France