W3C home > Mailing lists > Public > www-style@w3.org > January 2011

RE: [css3-font] Extension of font-stretch property

From: Koji Ishii <kojiishi@gluesoft.co.jp>
Date: Wed, 19 Jan 2011 15:55:28 -0500
To: John Daggett <jdaggett@mozilla.com>
CC: Brad Kemper <brad.kemper@gmail.com>, MasaFuji <masa@fuji.email.ne.jp>, www-style list <www-style@w3.org>, Thomas Phinney <tphinney@cal.berkeley.edu>, Ambrose LI <ambrose.li@gmail.com>, "www-font@w3.org" <www-font@w3.org>
Message-ID: <A592E245B36A8949BDB0A302B375FB4E0AAF009AAB@MAILR001.mail.lan>
> You're generalizing too far here, Morisawa has designed the font for a very specific *print*
> workflow, one that clearly involves rasterizer-level support for asymmetric scaling of font
> glyphs.

You're right, this font is a little special case to talk common scenario. I'm sorry to talk these two together.


> This is precisely my original point, asymmetric scaling of glyphs won't yield quality results
> for screen rendering unless the font rasterizer is involved.  I believe that's possible with
> CoreGraphics on OSX or FreeType on Linux but it's certainly not possible with GDI on
> Windows XP.
>
> In this vein, I should point out that it would be far more interesting to support some form
> of arbitrary font axes, such as Quickdraw GX variation axes or Adobe Multiple Master fonts.
> These formats allow a font designer to design specific parameterizations into the font data,
> similar to the way hinting data allows a font designer to add adjustments for display at
> specific resolutions. The Skia font that still ships with OSX supports 'weight' and 'width'
> axes, allowing a single font to support an arbitrary range of weights and widths.  While this
> isn't currently widely supported technology and would require a new version of OpenType,
> it has a lot of size benefits for web use.
> 
> With such a capability, an arbitrary axis such as 'contrast'
> could be supported for a family like Axis Mincho:
> 
>   http://www.typeproject.com/demo/axis_mincho.html

> 
> Note how the glyphs vary by weight (vertical axis) and contrast (horizontal axis).  Neat,
> eh?
> 
> Far better to have something like this that can unleash the creativity of font designers
> rather than try and perform imaging acrobatics with fonts simply not designed for a given
> effect.

Your points taken. I was talking about how often asymmetric scaling are used in printing, but I guess you already know that, and I have to admit that I didn't pay much attention on how on-screen quality differs from printing.

The information you gave me is so valuable, thank you. I'll investigate what would be the best way to bring those printing requirements to on-screen world. Not very soon though.


Regards,
Koji 
Received on Wednesday, 19 January 2011 20:54:05 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:36 GMT