W3C home > Mailing lists > Public > www-style@w3.org > April 2008

Re: My solution of webfonts problem

From: David Woolley <forums@david-woolley.me.uk>
Date: Sun, 20 Apr 2008 10:31:04 +0100
Message-ID: <480B0D58.8050404@david-woolley.me.uk>
To: www-style <www-style@w3.org>

Dave Crossland wrote:
> In both cases, using the one with the newest date or the largest
> number isn't useful, because it is not what the designer intended,
> which is always the web font, so that should always be the primary
> choice. I say the rationale for IE's approach is presumptuous.

Taking designer in a more general sense than visual artist, I think you 
will actually need to have mechanisms to signal the designer's intent.

One time, when I played with EOT fonts, it was to provide a subset of a 
Windows Chinese font that is an optional download for Windows 98 and and 
optional install for Windows XP.  I would have preferred the resident 
font to be used both for bandwidth reasons and to avoid the temporary 
display of missing glyph boxes (what to do before a font loads hasn't 
been discussed here).

In the end, I compromised, by renaming the subset font and making it 
second choice.  If I remember correctly, I didn't save on bandwidth, but 
I did get fast rendering on systems that did have the font.

(There was another time when I thought that appearance was much more 
important, and I did make a downloaded Arphic font the first choice.  I 
would still have been quite happy to use a pre-installed one, but didn't 
  think the likelihood of it being pre-installed sufficient to justify 
using the above tactics.)

A more difficult variation of this might be trying to serve Euro's to 
Windows 98 without upgraded fonts.  In that case, you might want the 
resident font to be used where possible and be backfilled with the 
additional characters from the download.  (I use this because it is a 
case where I know that a resident font got upgraded with additional 
characters, however it also true that Windows 98 is still in use.)

Even when the difference is only aesthetic, designers, in the general 
sense, need to consider rendering time and bandwidth, so may well prefer 
to use a resident even if it differs slightly from the version that they 
would download.  If exact reproduction is essential, HTML/CSS is not the 
optimum tool.

With a proper set of hints, users should be able to override them so as 
prefer faster download, or to prefer more faithful rendition.

David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
Received on Sunday, 20 April 2008 09:31:54 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:35 UTC