W3C home > Mailing lists > Public > www-font@w3.org > July to September 2009

Re: cutting to the chase

From: Chris Fynn <cfynn@gmx.net>
Date: Mon, 03 Aug 2009 13:01:42 +0600
Message-ID: <4A768B56.3040604@gmx.net>
To: www-font <www-font@w3.org>
CC: John Daggett <jdaggett@mozilla.com>
John Daggett wrote:

> John Hudson wrote:

>>> TTF/OTF is widely implemented and used.
>> It is reasonably widely implemented. It is hardly used at all.

>> EOT is more widely implemented, in terms of the number of active
>> browsers being used, and probably still more widely used in terms of
>> the number of sites making use of EOTs (especially in countries whose
>> scripts have not been well-supported by the availability of 'web safe'
>> fonts).

> I'm assuming the last line means "India".  The usage of EOT in Indian
> has more to do with making the best of poor situation than anything that
> should be suggested as an example of ideal usage.  My understanding is
> that most of these EOT fonts use custom encoding schemes made up of
> precomposed glyph sequences. These are used because the structure of
> OpenType requires some level of OS support for complex scripts (as
> opposed to AAT which embeds all the information in the font itself) and
> that OS-level support is lacking in older versions of Windows.  As
> OS-level support improves, the need for burdensome hacks like this will
> decline.

Windows 2000 supported two of the most widely used Indic scripts 
Devanagari & Tamil ~ and it is fairly easy to update Windows 2000 to 
support additional complex scripts.

Out of the box, Windows XP supported several more Indic scripts than 
Windows 2000 did ~ but it is somewhat more difficult to update to 
support more than those.

People in India, especially in smaller centres, will run Windows 2000 
and XP systems until the hardware fails and is *completely* beyond 
repair. I'm told the situation is similar in parts of China etc.

> The main competition for TrueType usage is not EOT, it's sIFR and other
> Flash-based techniques, those have far more traction in the minds of web
> authors.  Right now there's a lot more fonts that allow sIFR usage but
> don't allow direct linking.  The current EOT font market is a far
> smaller, niche market at best.

sIFR and similar are *not* well suited for body text - in addition I've 
observed opening pages using Flash in Firefox on some older systems 
frequently leads to crashes.

> The existing use of EOT is not relevant to the EOT-Lite proposal.

I'm not quite sure why not - but anyway supporting users of non-latin 
scripts in developing countries as best as possible with the 
hardware/software systems they have available should be a concern of 
W3C. Since many of these users are using older versions of Windows and 
IE - and will be for quite some time to come -  a solution that is 
compatible with those systems would be helpful.

- CF
Received on Monday, 3 August 2009 07:02:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 11 June 2011 00:14:03 GMT