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

Re: A way forward

From: John Hudson <tiro@tiro.com>
Date: Mon, 27 Jul 2009 12:16:58 -0700
Message-ID: <4A6DFD2A.3090105@tiro.com>
CC: www-font@w3.org
Dave Crossland wrote:

> Any foundry that signs up to TypeKit or similar demonstrates that,
> although they might prefer something else, raw TTFs are acceptable to
> them. They might even SAY that they won't allow raw TTF web fonts, but
> (as I said to TP) I fail to see the distinction between raw fonts
> served directly or by a service provider in the context of format
> discussions such as these.

There is no relevance at all of services to the discussion here, because 
here we are talking about standard interoperable formats. The 
suggestions that signing up for a service such as Typekit implies 
acceptance of naked font linking is disingenuous, because part of the 
service offered by Typekit to foundries is precisely to obfuscate the 
font data in such a way that the font is not nakedly exposed. Like the 
Typotheque service, Typekit currently provides an interim solution to 
the format problem using methods the might be acceptable to some font 
foundries *insofar as these methods address concerns about the exposure 
of font data*.

Put in other terms, Typekit tries to apply on the server side something 
as good or better than the level of simple protections that we want to 
see as part of the format. This is similar to the distinction of 
server-side compression vs. format compression, with the difference that 
server-side compression does not rely on a third party commercial service.

Again, I believe these services and what they are doing today are 
irrelevant to the discussion of an interoperable standard format, except 
insofar as they too have expressed a desire for a single format that is 
acceptable to font foundries, since this will reduce their own overheads 
by removing the need to serve multiple formats and take extra steps to 
try to protect naked font data. Such services may choose to continue to 
apply more protections than the minimal obfuscations of the proposed EOT 
Lite, .webfont or ZOT formats, and that will be a means of competition 
between them in attracting foundries to sign-up for their services.

John Hudson
Received on Monday, 27 July 2009 19:17:40 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:01:40 UTC