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

Re: CSS3 @font-face / EOT Fonts - new compromise proposal

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Thu, 13 Nov 2008 12:54:58 -0600
Message-ID: <dd0fbad0811131054l77774a6m67fa1cca35f4d01b@mail.gmail.com>
To: "Adam Twardoch" <list.adam@twardoch.com>
Cc: "www-style@w3.org" <www-style@w3.org>, "Thomas Phinney" <tphinney@adobe.com>, "Mikko Rantalainen" <mikko.rantalainen@peda.net>
On Thu, Nov 13, 2008 at 10:49 AM, Adam Twardoch <list.adam@twardoch.com>wrote:

> Thomas Phinney wrote:
> > But I still consider this a side issue. Web designers want to be able to
> use retail fonts, and they want to be able to use most any font. Ergo, a
> solution which doesn't make font vendors happy won't make web designers
> happy either.
> I agree. If the consequence will be that instead of a handful fonts now
> (Verdana, Georgia, Trebuchet, Arial etc.) the web designers will be able
> to use maybe 3 or 4 times as much, tops (that is, only the ones that are
> freely licensed), then the whole issue is not worth the effort.

You say that, but there are many who would disagree.  Multiplying our
options 3 or 4-fold would be wonderful in my opinion.  As well, the fact is
that free fonts don't have much opportunity to make someone (relatively, of
course) famous right now, since someone has to download the font file to use
it.  Part of the power of the web is the ability to distribute a person's
effort widely with little effort, and a font-linking system that allowed
that would likely do wonders for the free-font industry, just as it did for

Even for alphabets/writing systems that do not yet have lots of fonts
> available, it is the Web that can contribute to the typographic
> development of those scripts  but IMO only if there are reasonable
> revenue opportunities associated with it. Right now, making Indic or
> Arabic fonts does not offer huge commercial prospects because of the
> widespread piracy. But web fonts, if done right, can offer some
> typographic talent in those countries some career paths and
> opportunities that they would otherwise not follow.

I severely doubt that offering people career paths in typography is a
commonly agreed upon goal of this effort.  If people can make a living off
of new opportunities that we open up, great.  But I feel no need to be
responsible for economic development.

This links up with my previous point even more strongly, of course.  Arabic
fonts, frex, don't have a lot of reason to exist right now, because they're
only really usable in print in the Arabic-speaking world.  You have to have
a computer which has downloaded the appropriate font to see Arabic text in
that font, and that's not very likely.  On the other hand, with appropriate
font-linking supported by all browsers, we could have an explosion of Arabic
typography on the web, as people in countries which primarily speak another
language (to be honest, I mean English-speaking countries) will actually be
able to see the beautiful fonts without having to go to any effort

Both my contention and yours are hypothetical, but I think that network
effects can be *very* powerful here.

This is the major issue that some of the "free software" proponents seem
> to forget: very often, there is no choice between so-called "will there
> be free software" and "will there be proprietary software". Instead, the
> choice is "will there be software at all" or "will there be no software
> at all".
> If there is a chance that a larger part of the human society worldwide
> can read and write their own language on the internet, all stakeholders
> should sort their priorities and look for a model that will give those
> people that chance under _any_ model.
> Where I come from, freedom for people should come first _before_ the
> freedom for other things (like software or capital).

I'm making the argument that even a free-fonts-only linking would accomplish
the goal of making minority languages (that is, languages that aren't
English, or at least don't use English script) much more prevalent on the

Of course, the issue isn't even as dire as you make it out to be.  Most
people have Arial Unicode installed on their systems, and so *can* read the
majority of scripts on the planet.  Among other things I shepherd a
translations program for my company's software, which has to display (among
others) Arabic, Chinese, Hebrew, and Vietnamese in the user's browser, and
just telling the browser to render the text in Arial does the job.  The most
essential job of making it *possible* to communicate in your native language
is done; all that's left is giving you more choice about how this will look.

Received on Thursday, 13 November 2008 18:59:13 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:41 UTC