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

RE: EOT & DMCA concerns

From: Sylvain Galineau <sylvaing@microsoft.com>
Date: Wed, 5 Aug 2009 18:25:19 +0000
To: Jonathan Kew <jonathan@jfkew.plus.com>
CC: John Daggett <jdaggett@mozilla.com>, www-font <www-font@w3.org>
Message-ID: <045A765940533D4CA4933A4A7E32597E021637EB@TK5EX14MBXC111.redmond.corp.microsoft.com>
>From: Jonathan Kew [mailto:jonathan@jfkew.plus.com]
>Sent: Wednesday, August 05, 2009 10:10 AM


>If everyone implements EOTL now, then authors will naturally deploy
>EOTL fonts, because they'll work everywhere. That means that even if
>some of us implement another "better" format as well, authors are
>unlikely to bother deploying it because it won't work in legacy IE (or
>even current IE for a couple of years, perhaps). So the new format
>will not get any significant level of testing in the marketplace; and
>because EOTL will be the "universal" format in actual use on the web,
>there'll be little incentive for slower-moving vendors to bother
>supporting the new format at all.

If lack of support in legacy IE is enough to make authors not bother
deploying a new format then surely that will be true for .webfont or ZOT
as well.

>On the other hand, if we define a web font format that provides
>tangible benefits (such as compression to reduce bandwidth
>requirements, and improved metadata support that many vendors would
>like), and do *not* rush to implement EOTL, there's a strong incentive
>for all parties to work towards support of the new format. Meanwhile,
>authors who wish to support web fonts on legacy IE browsers will need
>to deploy two formats for some time -- how long depends on Microsoft's
>business decisions, and on how slowly users upgrade, and on how
>concerned authors are about getting their chosen fonts to display in
>obsolete browsers. Is that such an unreasonable scenario?

First, I must note that your argument so far implicitly assumes EOTL has succeeded
i.e. it is both widely deployed and used. Which would arguably be far better
for authors and web typography than the current status quo. I find it encouraging you'd
assume this could in fact happen and will candidly state this is a problem I'd love to
have to confront in practice; I prefer struggling with the aftermath of success rather
than the alternative but maybe that's just me :)

Overall, I think I agree that if EOTL is widely used then this will raise the bar for
the benefits any new format must provide in order to gain traction. But I'm not sure I
understand how this argues against deploying EOTL or any other format, either of which
may need improvement someday. If the incentives inherent to the new proposals
do not provide clear enough benefits to all parties involved - browser vendors, font
creators and authors - over adding 100 bytes of well-defined header data to a raw font file,
then I must wonder why we should dump the latter in favor of something that will
ultimately take much more time and effort.

If we are confident in the merits of .webfont and ZOT then they can and will succeed on
those merits given time and interoperable implementations regardless of what is already
there.

>I designed the .zot format, implemented sfnt2zot and zot2sfnt tools,
>and patched Gecko to support .zot fonts in less time than I've spent
>reading www-font messages in the last few weeks. It's not complex. And
>you're welcome to use the code. :)

But I'm willing to bet that still took a lot more code than EOTL...

>(Not that I'm asking for that prototype .zot to be the final answer;
>it was merely a proof-of-concept, and I expect it to be adapted in the
>light of feedback from other parties, if we decide to go in that
>direction at all.)

Understood.

>Perhaps if those corporate/government 'seats' are so slow to upgrade,
>they might also have relatively little interest in an explosion of
>downloaded font use, and be happy for the stability of continuing to
>see everything in Verdana et al.

The IT policies of these accounts should not imply that they have
no interest in using features their software already has.

>And those who upgrade at a very slow pace should not expect to see the
>latest glitzy web features in their browser.

Web fonts may be the latest glitzy feature in Firefox but EOT is a bit
older than that. I don't see why an IE6 or IE7 or IE8 user should not expect
to see headlines and text in the same font as in Firefox.

But whether these users should be left behind or not is really up to authors,
and those authors who choose to support them will appreciate doing so
at lower cost if that's possible.

>Fine. That's their choice. Then why are we trying so hard to deploy
>new fonts to them?

We don't have to 'try'. Their browser already supports web fonts. It already
did a decade ago. Why not use it if we can do so at low cost and effort ?
Received on Wednesday, 5 August 2009 18:26:05 GMT

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