RE: Next step?

From: [] On Behalf Of Robert O'Callahan
Sent: Wednesday, October 21, 2009 4:13 PM
To: Sylvain Galineau
Cc: Ricardo Esteves;
Subject: Re: Next step?

On Thu, Oct 22, 2009 at 11:59 AM, Sylvain Galineau <<>> wrote:
As for choosing a single format, this will not ensure interop until a large share
if the browsers used in the field support it and is thus of limited practical use
for real world authors.

Dealing with the legacy IE users is a separate problem that the Fonts WG can do nothing about. The best thing the Fonts WG can do is recommend a single font format that browsers should support going forward. Then authors can use that and, if they wish, support legacy IE in some other manner.

Asserting it is the best thing does not make it so. The best thing for whom ? For authors, it would arguably be a solution that lets them use the fonts they want across all browsers in the field today. Will there be browser bugs to deal with ? Sure. (And not just with IE and EOT, by the way). But I don’t see why the existence of such bugs should close the door on specifying an *optional* format compatible with 2/3 of the installed base. Some also believe the best thing for HTML5 would be to define what HTML should ideally be regardless of what’s already out there but that would not be very useful either.

The legacy is out there whether we like it or not. Specifying a simple, lightweight – and, again, optional – subset format that is both compatible with the majority of web browsers and can implemented by others if needed/wanted without any patent and other issues is well within the remit of a W3C working group and an interoperability charter.

I’m fine with individual browser vendors deciding they will pursue the ideal future right now at the expense of the short to medium-term needs of today’s real world authors. But that should be their own explicit choice, not the W3C’s. You may have the option of not dealing with the legacy. Most web authors do not have that luxury. And I assume the W3C cares about them.

Received on Wednesday, 21 October 2009 23:49:16 UTC