Re: Minutes, 16 February 2011 WebFonts WG telcon

On Feb 17, 2011, at 11:49 AM, Levantovsky, Vladimir wrote:

> 
> On Wednesday, February 16, 2011 11:55 PM Maciej Stachowiak wrote:
>> 
>> I believe there are three separate dimensions to the From-Origin
>> proposal:
>> 
>> (1) Define any embedding restrictions related to fonts as part of the
>> CSS3 Fonts spec (the place where @font-face is defined) instead of as
>> part of the WOFF file format spec, so the rules apply consistently to
>> all fonts.
>> 
>> (2) Instead of using CORS headers to change the defaults for allowing
>> font embedding, use a proposed new mechanism for limiting hotlinking
>> (From-Origin) that can apply to any resource type.
>> 
>> (3) Change the default to be that cross-site font embedding is allowed
>> (as opposed to presuming "From-Origin: same" in the absence of a From-
>> Origin header for @font-face).
>> 
>> I think that on the call, we had rough consensus on #1 and #2. Everyone
>> seemed to either think these are improvements, or was indifferent to
>> the outcome. #3 still  seems controversial. Most of those who spoke up
>> on today's telecon believed that the default should still be to forbid
>> cross-site font embedding by default. I personally disagreed and
>> thought it was better to make fonts consistent with other resource
>> types.
>> 
>> If we do indeed have consensus on #1 and #2, I think that would greatly
>> reduce the scope of remaining disagreement.
>> 
> 
> I think that we also need to define the near-term action plan to finalize the promotion of the Last Call to CR.
> 
> Since the technology under consideration in #2 ("From Origin") hasn't been even drafted yet, and the consensus on #1 involves the broader agreement with CSS WG, and taking into account that the Web Fonts WG decision to revisit the issues related to SOR and CORS are made as a good faith effort that is not driven by any of the official Last Call comments - I would suggest to finalize the editorial work related to LC comments and promote the spec to CR, possibly with the additional note that the WG is going to discuss the issues related to general requirements with CSS WG and other interested parties.
> 
> As you know, the WOFF spec has already been referenced by the EPUB 3.0 Public Draft (http://epub-revision.googlecode.com/svn/trunk/build/spec/epub30-contentdocs.html#sec-css-fonts), which is planned to be finalized in May this year. The sooner we go to CR, the easier if will be for the EPUB WG to accomplish their goals.

Making these changes would almost certainly require dropping back out of CR. So I don't think it will speed up the process overall to go to CR without them, if we are agreed that these changes are the right thing to do. If we make the changes to WOFF now, then WOFF could proceed all the way to REC while we work out the rest of the details, since there would be no actual dependency from WOFF to either From-Origin or CSS3 Fonts. It's actually failing to take action promptly that will create delay.

Regards,
Maciej

Received on Thursday, 17 February 2011 20:48:38 UTC