W3C home > Mailing lists > Public > public-webfonts-wg@w3.org > February 2011

RE: Minutes, 16 February 2011 WebFonts WG telcon

From: Sylvain Galineau <sylvaing@microsoft.com>
Date: Sun, 20 Feb 2011 23:21:55 +0000
To: John Hudson <tiro@tiro.com>
CC: "public-webfonts-wg@w3.org" <public-webfonts-wg@w3.org>
Message-ID: <045A765940533D4CA4933A4A7E32597E2AB62D3A@TK5EX14MBXC120.redmond.corp.microsoft.com>
> My point re. the toin coss is that we don't seem to be progressing any
> further in this discussion than the initial positions established during
> the teleconference. As Sylvain pointed out then, there is good pragmatic
> reasoning on both side of the debate, and we don't seem to be moving
> anyone to significantly alter their opening positions.

We've add a good discussion with Maciej, established a proposal he finds
better than the current CORS-based solution in IE and Firefox so it was
worth trying. But the core issue remains the default for fonts. As I don't 
think it can be bridged at this point, the current state of cross-browser
SOR incompatibility will stand. This means authors have nothing to do for
IE and Firefox, but Referer work to do for everyone else. It's unfortunate
but if some implementors believe putting this burden on authors - or the 
equivalent with some other header - is good for the web then it's probably
all we'll get today. 

> Where we have had significant movement is away from CORS as a mechanism,
> which presumably makes Anne and Håkon happy. 

It'll remain in use for this purpose in Firefox 4 and IE9, at least. On
the IE side, we have no plans to disable or remove this in future releases 
as of yet.

I don't think moving off CORS was really the issue. Same-origin by default 
has always been the friction point. Broadening the discussion to the reso-
-lution of the general same-origin control problem conveniently allows for 
the re-opening of the 'fonts should be like any other file' argument, except
 in the service of so-called consistency. So we went from 'fuck the foundries' 
to 'we know what's good for the web'. It's progress of a kind, but the 
substance and outcome are essentially the same. 

> It makes me happy too in that I think FO looks like a better overall 
> mechanism. 

There is nothing better about it for fonts imo. It just happens to solve
another more general problem for other resource types but in a manner
that is incompatible with current implementations and the requirements
agreed when this WG was set up.

> This move has implications for the schedule of WOFF standardisation, 
> and I think our time now would be better spent working out how to 
> administratively minimise the delay involved in spec'ing FO, finding 
> a home for it, and getting the CSS3 font module to reference it. 

I don't really see any reason to believe this can be specified so quickly. 
Given that Anne is about to leave on a trip for three months I suspect 
little will happen until he's back, at the least.

Unless someone else is available to pick this up and run it ?

> As I understand it, there is general agreement in our group that the 
> default interpretation of no-FO header for fonts will be @font-face 
> rather than WOFF specific, so is ultimately something that we'll only 
> be referencing in the WOFF spec. Yes? As the web fonts working group, 
> I think we can and should present a clearly worded recommendation on 
> this default interpretation, but the debates we are having here will 
> doubtless also take place in CSS circles.

And the exact same people will argue against it in the CSSWG. They'll 
probably also point out that as their current implementations load
raw fonts and SVG fonts from anywhere, applying this new requirement 
on all formats will break some web sites.

I am comfortable sticking to same-origin by default. Whether that will
remain conformant and/or interoperable (and with whom) is, at this point,
Received on Sunday, 20 February 2011 23:22:32 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:34:15 UTC