Re: css3-fonts: should not dictate usage policy with respect to origin

On Sat, Jun 18, 2011 at 4:32 PM, John Hudson <tiro@tiro.com> wrote:

> Surely if a same origin mechanism of some kind is a Webfonts compliance
> requirement, then a UA such as you describe could not be compliant?


A UA implementer may implement WOFF, and be compliant with WOFF, without
adhering to an independent WebFonts Compliance specification; i.e., they can
implement WOFF and ignore the latter.

Actually, I find it quite odd for a W3C WG to be talking about "compliance",
since the W3C has, historically, not performed or mandated any form of
compliance, nor has it created or promoted a certification process by means
of which implementations may be determined to be compliant or not. On the
other hand, it has defined "conformance", which is distinct from
"compliance".

If you want to define formal conformance requirements, and you define two
types of conformance:

(1) WOFF with same-origin conformance
(2) WOFF without same-origin conformance

(and likewise for CSS3-FONTS) then I could accept such a position, though I
would continue to believe that the introduction of same-origin (in any form)
to WOFF or CSS3-FONTS is a violation of best practices with respect to
proper specification layering.

I would also suggest the WG refrain from using the word "compliance", since
I very much doubt that the W3C is going to start testing and certifying
compliant implementations of conformance requirements in products of any
sort (whether paid or free); though, if you know something new that I don't
know about the W3C's intentions in this regard, then I would be very
interested to learn about it.

G.