Re: Including WOFF in ACID3

On Wed, Oct 13, 2010 at 10:59 AM, Håkon Wium Lie <howcome@opera.com> wrote:
> Also sprach Sylvain Galineau:
>
>  > The current test loads this font as a raw TTF only. It would seem
>  > appropriate for the test to be updated with a WOFF version of the
>  > test font i.e. instead of just:
>  >
>  >      @font-face { font-family: "AcidAhemTest"; src: url(font.ttf); }
>  >
>  > ...the test rule would be:
>  >
>  >      @font-face { font-family: "AcidAhemTest"; src: url(font.woff), url(font.ttf); }
>
> I don't support this change. I think there's a value to keeping things
> stable. Errors should be corrected, but -- in general, unless there
> are obvious reasons -- I don't think features should be added or
> removed.
>
> Aslo, by making the proposed change, it becomes possible to pass Acid3
> without supporting ttf. We could end up in a situation where browser x
> support ttf only and browser y support woff only, but both of them
> pass Acid3. As a result, interoperability would suffer.

At the moment interoperability suffers anyway, because the Ahem ttf
doesn't have the embedding bits set appropriately for IE9, so it
doesn't work there.

Acid3 isn't testing for ttf support; it *can't* be, because there's no
spec that requires it.  It's testing @font-face support.  Failing one
of the Acid3 tests even though you pass what it's *actually testing*
is kinda sucky.

If you have a problem with one browser passing due to support for ttf
and another passing due to support for woff, perhaps we can just drop
the ttf part of the test entirely and only test woff?

~TJ

Received on Wednesday, 13 October 2010 18:09:13 UTC