Re: .webfont Proposal 2

karsten luecke wrote:

> John Daggett wrote:
>> Without defining user agent behavior, it's hard to evaluate this
>> format.
> 
> Håkon Wium Lie wrote:
>> Also sprach John Daggett:
>>>> - The <allow> element would list domains that are licensed to
>>>> use the font. A meta URL, "any", would signify that the font
>>>> could be used on all domains.[1]
>>> This is a root string proposal in another form and suffers all
>>> the same problems
>> Agreed. I do not think we will find consensus around a format with
>> root strings. 
> 
> What is not clear to me is: Are you objecting to the format itself (xml,
> vs binary as with EOT)? Or do you accept the format and are merely
> discussing metadata pieces?

I'm concerned about what behavior is implied by the contents of the
metadata, the format doesn't matter so much to me as long as it's simple
and efficient.

Is a user agent required to treat the contents of the <allow> element
the same way EOT root strings are supposed to be handled?  If it is, I
would object, otherwise I'm quite happy with a format that includes
metadata that indicates any manner of licensing information, including
root strings. It would be nice if it was simple and easy to understand
by mere mortals, but that's just a wish not a requirement.

> As to user agent behavior: IF root strings have been defined, of course
> this MUST be respected by the browser.

This is precisely what has been objected to in the past.

Note that both the original Ascender proposal and what they referred to
as EOT-Lite removed the requirement for user agents to enforce root
string restrictions.

John Daggett
Mozilla Japan

Received on Friday, 17 July 2009 19:02:42 UTC