RE: Minutes, 16 February 2011 WebFonts WG telcon

[David Crossland:]
> "The web wants to be inconsistent with the default [state]" is illogical
> statement, since a default is not a consistent state but merely an initial
> one that can by definition change. The web wants to change the state of
> SOR often enough that we want a well-defined mechanism to allow that, and
> that mechanism is FO. If a site wants to set FO:same for everything,
> that's cool!

It's no more illogical than claiming every new resource type should be
consistent with all those that came before even though many of the 
latter are being banned from loading cross-domain using a number of 
techniques. That consistency is only an initial state is *exactly* my 
point. When it is being claimed that 'the web' is 'cleaner' and more 
'consistent' when all resources default to a common load policy, we are 
only talking about the consistency of browsers' code base. Not that of 
the real web.

 
> But defaults are a very different issue, and so I understand why the
> default for @font-face is so worrisome.
> 
> Nomen est omen, and _the web_ has an interlinking spiders-web nature at
> its core, which is teased undone by default same origin restriction. But
> the ability to cut yourself off from society is a right worth respecting
> deeply.

Absolutely agree. But that ought to be followed to its logical conclusion.

> 
> So I expect FO to be an uncontroversial and quick W3C REC, because
> Referrer: checking sucks and supporting FO for SOR for all resources types
> not just font formats will help make the web a better place. I hope FO
> will not need the help of the resources available to this WG and that it
> will happen fine by itself! :-)

So you believe that the web's 'interlinking spiders-web nature' to be 'undone
by default same origin restriction' but you also expect that a mechanism that
will make it easier to undo this initial, precious  interlinking state to make
the web a 'better place'. I find the reasoning confusing as it implies the
initial state is what makes the web great, irrespective of its runtime state.

If that were true then, ad absurdum, all servers could flip the From-Origin:same
switch on every URL one morning and the web would be a better place as a result ? 



> 
> If WOFF comes out with FO before FO is ready, we'll see if font publishers
> are really gating on SOR or not. I expect only a few will hold out,
> proving that it is not that important, really, to most of them, but it is
> worth having instead of not having, which seems to be the unspoken
> alternative.

I don't know and I don't really care. The vast majority of font licenses
today would require the FO header to be set to the same dumb static value 
by anyone who licenses a font under one of those licenses. I don't see why
tens of thousands of people should have to handle manually what browsers can
trivially set with a few lines of code. Given that sites can override this
default, it need not me a burden.

Received on Sunday, 20 February 2011 17:08:25 UTC