On 11/28/06, Rogers Cadenhead <[EMAIL PROTECTED]> wrote:
I have no idea how to gauge the likelihood they'll achieve it. Or whether they'll respect current autodiscovery functionality in MSIE 7 and Firefox 2.0.
My experience is that the IETF is essentially unresponsive to backward compatibility concerns, to the extent that they will debate the meaning of the term "MUST". It's like existential poetry. In my experience, the WHAT-WG doesn't make any changes that break compatibility unless users are already having problems caused by implementation divergence. To make sure this is true, they research and make decisions based on metrics, rather than ill-defined consensus handwaving and individually-authored drafts with no support from client implementors. I've even seen it claimed that "servers are clients too", in the IETF.
> Why not provide feedback there? I will if that's where autodiscovery goes.
It is already there.
Seems like a good idea to tell Atom publishers how to support autodiscovered feeds.
They already know how, in general. The WHAT-WG is the place to work out edge cases in HTML semantics. Sylvain Hellegouarch wrote:
there will be little harm done
Actually, the proposal seems so poorly researched and poorly coordinated with WHAT-WG that I don't see how you can make that claim. When Pilgrim wrote the draft, there weren't as many existing implementations, so his approach made more sense at the time. -- Robert Sayre
