Re: [webfinger] Possibly speeding up the RFC process

["Paul E. Jones" <[e-mail]>]

Walter,

> > So, as a way forward, we have a couple of options:
> >    1) Allow the assumption to be made that the account for
> >       the above address is "acct:"
> >    2) Don't allow the assumption and require a query to
> >       "mailto" and discover the "acct" URI
> 
> I am not sure we have to standardize that specific behaviour. It would
> be impossible to do for all use cases with a normative language. That
> said it could be clearly mentioned in the examples to guide implementors.
> My preferred option for the example would be with mailto: (optionally
> mentioning that the answer could provide a proof of existence of the
> equivalent acct: uri through an acct rel.

My preference is that we leave the examples we have in the text as they are and encourage use of the "acct" URI if a query relates to a user's account.  Concrete recommendations might also come about through a little more real-world experience.  I suspect the first major service provider to support WF will dictate how it works.

Paul