Re: [webfinger] Revised WebFinger spec (draft -15)

[Barry Leiba <[e-mail]>]

> Having reviewed the thread below and taken other input into consideration, I
> revised the WebFinger document and it now appears as draft -15.

Great.  How do you think this version applies to the DISCUSS positions
from Pete and Stephen?  What are the next steps?

Barry

> Here's a summary of changes and responses on some comments made:
>
> * Most significantly, I removed all of the contentious examples.  I kept
>   the OpenID Connect example and brought over an "http" example from RFC
> 6415.
>   Neither should be contentious, since the first is a concrete example
>   from OpenID and the second is an example that existed already in an RFC.
>   No example email auto config, device: URI, or locating a blog given an
> email
>   Address.
>
> * I did not change the language to say that WF is a framework.  It is a
> protocol
>   with a well-defined syntax, not unlike HTTP. What one requests and what
>   payload, like HTTP, is really outside the scope of the spec.  That said,
>   Section 1 states that applications that desire to use WF must specify
>   "properties, titles, and link relation types that are appropriate for the
>   application".  Is this sufficient to address this concern?
>
> * There was a suggestion to define a "rel" registry.  Such a registry
>   already exists: http://www.iana.org/assignments/link-relations/
>
> * Clarified that some server responses may have either an empty links array
>   or no links array.  (No change in behavior, but this clarification was
>   requested.)
>
> * I removed text in section 4.5 that recommends the use of the acct URI
>   scheme.  Applications / usage specifications will indicate what scheme
>   to use and we do not need to dance around this issue in the protocol spec.
>
> * A few minor edits (e.g., references and word choice)
>
> You can find the current text here:
> http://tools.ietf.org/html/draft-ietf-appsawg-webfinger-15
>
> Paul