Re: [webfinger] Server Response language

The 404 bit is needed, since the "webfinger" server was found. just not the
resource being queried.  That question absolutely will come up.

The new stuff (401, 2xx), I agree: it's re-stating what HTTP does.

If others agree, I'll not put that into the spec.

Paul

From: Tim Bray [mailto:tbray@textuality.com] 
Sent: Thursday, December 20, 2012 12:30 PM
To: Paul E. Jones
Cc: webfinger@ietf.org; webfinger@googlegroups.com
Subject: Re: [webfinger] Server Response language

As in every other case where the WebFinger spec is merely re-iterating
standard HTTP rules, I suggest just removing this language. -Tim

On Thu, Dec 20, 2012 at 8:28 AM, Paul E. Jones <paulej@packetizer.com>
wrote:

Folks,

We had this previously:

"If the client queries the WebFinger server and provides a URI for which the
server has no information, the server MUST return a 404 status code."

Someone posted to the list that we should talk about positive replies and
mention that a client might be rejected with a 401.  So, I wrote this text
to be appended to the end of that above paragraph:

"If the server is able to provide information in response to a request, it
MUST do so using an appropriate 2xx HTTP status code and including the
requested representation in the body of the response.  A server MAY also
return other HTTP status codes, as appropriate, such as a 401 to indicate
that the client is not authorized to issue a request to the server."

Is this agreeable?  Please suggest wording changes, if not.

Paul

_______________________________________________
webfinger mailing list
webfinger@ietf.org
https://www.ietf.org/mailman/listinfo/webfinger