Re: [CHANNEL-BINDING] Re: draft-ietf-sasl-gs2 AD review comments

[Nicolas Williams <Nicolas.Williams@xxxxxxx>]

On Mon, Oct 22, 2007 at 07:26:04PM -0400, Sam Hartman wrote:
> I just had a quick phone call with Nico.
> 
> 
> He's still been thinking about this from the API standpoint.  I was
> asking him why we wanted to support separate slots in the protocol for
> channel binding type and channel binding data.I didn't understand the
> complexity.  During the conversation it became clear that Nico
> believed that at the end of the day you want to end up with a channel
> binding type, a colon and some stuff.  I like that too.  I don't care
> how it works in the API at all.
> 
> 
> I propose  we accomplish this by adding the following requirement:
> 
> "Under this framework, channel bindings MUST start with the channel
> binding unique prefix followed by a colon (ASCII 0x3A).
> "

I second this.  Note: Sam's text should be added to either the third
bullet item in page 7, or as a separate item below it.

NOTE: draft-williams-on-channel-binding is now in AUTH48.

Nico
--