Re: gs2 and qop

[Sam Hartman <hartmans-ietf@xxxxxxx>]

>>>>> "Simon" == Simon Josefsson <jas@xxxxxxxxxxx> writes:

    Simon> I don't understand why an authentication-only GS2 mechanism
    Simon> is any better than a "pure" SASL authentication-only
    Simon> mechanism.

It isn't but we want people to write gss mechanisms not sasl
mechanisms.  So, we'd like gss to be a superset of the functionality.
And we'd like all gss mechanisms to be sasl mechanisms.

    Simon> A simple question that may illustrate the situation: Will
    Simon> there ever by any more authentication-only SASL mechanisms
    Simon> on the standards track?

I hope not.

    Simon> If the answer is no, I'd argue that they shouldn't be
    Simon> permitted to be specified through GS2 either.

I don't expect there to be any authentication only gss mechanisms on
the standards track either.

But there are mechanisms not on the standards track.  And my point is
that as an individual I want a standards track solution for using any
gss mechanisms as sasl mechanisms--even if those mechanisms are not on
the standards track themselves.