Re: [security-services] disposition of the HoK profiles

[Tom Scavo <trscavo@gmail.com>]

On Sun, Jul 5, 2009 at 1:39 PM, Scott Cantor<cantor.2@osu.edu> wrote:
> Tom Scavo wrote on 2009-07-01:
>> I may be missing something but this seems like the best way to get the
>> HoK Web Browser SSO Profile out the door now without having to
>> implicitly depend on the successful completion of the kerberos profiles.
>>
>> Thoughts?
>
> My only comment is that I don't think we've had many, if any, specific
> examples of specs that we've "revised" following a CS or OS vote. The
> exception might be the errata, which is now headed to its second published
> revision, but errata are by their nature accumulative.
>
> It becomes a "signaling" problem at runtime if you have behavior that
> changes between V1 and V2 of some spec that's sharing syntax.

I guess I don't see the problem.  The HoK Web Browser SSO Profile
normatively depends on the HoK Assertion Profile.  Currently, CD-02 of
the former depends on CD-02 of the latter.  If we take both of them to
CS at the same time, then CS-01 of one depends on CS-01 of the other.
So far so good.

Now suppose we take CS-01 of the HoK Assertion Profile back through
the review cycle.  It is anticipated that this will produce CS-02 of
the HoK Assertion Profile, but that work is totally independent of
CS-01 of the HoK Web Browser SSO Profile.  Even if the latter becomes
an OASIS standard, it can remain dependent on CS-01 of the HoK
Assertion Profile since there's nothing in CS-02 of the latter that
has any effect on the HoK Web Browser SSO Profile.

I could be missing something, I suppose, but I don't see how we could
get into trouble by pushing both HoK profiles to CS at the same time.

Tom