Re: [Curdle] Some work for the group

I don't much care from a CA point of view unless there is going to be
difficult getting the hardware we need to implement the signatures. Since
that hardware does not currently exist and since we would assume that any
suppliers would make sure they support the required mode, I doubt that will
be an issue.

There is a modest efficiency hit.

As far as making the system more robust goes, we are going to have to drop
SHA-2 if there is any significant compromise regardless of whether this
detail makes an exploit infeasible. It is far more expensive to explain why
faulty algorithms are safe than to switch to a safe one.

We are not shutting down use of SHA-1 because there is a real risk of a
bogus cert being generated by a competent CA. The security justification
such as it is is that the system should be fail safe and require at least
two failures before a catastrophe.

What the collision resistance property does provide is a cushion that makes
a compromise during the transition highly unlikely. And given the refusal
to consider issue of code points for SHA-3, we will probably need quite a
cushion.

On Thu, Dec 8, 2016 at 7:23 PM, Salz, Rich <rsalz@akamai.com> wrote:

> Hi there!
>
>
>
> We are looking for feedback on the following items, my memory spurred by
> the draft minutes found here:  https://datatracker.ietf.org/
> doc/minutes-97-curdle/
>
>
>
> Please review the minutes by the end of the month.  Speak up on the list
> if anything’s wrong.  Thanks to Ben Kaduk for taking them.
>
>
>
> Has anyone (besides Kyle Rose) reviewed curdle-rsa-sha2 ?  Are we ready
> for WGLC?  Please respond within a week.
>
>
>
> The text in draft-ietf-curdle-pkix says CA’s MUST NOT use pre-hash version
> of signatures.  Does anyone object to this?  There is a mention of the
> trade-offs in doing that at the end of section 5.  Please respond within a
> week.
>
>
>
> curdle-ssh-ext-info, curdle-ssh-kex-sha2, curdle-ssh-modp-dh-sha2
>
> Has anyone read these?  Will one or more people commit to doing so within
> a week?  (Commit, not actually read)
>
>
>
> Contexts – has anyone got a use for signature contexts? See the minutes.
> If you’re opposed to the (slight) consensus, speak up.
>
>
>
> Thanks.
>
>
>
> --
>
> Senior Architect, Akamai Technologies
>
> Member, OpenSSL Dev Team
>
> IM: richsalz@jabber.at Twitter: RichSalz
>
>
>
> _______________________________________________
> Curdle mailing list
> Curdle@ietf.org
> https://www.ietf.org/mailman/listinfo/curdle
>
>