Re: [TLS] SHA-3 in SignatureScheme

[Yoav Nir <[e-mail]>]

> On 5 Sep 2016, at 11:17 AM, Nikos Mavrogiannopoulos <[e-mail]> wrote:
> 
> On Fri, 2016-09-02 at 10:04 -0700, Eric Rescorla wrote:
> 
>>>> I also am not following why we need to do this now. The reason we
>>> defined SHA-2 in
>>>> a new RFC was because (a) SHA-1 was looking weak and (b) we had
>>> to make significant
>>>> changes to TLS to allow the use of SHA-2. This does not seem to
>>> be that case.
>>> 
>>> I don't think we strictly _need_ to do this now, however I think
>>> it's a good idea given that we'll need to do it eventually 
>> 
>> I'm not sure that that's true.
> 
> It is unclear to me what is the intention. Due to the semantics of the
> signatureAlgorithms extension in TLS 1.3, if the TLS 1.3 draft doesn't
> define SHA3, it effectively _bans_ the usage of SHA3 in all certificate
> chains intended to be used by TLS 1.3. If that's the intention then
> yes, SHA3 should not be included.

I don’t think that’s anyone’s intention.

> In that case implementations of TLS 1.3 will have to wait for a SHA3
> RFC to be published in order to enable the algorithm. That would
> introduce a delay, and in certain occasions (e.g., firmware) we will
> have TLS 1.3 implementations which may never support SHA3.

Not necessarily. If you write a SHA-3 document now, and it is simple as such documents tend to be, it could go through the last calls before the TLS 1.3 document. If it makes a reference to the TLS 1.3 document it will end up waiting for TLS 1.3 in the RFC editor’s queue.

The upside is that this adds support for SHA-3 certificates even for TLS 1.2 implementations, which makes sense to me. If there are SHA-3 certificates out there, their use and deployment should not depend on TLS 1.3 implementations.

> IMO, unless there are doubts about SHA3's adoption as a certificate
> algorithm, it should be part of the TLS 1.3 spec.

I have doubts. If you’re a CA and you’re going to issue a certificate, are you going to use SHA-2 that is supported by nearly everything, or SHA-3 that is supported by nothing. How about in 5 years when SHA-3 is supported by all updated browsers, but not by old browsers and dozens of different programmatic clients?

Besides, maybe we’ll all be using EdDSA and won’t need a hash at all.

I think the hard part for SHA-3 is not using it as a certificate hash but using it as a PRF.

Yoav