Re: [Trans] [trans] #77 (rfc6962-bis): normative client behavior specified in Section 5

On 6 July 2015 at 19:33, Stephen Kent <kent@bbn.com> wrote:
> I see your point. 6269-bis is mostly a document specifying the
> operation and interface to the log. Other parts of the document attempt to
> describe how clients interact with the log, but they do so only
> superficially.
> That's why I believe we need additional document specifying the behavior
> of the other elements of the CT system.
>
> You refer to "the protocol" but don't indicate which protocol you have in
> mind.
> I might agree that specs for TLS client behavior in the context of TLS
> exchanges
> are in scope. If so, 6269-bis needs to state that it updates TLS.
>
> Your interpretation of the WG's agreement to expunge client behavior from
> 6269-bis
> seems to encompass only the issue of whether a client rejects or accept a
> cert
> accompanied by a (valid) SCT. That strikes me as rather narrow.
>
> If a client that claims CT compliance MUST perform a series of checks to
> verify
> CT compliance of a cert, what does the client do if any of these tests fail?
> It's
> not useful to require a set of checks to be performed and yet not indicate
> what
> happens if they fail. If no action is required, then why bother performing
> the checks?
> If some action is required, or even recommended, then a spec needs to state
> what that
> behavior is.

To restate our position on this: we don't know what clients will do in
response to failure to comply with CT, and nor are we in a position to
mandate it. We feel it is helpful to provide a mechanism by which the
issuance of certificates can be made open to public scrutiny. Clearly
adoption of this mechanism does ultimately require that clients behave
somehow differently for certificates that do not conform, but we can't
dictate what that behaviour is.

For example, right now, the only client I know of that reacts to CT is
Chrome/Chromium, and the behaviour it (at least for Chrome - Chromium
comes in multiple flavours) currently exhibits is that text in some
dialogues is changed depending on whether CT is present and meets
Chrome's policies or not, and in the display/non-display of the EV
indicators for EV certificates. This behaviour has already been
through several versions and policy changes, and will continue to
evolve - and is not arrived at by IETF consensus, nor will it ever be.
So, I don't see how it can be a part of the I-D.

>
> Steve
>
>> #77: normative client behavior specified in Section 5
>>
>> Changes (bybenl@google.com):
>>
>>   * milestone:   => review
>>
>>
>> Comment:
>>
>>   What has been ruled out of scope is what action clients take, not how
>> they
>>   conform to the protocol.
>>
>>   Suggest close wontfix.
>>
>
> _______________________________________________
> Trans mailing list
> Trans@ietf.org
> https://www.ietf.org/mailman/listinfo/trans