[cabfpub] FW: [websec] #58: Should we pin only SPKI, or also names
jeremy.rowley at digicert.com
Mon Aug 12 21:40:06 UTC 2013
Thanks Ryan - why are you against the CAB Forum assuming responsibility?
Also, I took Chris Palmer's comments as more neutral/in-favor of the
proposal than against it. Is Google opposed to pinning CAs as an option in
addition to pinning specific keys?
From: Ryan Sleevi [mailto:sleevi at google.com]
Sent: Monday, August 12, 2013 2:01 PM
To: jeremy rowley
Subject: Re: [cabfpub] FW: [websec] #58: Should we pin only SPKI, or also
On the User Agent side, we have significant reservations with this proposal,
which have been expressed during these discussions.
I would invite interested CAs to comment further on the IETF list, rather
than splitting the discussion here.
For what it's worth, we would prefer NOT to see such an activity in the CA/B
Forum IF it was adopted.
On Mon, Aug 12, 2013 at 12:54 PM, Jeremy Rowley <jeremy.rowley at digicert.com>
> See below. Is this a project of interest to the CA/Browser forum?
> The Forum would be responsible for maintaining a registry of CAs and
> the mapping to their roots for the purpose of key pinning.
> -----Original Message-----
> From: websec-bounces at ietf.org [mailto:websec-bounces at ietf.org] On
> Behalf Of Yoav Nir
> Sent: Monday, August 12, 2013 12:50 PM
> To: Trevor Perrin
> Cc: websec
> Subject: Re: [websec] #58: Should we pin only SPKI, or also names
> On Aug 12, 2013, at 7:11 PM, Trevor Perrin <trevp at trevp.net> wrote:
>> On Mon, Aug 12, 2013 at 1:17 AM, Gervase Markham <gerv at mozilla.org>
>>> On 11/08/13 05:25, Trevor Perrin wrote:
>>>> Could we just say:
>>>> - The holder of a domain name is responsible for specifying the
>>>> SPKIs that it maps to.
>>>> - How the domain holder communicates this to the UA is out of scope.
>>> In other words "Don't set up a registry; just punt the problem and
>>> hope something works itself out organically"?
>> If people hate this, someone should make a proposal for a registry:
>> - Who maintains it?
>> - How are requests to add or remove CA names authenticated?
>> - Does the registry map CA names to actual keys?
>> - If so, how are change requests authenticated?
>> - What are the timing rules to ensure changes are propagated to
>> browsers as needed?
>> - How can the registry be monitored and double-checked to avoid it
>> becoming a single point of failure?
>> - Should these process details be defined in the HPKP spec or
> As you've said, before this is for the CAs and browsers to come up
> with such a solution (assuming they want it, and I'm not hearing this
> from either the Mozilla people or the Google people on this list). CAs
> and browers. Now, if only there was some forum where both of these come
> Joking aside, The CA/Browser forum is not currently in the business of
> running registries. IANA is, but I don't know how to specify in a
> draft an IANA policy that would include following mergers,
> acquisitions, and branding, and settling trademark disputes. Not do I
> have any reason to believe that IANA would be willing to do this. So
> unless the CA/Browser Forum agrees to take on this responsibility, and
> provide stable link for both machine and human readable mappings, I
> think this proposal should be shelved until we can find someone who will
answer your questions above.
> websec mailing list
> websec at ietf.org
> Public mailing list
> Public at cabforum.org
More information about the Public