On 8/20/14, Paul Wouters <[email protected]> wrote:
> On Wed, 20 Aug 2014, Jacob Appelbaum wrote:
>
>> Paul - perhaps this suggests that all stub and recursive resolvers
>> should log keying information, even if it isn't used for
>> validation/authentication/etc?
>
> That is one "out of band" authentication mechanism called TOFU (trust on
> first use) or LOF (Leap of Faith)

I'm suggesting that regardless of authentication, we should ensure
that we have some information for a user to report their experience.

>
> While possible, it will see a lot of false positives, like when going to
> a different starbucks using the same wifi ESSID.
>

There is no false positive - only an actually observed certificate. :)

> It could be done if one also logs mac address and/or lat/long info.
>
> But these are all local policy and local implementations issues.

I agree. I suggest that resolvers SHOULD log resolver certificate information.

All the best,
Jacob

_______________________________________________
dns-privacy mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dns-privacy

Reply via email to