Re: [Add] new draft: draft-grover-add-policy-detection-00

[Vittorio Bertola <[e-mail]>]

> Il 8 luglio 2019 22:42 Peter Saint-Andre <[e-mail]> ha scritto:
> 
>  
> A point of clarification...
> 
> Paul Hoffman was kind enough to point us to
> draft-sah-resolver-information and draft-sah-resinfo-doh just now, which
> intend to solve a similar problem. It's not that we think those I-Ds
> define a bad approach, instead we missed them entirely. We will read
> them between now and the Montreal meeting so that we can discuss them
> intelligently in person and on the relevant discussion lists.
> 
> Sorry about the lack of due diligence!

Still, if I remember the two sah drafts correctly, they define the basic framework for discovering information about the local resolver, but neither deals with discovering local policy, so we could still turn this effort into an additional document building on sah-resolver to add that piece of information.

If we do, however, I would suggest that we work on a much more nuanced expression of policy behaviour, including the origin of the policy (user, network or law) and perhaps even a description of the reasons and of the types of destinations that are affected. While "neutral" applications might just want to discover that policy exists and avoid Do*-to-the-cloud in any such case, I would expect "opinionated" applications ;-) to behave differently depending on specific policy types and sources, as well as, maybe, the location of the user and other factors. In such case, and barring the problem of lying resolvers, having more detailed information would lead to matching the expectations of the user much more precisely.

Ciao,
-- 
 
Vittorio Bertola | Head of Policy & Innovation, Open-Xchange
[e-mail] 
Office @ Via Treviso 12, 10144 Torino, Italy