Re: Comments on draft-mm-wg-effect-encrypt-11

[Randy Bush <[e-mail]>]

Benoit Claise wrote:
> The operators are used to manage their network in a certain way.

the ways operators manage their networks is highly varied.

but, in reality, what packet data do i need beyond the basic four-tuple
and congestion markings?

> The change for more encrypted traffic will force a change of those
> operational practices.

not necessarily.  of course it will seriously impact those operators
doing dpi, http header insertion, etc.  many of us consider this a
feature of encryption not a bug.

> This document should serve as for a starting point to have this debate
> at the IETF> practices.

it is.  and it should not go forward until we have had this debate.  and
we clearly have significant differences today.

> based on the documentation of those operational practices.

the set of operational practices is manifold.

Kathleen Moriarty wrote:
> There's an explicit statement that says the IETF does not endorse the
> documented practices.

standard wiggle 14.3.  please specifically call them out as negatively
affecting privacy and dis-recommended.

> It's not the practices, but the overall document that we should have
> consensus on - that it is important to document these practices so we
> have a starting point for discussion.

as christian alluded, the practices are not a closed set.

Pete Resnick wrote:
> I cannot come up with any way to read the mention of super cookies in
> section 8 as an endorsement at all.

or as an anti-endorsement.  so you would lay out the road map with no
marking of the evil paths.

> Either way, lack of overt disapproval is not endorsement.

in the real world of "buy our X device which implements rfc 666's
description of how to murder users" it is endorsement.

randy