On Tue, Aug 21, 2018 at 1:10 PM Ted Hardie <ted.ietf@xxxxxxxxx> wrote:
On Tue, Aug 21, 2018 at 10:03 AM, Paul Wouters <paul@xxxxxxxxx> wrote:On Tue, 21 Aug 2018, Ólafur Guðmundsson wrote:
Ted, Would it be acceptable to just do
s/TCP/Connection oriented Transport/
For RFC 7901 we used "source-IP-verified transport"
If this is the only characteristic that the working group believes will cause variance in ANY responses, then reusing this terminology seems fine. I suspect, however, that it is not the only characteristic that will in practice. I can easily imagine cases where ANY returns full answers when the transport is confidential and minimal ones when it is not.I have absolutely no data to back this intuition, though, so I will understand if the working group and authors disagree.regards,Ted HardiePaul
As I see it, the data is 'public' in the sense that any other client could also request "ANY" over a confidential transport and get the same info, so I don't see why the server would care that the transport is confidential. It only protects the privacy of the client.
Preventing amplification, or unnecessary traffic in general, is the basic use case that I see.
--
Bob Harold
