> On 28 Oct 2016, at 16:40, Paul Hoffman <[email protected]> wrote:
> 
> On 28 Oct 2016, at 8:27, Bob Harold wrote:
> 
>> In Section 5 on Opportunistic Privacy, I am not sure "MAY" is correct.  If
>> the user chooses Opportunistic, I would think the server MUST try to be
>> secure, in whatever ways are possible, but MAY fall back to less secure,
>> only if those fail.

I think it is this text that Bob is referring to:

“Clients using Opportunistic Privacy SHOULD try for the
   best case but MAY fallback to intermediate cases and eventually the
   worst case scenario in order to obtain a response.  It therefore
   provides no privacy guarantee to the user and varying protection
   depending on what kind of connection is actually used.”

One reason I left this as a SHOULD not a MUST because forcing a client to try 
all available servers for a secure encrypted connection can introduce 
significant latency into the resolution process which is probably undesirable 
for Opportunistic. For example, the client has 5 recursive resolvers 
configured. Authentication fails on the first, but encryption is ok. MUST the 
client try everyone of the other server to try to get an authenticated & 
encrypted connection or can it just use the encrypted connection it already has 
set up, given that it is using Opportunistic Privacy anyway? Appendix A touches 
on this a little bit too in terms of probing for server capabilities. 

And yes, if we decide to go for keeping it simple then I think this should be  

“Clients using Opportunistic Privacy SHOULD try for the
   best case and SHOULD fallback to intermediate cases and eventually the
   worst case scenario in order to obtain a response. “

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

Reply via email to