Re: New Version Notification for draft-thomson-http-encryption-00.txt

--------
In message <55522FBA.2010108@treenet.co.nz>, Amos Jeffries writes:

>> If you use different keys for different users, then you have to add
>> Vary: Encryption.
>
>Encryption is defined as a response header.

Why would it be limited to responses ?

>Vary lists request headers.

Which is why we should think about "Accept-Encryption"

>You cannot be both cacheable and use per-client keys (or payload for
>that matter).

"per-client" can trivially mean "per-group-of-clients" -- particularly
so since this does not at all address or limit key-management.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.

Received on Tuesday, 12 May 2015 17:18:20 UTC