Re: Last Call: <draft-nottingham-safe-hint-05.txt> (The "safe" HTTP Preference) to Proposed Standard

[Dave Crocker <[e-mail]>]

On 10/27/2014 4:45 PM, Brian E Carpenter wrote:
> No. I mean that a badly motivated web site can pretend to offer safe material
> using this but actually offer objectionable material (for whatever definition
> of safe or objectionable you care to adopt).


Brian,

That's a pretty surprising and quite disturbing criterion.  Apply it
consistently and we get no standards at all.  Ever.  Any site can choose
to be deceptive.

For example, when a message is relayed to a site and it accepts it, we
can't be sure it won't choose to mis-route it.

The purpose of the safe mechanism is to provide a standard way that a
user can state a basic desire to a server.  It is not the purpose of the
mechanism to guarantee that the server will behave honorably.

   1. The mechanism already has plenty of field experience demonstrating
basic utility.

   2. The proposed mechanism opts for simplicity.  More complexity would
actually increase the likelihood that a user's expectation's are not
matched.

Criticisms of the proposal are tending to miss the established
experience, or to propose entirely different designs that have no basis
from that experience, or to raise concerns that are frankly outside the
proper scope of the specification.

d/
-- 
Dave Crocker
Brandenburg InternetWorking
bbiw.net