Re: [dane] [saag] Need better opportunistic terminology

[Joe Touch <[e-mail]>]

Hi, Steve,

On 3/11/2014 2:53 PM, Stephen Kent wrote:
> Joe,
>
>> On Mar 6, 2014, at 1:23 AM, Phillip Hallam-Baker <[e-mail]
>> <mailto:[e-mail]>> wrote:
>>
>>> The term opportunistic has become the new synonym for 'Good' but it
>>> is being used for many different things.
>>>
>>> A) Unauthenticated key exchange
>>
>> Fwiw, this is IMO an error since I first introduced BTNS, and I had to
>> clear it up on Wikipedia multiple times. I see nothing opportunistic
>> about this mode as a stand-alone concept.
 >
> The original use of the termappears to be from RFC 4322, Micheal
> Richardson's document. He describes how to use keys retrieved from
> the DNS with IPsec/IKE, without prior, bilateral arrangements for
> access control, via the SPD. He defined OE that way, and noted that
> it was not an unauthenticated mode of IPsec.

RFC4322 defines OE.

Section 1.2 describes "anonymous encryption" which is basically 
unauthenticated key exchange.

 From later in that section:

    Although it is a useful mode, anonymous encryption is not the goal of
    this project.

Michael and I discussed the difference between the two (OE and anonymous 
encryption) on many occasions, and I don't think either of us ever 
confused the two (though someone who edited Wikipedia did once). The 
sentence above confirms that, AFAICT.

 > I prefer that we stick
> with that definition of the term, which is IPsec-specific.

I'm not quite sure what term or what definition you're referring to: OE, 
anonymous encryption, or unauthenticated key exchange. Can you clarify?

> I have
> suggested "opportunistic keying" as a preferred term, since its the
> key management, not the encryption per se, that distinguishes other
> proposed modes of operation for IPsec, TLS, etc.

I agree if you're replacing OE with OK ;-)

> The breakout group at the STRINT workshop that discussed terminology
> suggested using the term noted above.

Sorry, but to clarify, which term?

Joe