[Cryptography] Certificates and PKI

[Paul Wouters]

On Sat, 27 Dec 2014, Tony Arcieri wrote:

> On Sat, Dec 27, 2014 at 10:25 AM, Paul Wouters <paul at cypherpunks.ca> wrote:
>       You are confusing "opoprtunistic encryption" with "anonymous
>       encryption". If it is opportunistic, it does not mean that there was no
>       authentication. It just means it was not a pre-configured trust.
> 
> Every IETF draft I've read on the matter describes it as opportunistic encryption, so I think you're the one that's confused 

The term OE originates from The FreeS/WAN Project of which I was a
member:

http://www.freeswan.org/freeswan_trees/CURRENT-TREE/doc/quickstart.html

Freeswan did not allow un-authenticated encryption. In fact it required
mutual authentication, which is one of the reasons it was too hard to
deploy.

The IETF actually avoided defining the term OE and instead settled for
the term OS ("Opportunistic Security"):

https://datatracker.ietf.org/doc/draft-dukhovni-opportunistic-security/

I was (and still am) a very active opponent for using that particular
definition, which discussion you can find back in the IETF SAAG working
group archives.

See also http://en.wikipedia.org/wiki/Opportunistic_encryption

People mean different things when they say OE, which is one reason to
stop using the term. Unfortunately, OS in my opinion just makes it
worse.

So I have about 20 years of experience with the term and its
evolution.

Paul