authenticating Real Life(tm)

[Jim Choate]

On Tue, 16 Jan 2001, Tom wrote:

> I love it when you conflict with yourself. :)

Then in general you must hate me...

> so you trust the proof. great. if you trust the proof, and the protocol
> has just been proven, then your trust extends to the protocol. and so
> on. web-of-trust.
> 
> please don't say you don't. because if a protocol that was just proven
> by a prove you trust has not earned your trust by that procedure, then
> obviously you lied when you said you'd trust the prove.

The 'proof' IS the 'protocol'. You act as if 'proof' and 'trust' are
equivalent. They're not. I 'trust' because I know the protocol won't
'lie'. That is the 'trust' and the heart of the 'proof'.

This of course, speaking of Real World (TM), raises the question of if a
protocol even exists. In most cases it doesn't. And in many cases even if
it did only Bill G. could afford it.

Here is the heart of commercial authentication services. Raise the cost of
tampering with the system while at the same time not raising the cost of
actually doing the system. Otherwise the customers can't afford it. It's a
horse race. And it will be until every iota of information is free (fat
chance).

For your assertion to be so you still need to prove:

A trust B, B trusts C, therefore A trusts C.

After all, simply because you and I trust the protocol still doesn't mean
I trust you. It only means I believe you haven't lied in this particular
case.

Another aspect is that the 'authentication' is good only for now. The fact
that we require the protocol to be repeated each time is a priori
admission of our trust. I mean if we've authenticated them once it should
be ok for now on if we 'trust' them.

You can't because it doesn't. This proof is central to your assertion.
It's fundamentally central of 'web-of-trust'. It's why this particular web
doesn't scale well.

I trust the protocol to fail if it is tampered with because of universal
access to its base components (ie isotropic and homogenious). Something
that no amount of money, time, or political influence can change.

I use the protocol not to decide my trust but to give me a reason to opt
out of the process. Fundamentally if you have to apply any of these sorts
of protocols to an exchange a reasonable person won't want to be involved
in the first place. There is a fundamental lack of trust already extant.
The key point however is to recognize the true function of any
authentication protocol, to stop now - to drop out.

Speaking of examples of trust,

In the made-for-tv SG-1 pilot movie they must identify if the stargate is
a trap. They do this with a simple empty cleanex box. This is a great
example of 'trust' and 'authentication'.

    ____________________________________________________________________

           Before a larger group can see the virtue of an idea, a
           smaller group must first understand it.

                                           "Stranger Suns"
                                           George Zebrowski

       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage at ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------