You got my position completely backward on this.

I feel much better now. :-). I really didn't want to cross swords with someone who's S/N ratio is as high as yours is around here.

You may be thinking of what I said about the cost of defeating traffic analysis.

That must have been it.

The natural state of the Net seems to be a kind of semi-anonymity.

Ah. I believe this is what I'm talking about...

Trying to push it in either direction (complete traceability or anonymity) is costly.

I'm going to build a rant about this pretty soon, but I think that in a geodesic network, audit trails of any kind cost more money, particularly in the "fringes" of the network, where all the processor growth is going to happen, assymetries or not. As processors handle smaller and smaller stuff faster and faster, it becomes harder and harder to "control" and monitor all of them. To operate efficiently, they have to be more and more autonomous. One way to provide decision rules for autonomous processors is to introduce micro-e$ auction markets into the net's infrastructure. A good example might be packet routing. Suppose you attach some very skinny money to a packet (something issued with Micromint, maybe?), and strip it off as the packet goes through the network, sender pays. Routers price their throughput based on their load at any given moment. When you create this "economy of switches", it becomes harder to establish the hierarchical book-entry control/coodination models we all know and love, certificate hierarchies among them. I think that something like this might work as a way of randomizing traffic in a network of remailers, by the way. Let the market determine the remailer path. Specify the number of hops left, and attach enough money for that number of hops. This is one way for people to pay for the use of remailers, and eventually, it could map to packet-level-anonymity someday. For the moment, I'm going to claim that anonymous auction markets are always more efficient than identified, command economies, just like flight is faster than surface travel. At least that's what I think this morning. :-). More as I get some time. However, I do agree that it's foolish to simply declare that everything must be unauditable, just as it is to declare universal auditability. It's better to let the market figure this out. ;-). Cheers, Bob Hettinga ----------------- Robert Hettinga (rah@shipwright.com) e$, 44 Farquhar Street, Boston, MA 02131 USA "Reality is not optional." --Thomas Sowell The e$ Home Page: http://thumper.vmeng.com/pub/rah/