Wei Dai <weidai@eskimo.com> writes:
On Fri, 21 Mar 1997, Timothy C. May wrote:
The best arguments are not those couched in terms of "benefits" to society/markets, but in terms of "lowest common denominators." I favor the "Schelling point" view, and have espoused it here often.
In most Western systems, a reasonable Schelling point is that various actors agree not to try to enter the "private spaces" of others. For example, my neighbors and I agree not to enter each other's houses to pry and inspect.
(The Schelling point is that this "boundary" point is essentially agreed upon even though my neighbors and I have never formally negotiated it...as with animals and their "territories," certain equilibrium points are reached.)
In Western systems, it is recognized that the _costs_ of coercively entering another's home and/or private space are too high, and so we declare there to be a "right of privacy." In reality, what we mean is that we accept a private domain as a Schelling point.
I think you're answering the third meaning of "why privacy", "why DOES privacy exist?" This is a positive rather than a normative question (which is what I was asking). I agree with you that most likely privacy exist because in some sense it is an equilibrium in the game of life. But what I'm asking is whether it is a desirable equilibrium, and whether a new equilibrium involving more privacy is better than what we have now.
Well things aren't standing still. Absent continual watchfulness, political lobying, and cypherpunkery (writing freeware crypto code, and providing privacy memes), privacy is taking a huge negative hit. You requested an argument couched market economic terms as to why reduced privacy might be a bad idea for the market efficiency. Consider: Less privacy is a bad thing in market economics terms, because as privacy takes a downwards spiral, which leads to ever increasing government intervention, increasing sizes of governments, fascism, etc. the free market economy will go to hell. Poverty, and food shortages will result, a la the former USSR, which is slowly recovering from the decline caused by statist, facist policies. We on the other hand, absent pressure from outside government, law enforcement and secret services circles are collectively headed into that fascist driven downward spiral.in economics.
I reject the argument Wei Dai mentions, that market benefits could obtain if society rejected privacy, for two reason. First, because in an important sense rights are not the slaves of corporate or personal business interests. (E.g., my interest in Alice's television viewing habits does not trump her right to keep me out of her house.). Second, because so-called "market efficiency" is largely an illusion anyway.
Certainly no real market is perfectly efficient, but some markets are relativly more efficient than others, and most markets are more efficient than alternative forms of economic organization (e.g., socialism) so I don't see how you can call "market efficiency" an illusion. The inefficiencies I was talking about were relative to the case where both parties have the same information.
Also, I don't quite understand your first argument. It seems to suggest that privacy should exist for no reason in particular. If this is the case then it doesn't make sense to argue about the costs/benefits of privacy. But it is my understanding that most cypherpunks believe more privacy benefits everyone, and therefore work to making more privacy for everyone. What I'm looking for are arguments that support this belief.
Wide-spread strong privacy tends to support a freer economy, and reduced government overhead. Presumably the increased market efficiency you refer to as obtainable by reducing privacy refers to the loss of direct marketing opportunities for commercial advertisers. I'm not sure that this (putative) gain in the market efficiency would be significant compared to the huge efficiency gains to be made from dismantling or reducing governments. Is it not pro-competition, and supportive of a free market to have privacy preserving technology widely used? Then commercial advertisers who wish to target groups of potential purchasers of their wares, can pay the purchasers for the information. (The information belongs to the individual, so it seems fair that a commercial entity should have to pay for it.) Examples would be store loyalty cards which offer discounts, or points, and in return obtain customer spending patterns. If sharper commercial interests choose to buy the users spending patterns at reduced rates by preserving their privacy using non-crossreferenceable Chaumian credentials, well more power to them. The market can decide. This is a market based solution. Adam -- print pack"C*",split/\D+/,`echo "16iII*o\U@{$/=$z;[(pop,pop,unpack"H*",<> )]}\EsMsKsN0[lN*1lK[d2%Sa2/d0<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<J]dsJxp"|dc`