From: juan <juan.g71@gmail.com>
On Thu, 10 Nov 2016 14:14:50 -0500
Steve Kinney <admin@pilobilus.net> wrote:

>  Bounties for killing the
> operators of an AP system, offered through more old fashioned means,
> would be extraordinarily high - requiring bullet proof anonymity in
> the presence of uber-motivated adversaries with global network
> surveillance capabilities.

>    Hey, but Jim's system (which Tim May 'invented' before Jim I
 >   believe?) would be protected, by, GET THIS, TOR.

That depends on how much you want to distort reality.  I was unaware of the existence of the CP list prior to about May 1995, as I recall.  (Someone I don't recall forwarded a copy of AP part 1 to the CP list, and then alerted me to the list's existence.)

 Prior to that, I was unaware of anything about Tim May except that he had worked for Intel (in Santa Clara, California) in the late 70's and early 80's, and he had discovered that alpha particles (charged helium nuclei) were the main cause of 'soft errors' in DRAMs of that era.  
As I learned, much later, May (and others) invented the idea of an "Assassination Market", at least what I now call the "Anonymous Person A hires Anonymous Person B to kill Person C" version.  Entirely unaware of their specific work (but, as I vaguely recall, aware of this general concept; I'd probably heard of it, indirectly, from a third person whose identity I don't recall), I thought of the "Hundreds, or thousands, or millions of 'Person A's', make anonymous contributions to a general offer to potentially any 'Person B' to reward him for 'predicting' the date of death of 'Person C'.

Are these two models alike?  Kinda-sorta, I suppose.  But I think they would be enormously different in effect, for many reasons I need not go into here.  If 'Assassination Markets' were limited to the former model, very few people would be hated, enough, by only one person to obtain a donation sufficient to buy a death.  In the latter model, a few million 25-cent donations would get rid of nearly all potential targets.

I suggest that I did indeed advance the rhetorical state-of-the-world.  

>    Only very ignorant people would fail to realize that TOR
>    provides bullet proof anonimity, especially against the
>    pentagon.

I didn't mention TOR to imply that it is, in its current form, entirely suitable for use in a functioning AP system.  Rather, my intent was to show that the kind of tools necessary to implement AP are being considered and produced.  Just "the kind of tools", not necessarily the tools themselves.  TOR should be made stronger, with more hops, more exit nodes, and more transfer nodes, filler traffic, for some examples of improvements.  Bitcoin needs an upgrade, for example to Zerocoin, to provide true anonymity, rather than mere pseudonymity.  


> The betting pool itself would alert

No, it would not.  Unlike the Federal Government's short-lived proposal in 2003, PAM "Policy Analysis Markets", (FutureMAP),  https://en.wikipedia.org/wiki/Policy_Analysis_Market   in which the state of the betting itself alerts people to threats, a well-designed AP system would carefully avoid alerting the public (or anyone) to bets, ideally until later, after the event predicted had materialized or failed to materialize.  The 'money' for the bet might be inside an encryption envelope, without the name of the target or date.  Another encryption envelope, inside the first one, could contain the target and date information.  The AP organization could decrypt the first (outer) envelope, and be unable to decrypt the inner one, at least until the password is sent in by the predictor. 

 The AP organization would, however, publish the decrypted contents of the outer envelope, so that everyone would know that a prediction with $X of value came in on a specific date and time. Nobody, except the predictor, would know the identity or date.  Eventually, the inner password would be sent in, used to decrypt the inner envelope, with the results published online.  If the AP organization cheats, by not failing to perform one of these steps, the predictor could publish the inner prediction key himself, disclosing to the public that the long-since-published content of the outer encryption envelope was a valid prediction, and for some reason (fraud?) the AP organization did not play fair.   That would destroy the credibility of that specific AP organization; others would soon take its place.


> potential targets to take proportional defensive measures, which "at
> best" would inhibit the social progress promoted by the system.

The system would adapt.  Consider Le Chatlier's Principle.

A working AP system might, for example, authorize spending (for concreteness) 10% of donations on defensive contracts:  Consider the effect of a $250,000 reward on the prosecutor in a case alleging an AP action, or $500,000 for a judge.   Or perhaps a reward of $100,000 for each juror who participated in such a trial, and voted for acquittal, where the outcome was such that a retrial would be impossible, or at least did not occur.    Such rewards could become very high, in large part because there would rarely be legal cases in which they would have to be paid.  


> But other than that...\

"Other than that, Mrs. Lincoln, how did you like the play?"

        Jim Bell