December 2003 - cypherpunks-legacy

Re: Publicly Verifiable Anonymous Voting System
by tob＠world.std.com 17 Dec '03

17 Dec '03

JonWienk(a)ix.netcom.com writes: > That doesn't really do it. In fact, I'm not sure why you move in this direction at all. No offense. Seems to me that voter registration of some kind must be UNanonymous. I want to have confidence that my empowered political opponents aren't voting 10,000 times each. It also seems to me that if the entire system is the government's black-box, there is no anonymity. It is a trustworthy as the government itself. At worst, if they want to know how you voted, they can sum all votes, then sum all votes but yours, then subtract. Seems to me that the best thing would be a system where you the voter pick who is to handle your vote, in a sort of hierarchical tree. Each level has the same goal as a DC net, to make it difficult to tell who among many people communicated, and further levels confuse which group did, and which group-of-groups did, and so forth. Each group delivers its collected votes, publicly but unidentified, to a group-of-groups, and watches that groups' output. Obviously you would use multiple envelopes to prevent premature disclosure. At each level, you need to be able to identify your own vote and check that it is unaltered, but no-one else should be able to associate it to you. This assumes that there is no way for anyone to see you looking for your own ballot, which is a separate facet of the problem. Everyone needs to be able to see that the collected votes did not exceed the number of voters that group has. This is pretty simple and requires a fair bit of collaboration to defeat. But it can be partially defeated by partial collaboration and the rest guessed from parallel voting patterns. If your entire group is seen to vote the same way, your vote among them is obvious. Tom

1 0

Re: WaPo on Crypto-Genie Terrorism
by Deranged Mutant 17 Dec '03

17 Dec '03

On 27 Jul 96 at 19:21, John Young wrote: > The Washington Post, July 27, 1996, p. A22. > Speaking in Code on the Internet ... [Editorial] Some bothersome things about this editorial... [..] > security of their data. They also see it as a market in > which the United States maintains a comfortable lead, one > that is threatened if domestic encryption makers can't sell > their products elsewhere. The makers argue that foreign > encryption software will rush in to fill the gap, doing > nothing about the uncrackability problem -- indeed, making > it worse. The administration in turn is pursuing a wider IMO, the US does not have a comfortable lead. It's already falling behind considering some of the stronger crypto programs available (at least as freeware) are made outside the US. Many of the stronger algorithms were invented outside of the US (IDEA for instance). [..] > with wiretapping. Mr. Freeh, testifying at Thursday's hearing in > favor of an optional key escrow plan, noted that the point is not > to prevent all copies of uncrackable code from going abroad -- that's > clearly impossible -- but to prevent such high-level code > from becoming the international standard, with architecture > and transmission channels all unreadable to world > authorities. To software companies and Internet users who So why should criminals bother with using standards if they are readable by authorities? > have been clamoring for the right to encrypt as securely as > possible, Mr. Freeh and others argue, "the genie is not yet > out of the bottle" on "robust," meaning uncrackable, > encryption. Are they going to magically erase all copies of strong software that is already currently available? (Side note: the Pacifica news report on Friday notes that while Freeh gave his testimony, over 100 copies of PGP were downloaded from MIT's site.) [..] > Encryption, if widely used, could conceivably ease some > privacy problems concerning who gets to see personal and > financial data on individuals -- though such data usually > are vulnerable to being dug out of storage rather than > intercepted in transmission. But neither is it clear that And evidence cannot be encrypted. You cannot encrypt an airplane full of cocaine or an unusually expensive car baught by money from drugs, espionage, etc. Nor can you encrypt bomb-making materials, nor conversations in a room (from your mouth to a telephone receiver). Nor will encryption do anything about informants inside the communications loop. etc. etc.... > the encryption enthusiasts' desire for free development > should take precedence over the tracking of terrorism. At It's not clear that terrorism can be tracked, even if it's unencrypted. The OK and WTC bombings were apparently not encrypted, and there's some allegations that the authorities had advanced warnings of the latter. > the very least, Congress should be exceedingly cautious > about getting out ahead of administration concerns on > controls that, once lifted, are hardly reversible. The controls haven't done much to prevent free software from being exported. They only control commercial sales of software (and hardware). Particularly absent in the WaPo-ed is that many do not trust the authorities (in the US and elsewhere)--particularly the FBI, which has a long history of extra-legal surveillance. Rob --- No-frills sig. Befriend my mail filter by sending a message with the subject "send help" Key-ID: 5D3F2E99 1996/04/22 wlkngowl(a)unix.asb.com (root@magneto) AB1F4831 1993/05/10 Deranged Mutant <wlkngowl(a)unix.asb.com> Send a message with the subject "send pgp-key" for a copy of my key.

3 2

Re: Twenty Bank Robbers -- Game theory:)
by tob＠world.std.com 17 Dec '03

17 Dec '03

What the 20 Cypherpunk Bankrobbers do, aside from writing "Memo: Need better loot-division scheme. *URGENT*." Assume every player acts completely rationally, and unwaveringly trusts every other player to fully understand the situation and act solely to maximize their profit in this single situation. Also assume that each player announces an immutable, unconditional monetary distribution before anyone has voted, and the announced distribution constitutes the whole of the payoff (no way to make or enforce promises, no external loyalties, no side payments, etc.) Also assume no revolt against the rules is possible. Every proposer makes the most profitable deal for himself that will pass the vote, if they can predict it (As it happens, they always can with one exception, see below) Every player, when voting, compares what he is offered right now with what he will be offered between now and the first guaranteed successful proposal. He inspects the future possibilities and sees that the next offerer can make a successful proposal if they (the player) vote down the current proposal, and if his vote is needed this time, it will not be needed for the next offer to succeed. (You can check this assumption below). So he sees no reason to hope for a better result by voting against apparently favorable offers. ---- If we were to keep the condition that all other things being equal, players vote for the other players to live, the proposer always keeps the *entire* amount. If a player expects to be offered no money next time, it makes no sense to offer him anything this time, because you've got his vote for sure anyways (nice guys finish last). Working backwards, once a player will be shut out in the future, it never makes sense to cut him in. In fact, for each player it makes sense to some proposer to shut him out. The guy next in line will never vote for the proposer to live, so is always shut out. So nobody but the proposer expects any money and everybody but the guy next in line votes for the proposer to live. ---- So instead I'm going to assume that if a player does not do *better* than he will next time, he votes against the proposer. The result is that the first player not only lives, but makes out like a bandit, to the extent that the loot is subdividable. If there were 40 million and 1 guys in line, he'd be meat. Likewise, if the money was all in $2,500,000 notes he should start framing his last words. I'm going to write it in reverse order, where #0 is the last guy, #1 is second to last, etc. because it's easier to see the progression that way. 1 Player: #0: everything Voting for: #0 Trivially succeeds 1/1 2 players: #0: nothing #1: Everything Voting for: #1 Succeeds 1/2 3 players: #0: $.01 #1: nothing #2: Everything else. Voting for: #0 #2 Succeeds 2/3 4 players: #1: $.01 #0, #2: nothing #3: Everything else. Voting for: #1 #3 Succeeds 2/4 5 players: #1, #3: nothing #0, #2: $.01 #4: Everything else. Voting for: #0 #2 #4 Succeeds 3/5 6 players: #1, #3: $.01 #0, #2, #4: nothing #5: Everything else Voting for: #1 #3 #5 Succeeds 3/6 The other cases proceed similarly, grouping even/odd. In the end, the guy at the head of the list offers every other guy a penny and keeps the rest -- and then starts to sweat as he wonders whether our assumption of perfect rationality will hold up in real life. Paul Foley nearly got it, but it does matter who is offered money. If it didn't, then nobody could count on doing worse on the next proposal. -------- The case where the proposer does not get a vote is completely different: 1 Player: #0: everything Voting for: Vote is indeterminate 0/0 Player seals himself into a box with the money, Schroedinger's cat, and a vial of poison that has a 50% chance of being shattered. Everything that follows depends on how we resolve that case. If we assume some probability mixture of money & death, it is not given how to weight the player's negative payoff (death) vs positive payoff (money). Let's assume Schroedinger's player gets all the money and is not required to suicide. 2 players: #0: Everything #1: nothing Voting for: Fails 0/1 Remember the assumption that if a player does not do *better*, he votes against the proposer. This is a null move since the outcome is the same as above. 3 players: #0: nothing #1: $.01 (or nothing) #2: Everything else Voting for: #1 Succeeds 1/2 Note that we assumed that #1 requires strictly more money than he would otherwise be offered (His life means nothing to him). Otherwise he would be offered $0. (Yeah, I don't want to rewrite the analysis just for that). It would end up with the same pattern, though, just faster. 4 players: #0: $.01 #1: $.02 (or $.01) #2: nothing #3: Everything else Voting for: #0, #1 Succeeds 2/3 That fact that the proposer cannot vote means he needs to pick up an extra vote. He can't get it from the guy who would get almost everything, and a penny secures the vote of the player who would get nothing. So he has to offer the player who would get $.01 the next increment of money. If #1 valued his life, he'd be offered $.01 instead. 5 players: #0: $.02 #1: nothing (or group #0,#1 interchangeably) #2: $.01 #3: nothing #4: Everything else Voting for: #0, #2 Succeeds 2/4 ...leaving the next (previous) proposer a similar position. 6 players: #0: nothing #1: $.01 (or group #0,#1,#2 interchangeably) #2: $.02 #3: $.01 #4: nothing #5: Everything else Voting for: #1, #2, #3 Succeeds 3/5 7 players: #0: $.01 #1: ? #2: nothing #3: ? #4: $.01 #5: nothing #6: Everything else Voting for: #0, #4, ? Succeeds 3/6 For the first time here, one of the proposers has an arbitrary decision. He has #0 and #4 and he needs to pick up one more vote. He can't do it for $.01 and there are two ways to do it for $.02. We would have encountered this situation earlier if #1 valued his life. Here we have to add some notation to indicate a non-determinate distribution over a set: (set): (non-zero payoffs, all others are 0) average 7 players: #0: $.01 (#1,#3): ($.02) average $.01 #2: nothing #4: $.01 #5: nothing #6: Everything else Voting for: #0, #4, (#1 or #3) Succeeds 3/6 8 players: #0: ? (#1,#3): ? #2: $.01 #4: ? #5: $.01 #6: nothing #7: Everything else Voting for: #2, #5, ?, ? Succeeds 4/7 The proposer easily gets #2 and #5 and needs two more votes. #6 can't be reached. The rest all have an expected payoff of $.01, so he gives $.02 to any two of them. Our indeterminate-list seems to be growing. 8 players: (#0, #1,#3, #4): ( 2 x $.02) average $.01 #2: $.01 #5: $.01 #6: nothing #7: Everything else Voting for: #2, #5, 2 of (#0, #1,#3, #4) Succeeds 4/7 9 players: (#0, #1, #2, #3, #4, #5): ( 3 x $.02) average $.01 #6: $.01 #7: nothing #8: Everything else Voting for: #6, 3 of (#0, #1, #2, #3, #4, #5) Succeeds 4/8 The indeterminate-list has now swallowed up everything but the last 3 players and will continue to do so because the average payoff to the list is always more than $.01 and less than $.02. 10 players: (#0, #1, #2, #3, #4, #5, #6): ( 4 x $.02) average $.08/7 #7: $.01 #8: nothing #9: Everything else Voting for: #7, 4 of (#0, #1, #2, #3, #4, #5, #6) Succeeds 5/9 -------- Another set of assumptions is that the players do not trust each other to be rational or to see all outcomes. In this case, the guy at the head of the line says in a loud voice to the guy behind him: "Whoever would vote to kill me even though I offer them money would vote to kill you too even if you split the money with them, so if it gets to you you should cut out anyone who got my money but voted to kill me, but keep those who got my money and voted to keep me alive." He then allocates equal portions to half the group, including himself and not including the guy behind him. He probably lives. --------

1 0

Re: Questions...
by David Rosoff 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- At 12.12 PM 7/27/96 -0400, Alan Horowitz wrote: >> Hello, I would rather I not be too blunt, but despite my generous >interest in computer > hacking cracking, and other such related topics, I >have come to be confused by the > mailer, can I ask any questions that I >wish, or am I limited by some type of header > subject? > > > >Shaun, let me explain. There's a committee of seven people. Me, Tim May, >David Sternlight, some assination-politics guy, some guy named Vultis or >somesuch, and so on. > >Only if we're in unanimous agreement on the outcome of an issue, may you >start a thread on a new topic. which then continues till the first >posting which calls someone a Nazi. And if they *don't* like your suggestion, you get killed and they move on to the next guy's suggestion of how to split up the available issues so that each cypherpunk gets a fair share of ranting. =============================================================================== David Rosoff (nihongo ga sukoshi dekiru) ---------------> drosoff(a)arc.unm.edu PGP public key 0xD37692F9 -----> finger drosoff(a)acoma.arc.unm.edu or keyservers 0xD37692F9 Key fingerprint = 25 7D AA 01 85 41 43 89 50 5A 33 76 F1 F1 99 67 Do you know who's reading your email? ---> http://www.arc.unm.edu/~drosoff/pgp/ Is it a forgery? --- I have PGP signed all email and news posts since May 1996. =============================================================================== "Relax. It's not a real alarm. They can't crack _Pentagon_ codes. Can they?" :p -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBMfqXpBguzHDTdpL5AQHkjQQAoLcf75f6TcKCGUaoq0c1JiL+seDFgw2l LQNHY+P/coI8KmQeEpeZgdAVEmts+BNbhGfHSaHTFtyAoUP24OfRDyVr7Mn4d0gF 1wuDp4aVNtGAEMCVkg9dXL6Klf38NeglU1EH2522loYo6g6/ANiTdPX729KrMNMO 3iPydVCu0aI= =NhSg -----END PGP SIGNATURE-----

1 0

Re: "privatizing" phones?
by David Rosoff 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- At 04.11 PM 7/26/96 -0700, Eric Murray wrote: >It's a 46mhz analog model, newer phones use frequencies >in the 900mhz range and/or digital encoding. >The 900mhz range is one of those 'blocked' in most newer radio >scanners, this is required by law as of a few years ago. >Many scanners can have some or all of the locked-out >ranges restored by suitable modification (i.e. removing >a resistor). Many of these mods are posted on the net. One of my friends has a Radio Shack scanner (the cat. # of which I do not recall) which does pick up the 900MHz range. We tried (unsuccessfully) to modify it according to an internet posting. I don't know how old it is. =============================================================================== David Rosoff (nihongo ga sukoshi dekiru) ---------------> drosoff(a)arc.unm.edu PGP public key 0xD37692F9 -----> finger drosoff(a)acoma.arc.unm.edu or keyservers 0xD37692F9 Key fingerprint = 25 7D AA 01 85 41 43 89 50 5A 33 76 F1 F1 99 67 Do you know who's reading your email? ---> http://www.arc.unm.edu/~drosoff/pgp/ Is it a forgery? --- I have PGP signed all email and news posts since May 1996. =============================================================================== "Relax. It's not a real alarm. They can't crack _Pentagon_ codes. Can they?" :p -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBMfqQkRguzHDTdpL5AQFCuQP+ILE3NAbjEZ2p6d6WDIhoi9yNYafQOiSv u00hink6Ylz8pQGP1xhQ9Bsn5pLyrS9Mck/UJ4Qw6omJBrE2LkgNy5Du8xdkqlQu 9MZaVqPXYphnGdAJ+oORXhvSkI9G0qYZPJjByypMd4Da92vqmDogrdTJFKgK21vh OeTNyh8i+zQ= =55JX -----END PGP SIGNATURE-----

1 0

Re: Feinstein wants controls on Internet, Books
by Marshall Clow 17 Dec '03

17 Dec '03

>From <http://cnn.com/US/9607/28/clinton.speech/index.html>: > The president said he has invited congressional leaders to meet with the > head of the FBI on Monday. > > The leaders will discuss bills that would expand wiretaps and allow for > chemically "tagging" explosives, which could help track terrorist acts. > Such legislation is needed to "increase the protection of the American > people," Clinton said. > > Both measures were taken out of an anti-terrorism bill Clinton signed > into law earlier this year. He also said that a spate of recent acts of > political violence showed that law enforcement needed such powers. > > "As strong as the bill was, it did not give our law enforcement officials > some of the powerful tools I had recommended," Clinton said. and from a similar article on AOL (from Reuters): > ``We will ... do whatever is necessary to give law enforcement the tools they need to find >terrorists before they strike and to bring them swiftly to justice when they do,'' he said. > [snip; same as the cnn story] > > The expanded wiretap authority would allow law enforcement personnel to listen to all >communications devices used by a given person, including a home phone, mobile phone and >pager. At present, authority is given over a specific phone number, rather than for a >specific person. > > The second measure would require explosives makers to insert a chemical fingerprint >in their products that would help authorities narrow their search for a bomber. The White House web page does not have a transcript of the President's speech; in fact, under "Today's Top Issue" was 'President Clinton's Call for a National Community Policing Number', dated July 23rd. -- Marshall Marshall Clow Aladdin Systems <mailto:mclow@mailhost2.csusm.edu> "We're not gonna take it/Never did and never will We're not gonna take it/Gonna break it, gonna shake it, let's forget it better still" -- The Who, "Tommy"

1 0

your favorite poison recipes
by anonymous-remailer＠shell.portal.com 17 Dec '03

17 Dec '03

Please post your favorite poison recipes to this mlist.

4 3

Re: Twenty Bank Robbers -- CLARIFICATION
by James A. Donald 17 Dec '03

17 Dec '03

At 04:33 PM 7/25/96 -0700, Jeremey Barrett wrote: >-----BEGIN PGP SIGNED MESSAGE----- > > Assuming "perfect" intelligence on the part of the robbers (i.e. they will > follow deterministic behavior and do the "right" thing), then here's what > must happen IMO (1 being the first guy and 20 being the last): [...] > If 2 are left (19 & 20), 19 gets all the money. So 20 will vote for whatever > 18 says, which MUST include 20 in the deal. But here we run into the paradox, that it is not in each persons self interest to pursue is self interest. Example; Suppose only two are left: Then No. 19 get everything, and No. 20 gets nothing. So if only three are left, it will maximize 20's return to vote for a proposal by number 18, that number 18 gets 19,999,999 dollars, and number 20 gets one dollar But suppose that number 20 announces in advance that he considers an unequal division morally wrong, and will always vote against any unequal division. If this threat is credible, which it is, number 18 will have no choice but to propose an equal split, so number 20 now gets 6,666,666 dollars, instead of one dollar. Now if number 20 had threatened to vote against any proposal that did not split the money equally between himself, and number 18, thus going for ten million instead of six million, this threat would be less credible, and he would very likely have wound up with one dollar. This is a particular example of the various well known paradoxes of utilitarianism, that utility is maximized by a firm and credible promise to utterly disregard utility maximization. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.jim.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the state. | jamesd(a)echeque.com

1 0

Re: Game Theory and its Relevance to Cypherpunks
by James A. Donald 17 Dec '03

17 Dec '03

Llywarch Hen wrote: > > > What Timothy May espouses is not the appearance of craziness but actual > > > insanity itself. James A. Donald wrote > > if you organize your nuclear forces so that any > > serious war is likely to escalate uncontrollably into the battle of > > armageddon, regardless of your intentions and desires, which is how > > the American government organized its nuclear forces in Europe, > > then you can pretty much guarantee you will not have to face a > > serious war. At 03:09 PM 7/27/96 -0700, Jon Leonard wrote: > This was, of course, the logic behind the alliances before World War I. > It was obviously suicidal to start a war with any of the major powers, > but it happened anyway. Your history is false: War was romantic and a big vote winner before World War I, and after World War I and before world war II the "intellectuals" were still big fans of war, just as they are still big fans of communism and socialism today, though World War I was sufficient to kick sense into most normal people. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.jim.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the state. | jamesd(a)echeque.com

1 0

Re: your favorite poison recipes
by jim bell 17 Dec '03

17 Dec '03

At 12:11 PM 7/28/96 -0700, anonymous-remailer(a)shell.portal.com wrote: >Please post your favorite poison recipes to this mlist. I recommend a book called "Handbook for Poisoners," by Raymond Bond. While it is mostly a collection of (mostly fictional) poison stories, the forward is a rather extensive non-fictional summary of poisons as was commonly known when the book was written, in 1950 or so. The main problem is that it doesn't include the various anticholinesterases then known, such as Tabun or Sarin. Jim Bell jimbell(a)pacifier.com

3 2