December 2003 - cypherpunks-legacy

Cyphernomicon FAQ -- Review in "Wired"
by tcmay＠netcom.com 17 Dec '03

17 Dec '03

A couple of items of interest: 1. Rishab Aiyer Ghosh tells me he has committed to write a review of the Cyphernomicon FAQ for "Wired," for either December or January publication. I had nothing to do with this, and it surprised me when he contacted me to notify me. (I reminded him of the "pre-release" status, the incompletenesses and holes, but he feels it's still timely to review it. And by the time his review gets printed, a newer revision will be available.) I'll make some comments later, but I of course urge Rishab to make his own comments here. 2. The Cyphernomicon is available in two main places that I know of: * My own anonymous ftp site: ftp.netcom.com, in pub/tcmay - 6 files, including the full FAQ, listed as "CP-FAQ" - and 2 flavors of this in compressed and gzipped forms - Contents, MFAQ (Most Frequently Asked Questions), etc. - Netcom has been having performance problems, lately * The Karl Barrus/Chael Hall gopher and ftp site: gopher chaos.bsu.edu - Karl has split the full CP-FAQ file into 20 chapters, each of which can be individually retrieved/mailed. The Cyphernomicon can of course be put on other ftp sites, though this complicates the version control problem. I would appreciate it if anybody who does put it on an ftp site agrees to keep it current, to periodically check my own ftp site and copy over newer versions. (I think the approach of _pointers_ to a few sites is a better approach than copying the actual documents to many sites. The most popular docs will be replicated at more sites, but I think the Cyphernomicon FAQ is sufficiently specialized that a few sites will be enough...that might change if Rishab's review in "Wired" meets with interest...) 3. Hal Abelson, Professor of Computer Science at MIT, has asked me if he could convert the FAQ into a Web document for his class on "ETHICS AND LAW ON THE ELECTRONIC FRONTIER." I have great respect for Hal--co-author of course of the seminal "Structure and Interpretation of Computer Programs"--so I responded favorably. He says that if I agree, which I did, he'll leave the Web doc available permanently, even after the class is over. More on what the URL is when it's ready; you can try http://martigny.ai.mit.edu/6095, which is his home page for his class. 4. Yes, I still have concerns about multiple, out-dated copies floating around. I worry about people going to great lengths to "Webify" the version they have, and then being unwilling, naturally enough, to repeat the work each time a new version appears. I worry that the "pre-release" version, the version 0.666 I released, will become frozen/encrusted/trapped in the silk of the aptly-named "Web" and thus forever be the de facto public release. (Unless I abandon the writing tools on my own computer and instead concentrate only on adding links and making changes to the Web version. This I just won't do.) 5. Rishab is especially anxious to get CompuServe, AOL, etc., versions available, or referenced in his review in "Wired." Thus, if any of you want to make this conversion, download the FAQ, or the chapters (from Karl's site), etc., and _agree to keep the version relatively current_, then this would be a Good Thing. (I'm not sure the advanced stuff in the Cyphernomicon, stuff about remailers, digital cash, etc., is the most useful stuff for typical "Wired" readers, given the questions the material will generate (as it does on our own list!), but this is not my call to make. Rishab says the editors at "Wired" want his review, so it's out of my hands.) Rishab says he needs to have the CompuServe/AOL/etc. addresses by 27 September, as his review is due on 1 October. So, not much time. There are of course major _size_ problems. I tried to e-mail the FAQ to one list member without ftp access who assured me his mail system could handle it--but it couldn't. And another list member uses a VAX, and only the first line would transfer. The compressed versions have also caused problems for some folks (though not to others, so I know the versions are kosher). Frankly, I shudder to think about the AOL and CompuServe folks, not to mention the Prodigy customers, trying to get a megabyte-plus file, or a gzipped file! So, if someone makes this available, please be careful to make sure it is really retrievable and readable on these systems. Thanks. --Tim May .......................................................................... Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)netcom.com | anonymous networks, digital pseudonyms, zero 408-688-5409 | knowledge, reputations, information markets, W.A.S.T.E.: Aptos, CA | black markets, collapse of governments. Higher Power: 2^859433 | Public Key: PGP and MailSafe available. "National borders are just speed bumps on the information superhighway."

1 0

HIT MEN
by Sandy Sandfort 17 Dec '03

17 Dec '03

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ SANDY SANDFORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C'punks, I'm stymied. I keep arguing the real world, and everyone responds as though this were a MUD game. David Merriman wrote: If the escrow agent has a reputation for reliability, honesty, integrity, etc, then the nature of the transaction is irrelevant.... I'm sorry, but this is patently ludicrous. This is not a MUD or MOO. We're not talking about game theory and the "iterated prisoner's dilemma." This is about the real world. I could cite hundreds of examples where folks don't think "the nature of the transaction is irrelevant." Tree huggers won't buy from McDonalds, irrespective of quality, because Mickey D's "encourage the destruction of the rain forests." Gun nuts won't do business with Taco Bell because it's owned by Pepsico, and they've supported gun control. There are "green" shoppers and "investors of conscience." Hell, Duncan won't even buy the fine chickens produced by Foster farms because Foster is in bed with the Klintones (well not literally... well maybe). The burden of proof is on those who would tell us that human nature will change over-night when crypto-anarchy comes on line. People *do* care about the morality of those with whom they do business. I see no reason to believe that will change. Anonymous murder for hire, is not easy. It isn't going to be any more a threat under crypto-anarchy than it is now. Please, let's get out of this Tabloid, the-sky-will-fall mind set. It's going to be okay. S a n d y Who is signing off this thread. What's coming is coming, no matter what we do or say. History will have the last word. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

4 4

(Fwd) Internet Security: Secure Communications Over Untrusted Networks
by Ian Farquhar 17 Dec '03

17 Dec '03

> of effective cryptographic systems. He will highlight the > problem of providing strong encryption for users of computer > networks while meeting the legitimate needs of law enforcement ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Has anyone else noticed how this exact phrase keeps popping up time and time again? I for one don't consider it axiomatic, and have grave doubts that LEO's do have a legitimate SIGINT requirement at all. It looks like the opposition are formulating their own catch-cries now. Ian. #include <std.disclaimer>

2 1

Tedious Chomsky arguments, some small remailer relevance.
by jamesd＠informix.com 17 Dec '03

17 Dec '03

chen(a)intuit.com (Mark Chen) writes an amazing article worthy of Chomky himself I note that Chomsky fans, like Chomksy himself, have no shame in lying bare faced in public. > There is no mention anywhere of moral superiority. There > is a statement of the demonstrable fact that "the editors > and the ambassador, in the not very distant past, have > supported racist murder on a scale that exceeds Amin's > wildest fantasies," referring, perhaps, to the (then) > .... One: Chomsky's claim clearly implies a claim of extreme moral superiority. Two: Your claim, and Chomsky's claim, is a flagrant lie Amin murdered three hundred thousand people, ate a few of them, and daydreamed out loud of extending a similar rule over all of black Africa. 2.1. Even if the Ambassador and the editors of the New York times had personally commanded the repression and man made famine in East Timor they could not have clocked up as many as Amin did in real life, let alone in Amin's "wildest fantasies". 2.2 It is completely false that the New York times supported the genocide in East Timor. They were sympathetic to, or tolerant of, the wrong side in East Timor, but, unlike Chomsky, they did not support genocide. Nothing the New York times said or did remotely compares with Chomsky's enthusiastic support of Pol Pot's genocide in Cambodia. Chomsky compared Pol Pot's genocide to the denazification by the French Resistance after world war II. When did the New York Times compare Indonesia's repression in East Timor to reconstruction after the American Civil War? More rationally, Mark Chen claims that I have failed to make my case that Chomsky is seeking to achieve a totalitarian state in America by democratic, constitutional, and institutional methods, as the National German Socialist Workers party did in Germany, rather than by revolutionary methods, as the Bolsheviks did in Russia. Certainly it is true that Chomsky never says in so many words "Hey let us imitate the great methods that Pol Pot used in order to make sure the will of the people prevails over the immensely powerful secret conspiracy by evil racist capitalist imperialists." But Chomsky defines peoples free choice to say one thing rather than another thing, to listen to one source rather than another source, to be "extreme coercion and control". With this definition, it obviously follows that exterminating those who engage in "extreme coercion and control" is an act of self defence. When a socialist argues labor theory of value, I know he is planning to rob me. In the same way, when Chomsky argues that speech is coercion, and choice is submission, I know that he and his pals in the government are planning to enhance our civil liberties by protecting us from that speech, and to enhance our lives by rescuing us from that submission. -- --------------------------------------------------------------------- We have the right to defend ourselves and our property, because of the kind of animals that we James A. Donald are. True law derives from this right, not from the arbitrary power of the omnipotent state. jamesd(a)netcom.com

2 1

OK, *THIS* IS MY LAST WORD...HONEST
by Sandy Sandfort 17 Dec '03

17 Dec '03

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ SANDY SANDFORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C'punks, That sneaky Jeff Barber has lured me out for one last word; cross my heart and hope to be assassinated by a digitally anonymous crypto-murderer. He wrote: I'm sorry that you're signing off this thread, particularly since it was you that challenged others to "disprove" your scenarios. That was not my main *challenge*. My shoot-from-the-hip scenarios were meant to be illustrative. I wanted Cypherpunks to use their powers for good, not evil. My primary challenge was to think up ways to defeat digital murders for hire. Why? Because: (1) Cypherpunks support strong crypto and other technological means for securing personal privacy. (2) We have enemies who are attempting to suppress such techniques by inflaming the general public with sensationalistic claims (e.g., PGP pedophiles, Net narcotraffickers, anonymous assassins). (3) To offset these tabloid bugaboos, it is in our best interest to solve these "problems" rather than to think up new and better ways to create crypto-chaos. Now everybody can either waste their time thinking up really neat-o keen-o ways to whack innocent spouses, business enemies and rich uncles, or you can support strong crypto and privacy by thinking up ways to mitigate the potentially negative side-effects of crypto-anarchy. Don't forget why we tune into the Cypherpunks Channel. Let's keep our eyes on the prize. I now say adieu to this thread for the second, and last, time. S a n d y ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

1 0

It's MEME time!!!
by jim＠bilbo.suite.com 17 Dec '03

17 Dec '03

I thought I'd jump onto the meme-creation bandwagon before it gets too full.. Cryptography - it's not just for governments anymore! --- Clipper and the Wiretap Bill: salvos in the War On Privacy! --- The price of security is continuous surveillance. --- Denning's Dilemma: Privacy or Oppression. There's no middle ground. --- Surrender Dorothy!

1 0

MATH: Brands' cash, Hal's post #2
by Karl Lui Barrus 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- This post gives numerical examples to go along with Hal Finney's excellent description of Brands' digital cash, posted earlier. If the math is too much, just remember the whole point: > Blind signatures are, IMO, the key to anonymous digital cash, and in fact > to many forms of anonymity. The ability to engage in mutual information > manipulation with another person, while guaranteeing that no linkage will > later be possible between the data exchanged and the results of that > calculation, is the foundation for interacting in a complex way without > losing any privacy... > Vicki wants to end up with a non-interactive signature on m', which is a > special transformation of m. To do this, she engages in an interactive > signature protocol with Paul, getting him to sign m... the result is > that she ends up with a non-interactive signature on m' because Paul was > willing to participate in an interactive signature session on m Continuing along: > Now for the mathematics. Recall the g is the "generator" of the group, > the base of all of the powers. x is Paul's secret key, and GX=g^x is his > public key. I will use g = 10, n = 17389 as in the previous example. Paul will choose x = 351 to be his secret key, so GX = 10^351 mod 17389 = 16987 is his public key. In addition, the message is m = 1994. > As the first step of the interactive protocol, Paul chooses a random w > and sends Vicki MX = m^x, GW = g^w, and MW = m^w. Paul chooses a random w = 666 MX = 1994^351 mod 17389 = 11740 GW = 10^666 mod 17389 = 7115 MW = 1994^666 mod 17389 = 13262 > The relationship between m', which is what Vicki will end up > with a signature on, and m, which is the number that Paul sees, is > m' = (m^s)*(g^t). Vicki chooses s = 3694, t = 1243 m' = (1994^3694)*(10^1243) mod 17389 = 10313 > the challenge c is calculated as the hash of (m,MX,GW,MW). Vicki > must transform these numbers so that Paul will not recognize them, but in > such a way that the mathematical relationships are maintained. > To do this, Vicki chooses two (more) random numbers, u and v (along with > s and t above). Vicki chooses u = 5192, v = 100 > MX' = m'^x = ((m^s)*(g^t))^x = (m^(s*x))*(g^(t*x)) = (MX^s)*(GX^t) > GW' = g^w' = g^(u*w+v) = (g^(u*w))*(g^v) = (GW^u)*(g^v) > MW' = m'^w' = ((m^s)*(g^t))^(u*w+v) = [...] = > (GW^(u*t))*(MW^(u*s))*(m'^v) MX' = (MX^s)*(GX^t) = (11740^3694)*(16987^1243) mod 17389 = 10710 GW' = (GW^u)*(g^v) = (7115^5192)*(10^100) mod 17389 = 12113 MW' = (7115^(5192 1243))*(11740^(5192 3694))*(10313^100) mod 17389 = 9314 > Using these, Vicki calculates her hash c'= Hash(m',MX',GW',MW'). c' = hash(10313,10710,12113,9314) = 7672 (some hash function I made up) > Now, the c she sends to Paul... > c = c'/u c = (7672/5192) mod 17389 = 323 [ 5192 c = 7672 mod 17389 --> 5192 c" = 1 mod 17389 --> c" = 3520 ==> c = c" 7672 mod 17389 = 323 check: (323 5192) mod 17389 = 7672 ] > Paul will ... calculate r = c*x+w. r = (323 351 + 666) mod 17388 = 9711 > [Vicki calculates] r' = u*r + v r' = (5192 9711 + 100) mod 17388 = 11800 > The resulting signature on m' is (MX',GW',MW',r') So the resulting signature is (10710,12113,9314,11800) Okay, that should be an actual example of the protocol, unless I messed up somewhere ;) I hope to finish going through Hal's third post soon. Karl Barrus klbarrus(a)owlnet.rice.edu -----BEGIN PGP SIGNATURE----- Version: 2.6.1 iQCVAwUBLoImUMSF/V8IjI8hAQGRmAP/RojMlpm8rnnx4K6c3GEHsBoQL7hIhdBB bTiwBhkXbi8ZhHsZJtX9mFceIhTK7yIxVsq9y17d2m5NghGME1qtIN+MjbbvwHfp j9S9fWwF6/mIiRvV9IM1a23IGhyZi0ZQASLKRiPlStjbcwv6QoGxZQuTyGOD8pSn hpoKosUFbqY= =EIjf -----END PGP SIGNATURE-----

1 0

Re: HIT MEN
by tcmay＠netcom.com 17 Dec '03

17 Dec '03

[For some odd reason, I'm only seeing Sandy's posts on this subject, not those of Merriman, Barber, or Mays.] >Patrick missed my irony. Murders can't hurt any escrow's >reputation. To do so, they have to admit to being murderers who >were stiffed by the escrow. NOT BLOODY LIKELY. For murder >escrows, a positive reputation is meaningless. They can't--nor >can anyone else--risk exposure of such negative information. >Escrows that admittedly engage in abetting criminal acts can have >NO MEANINGFUL REPUTATIONS. That's just not true. Currently, many organizations--some known widely, some shadowing, some essentially anonymous--count on their reputations for being efficient, cold-blooded murderers as a means to scare off competitors, increase their market value, etc. From Jamaican gangs who execute the families of their targets to the CIA's Phoenix Program operatives who mutilated their victims horribly, such "reps" are highly useful. To paraphrase Sandy, "BLOODY LIKELY." Admitting to engaging in a criminal activity is not at issue--remember, all parties are cryptographically protected and what they "admit" to doing cannot reflect upon their physical/legal identities, only their digital reputations. And those who contract for such services, via their pseudonyms, can "admit" to wanting to buy such a service. (The issue of whether a well-respected nym like "Locke" would want to publicize a failed hit on his arch-enemy "Demosthenes" is a separate issue, which I won't conflate with this one.) Gambling is illegal in most places, unless run by the state. And yet people gamble, illegally. They use bookies. Bookies who are doing illegal things, as the gamblers are. And yet if they get stiffed by a bookie, which _sometimes_ happens, they tell their friends, family, etc., and the reputation ripples spread. Taking Sandy's "For murder escrows, a positive reputation is meaningless. They can't--nor can anyone else--risk exposure of such negative information. Escrows that admittedly engage in abetting criminal acts can have NO MEANINGFUL REPUTATIONS." argument, are we to assume that this applies to illegal betting? That stiffed bettors won't speak up because there are "Escrows that admittedly engage in abetting criminal acts can have NO MEANINGFUL REPUTATIONS"? Crypto barely changes things, except to make outside interference less likely. If, for example, Black Unicorn offers to transfer 100 Ghost Marks to Pr0duct Cypher, for some C programming, and he doesn't feel he got his money's worth, he can publicize it. Maybe we believe Black Unicorn, maybe we don't. Maybe we ask to hear Pr0duct Cypher's side of the story. Maybe we suggest that SOLONg act as a third party escrow agent. And so forth. Not perfect, in some abstract sense of ultimate truth always coming out, but reputations do indeed matter. And whether the deeds contracted for are heinous or noble depends on your point of view. To William Colby and the Viet Cong, the taking of ears and other body parts by the Phoenix assassins was a fearsomely reputable thing to do, regardless of what the 4H Club in Skokie might have thought about it. If I contract with "Sandy's Salvage--You Pay, We Slay," I want to hear that they've got some satisfied customers. (Yes, flooding of reputations is an issue. Same issues as arise in DC Nets. Same kinds of solutions.) Again, I've written too much, so I'll stop for now. --Tim May .......................................................................... Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)netcom.com | anonymous networks, digital pseudonyms, zero 408-688-5409 | knowledge, reputations, information markets, W.A.S.T.E.: Aptos, CA | black markets, collapse of governments. Higher Power: 2^859433 | Public Key: PGP and MailSafe available. "National borders are just speed bumps on the information superhighway."

2 4

Unix Expo Disk Distribution Plea
by rarachel＠photon.poly.edu 17 Dec '03

17 Dec '03

Forwarded message:

1 0

Re: Chomsky Arguments / Redefinitions
by Jeff Gostin 17 Dec '03

17 Dec '03

jamesd(a)netcom.com (James A. Donald) writes: [Blanc's statements Re: this argument about Chomsky, whomever the hell he is.] > By the way when I called this thread "more tedious Chomsky stuff" > I was perfectly serious - This thread has been beaten to death > time and time again, and many people have a kill file that > automatically kills anything with Chomsky in the header. I haven't had the pleasure of reading this thread before, but, thanks to the wonder of killfiles, I won't have to again. *plonk* to Threads-From-Hell

1 0