December 2003 - cypherpunks-legacy

Re: Another problem w/Data Havens...
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

> I could write a procmail recipe and a script in about an hour to > automatically secret-share-split and redistribute the incoming submission. > If the authorities attempt to indict you for possessing illegal > information / kiddie porn / whatnot, they have to prove that you > interfered with the automatic redistribution process and examined the > contents of the submission. If you in fact did not look at the submission, > they would have a difficult time doing so. The problem with a procmail script, unlike the ugly on-the-fly SMTP-like splitting method I proposed, is that the suspicious message is in your mail system, intact, and if they nail you before you can dispose of it, they can prove you had it. (Even though procmail does stuff quickly, sendmail still accumulates the material before delivering it, doesn't it?) They still have to prove that there's something illegal or forfeiture-supporting about the way you possessed the message, but by then you're hauled into court and your computer is stolen. The Feds argued in the Steve Jackson Games fiasco that intercepting unread mail in mailboxes is different than intercepting it in transit, and while we can hope that will be overturned, it's better not to be the guinea pig. If your machine never has more than a few bits of the message, they can't catch you in possession of the message - they can only argue that bits of it flowed through your machine, and that's much harder to build a case on, and given the abuse the judge gave the Bad Guys in the SJG case, it's probably easy to prevent them from keeping your machine as evidence, or at least get it back. Your paranoia may vary.... Bill

1 0

Re: Scientologys Attempts
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

> > > who has the authority to remove alt. groups? jamesd writes: > Anybody has the power. Nobody has the authority. More precisely, anybody has the power to send a rmgroup message, just like anybody has the power to send a cancel message. Aside from ettiquette and ethics, the question is whether most of the news servers out there will respond to them, which is a local decision. The last time I ran a news server was in B News days, so your mileage may vary, but responses to newgroup and rmgroup messages can be turned on and off. I used to leave them turned on; not only did new groups appear on my system without manual intervention, but the Great Renaming took care of itself automatically on my box. > [...] This is a classic case of net abuse. Yep. rmgroup wars happen occasionally on the net, usually about alt.joe.newuser.die.die.die and other spam groups run by (or about) people with more spare time than taste. The difference here is that it's part of an organized group-sponsored censorship attempt, rather than an individual squabble, so it retains some degree of classicity. Bill

1 0

On DH public key crypto
by eric＠remailer.net 17 Dec '03

17 Dec '03

Diffie and Hellman did have a public key cipher based on matrices in their original paper; they knew at the time it didn't work. This is not the same algorithm as the D-H key exchange protocol. Eric

1 0

JP vs Homer
by Nathan Zook 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- A side note: "The ultimate hacks of Wall Street." Now THERE is an idea... You think Donald Trump could see the usefulness of e*? Ivan Boskey? (sp) To the point: JP's withdrawal, and Homer's thread point to something that I've felt for a while: The classification of our tools as munitions has an important point--they should not be used by children. A.S.R not withstanding, (such a person has had their childhood robbed) what we are doing is _very serious_. Sometimes we talk about pulling down the entire structure of our governments and economy as if it we a Sunday afternoon parlor exercise, or a pipe dream. It is neither. We are on the verge of a restructuring of our society. Cypherpunk projects are the bleeding edge. Implementation calls for serious, considered efforts. Childish behavior is not only uncalled for, it must be anticipated and prevented, or as a minimum, found and fought. I have long been concerned that the remailers are far to vulnerable to either childish abuse or TLA attacks. The former is reaching the stage of threatening to swamp us, either by resource depletion, or by stigmatizing our efforts. There is little, if anything, we can do to stop a chained, PGP'ed mail bomb, "This is mail bomb number XXX. Boom!" It is therefore in our best interest to not encourage children to send such messages in such a way. I believe Homer's message was erroneous for this reason. We do have considerable interest, however, in maintaining a positive image. Positive image is a multifaceted jewel. We must look to preserve it as much as possible while pursuing our goals. We are capable of controlling mail bombs, for instance, in the following way: Take an incoming message, capture From: line. Strip header. MD5 body. Add to sorted table [From: MD5(message) date]. Check for repetition of first two fields. If reps = 1, forward message. If reps = 2, send message to From: "Possible error. Two copies of message <message> received." If reps = 0 mod 5, send letter to postmaster@From:. "Possible mailbomb or spam. <reps> copies of <message> received from <From:> at your site in the past week." Clear table of entries more than a week old every midnight. If all remailers did this, then no matter where the net was entered, the messages would be rejected. And spammers/bombers would be spamming/ bombing their own postmaster. Probably a bad idea. If not all did it, then add special handling to hit remailer-operator@. This would encourage the operator to auto-magically handle the spam-bomb himself. Dealing with huge files and/or TLA's is a topic for another day. Remember: spook(a)cia.gov, snitch(a)fbi.gov, and cracker(a)nsa.gov are probably all listening, maybe even denning(a)aol.com. Is it a stretch to consider that _they_ might spam-bomb us, trying to discourage the best? Finger or request keyserver for PGP 2.6.2 (tm) key. PGP<->Mail/News installation incomplete. Factors for modulous are not proven primes. Key may be far weaker than expected. Encode at your own risk. Key ID: 14712B4D 1994/12/26 Nathan H. Zook <nzook(a)bga.com> Key fingerprint = 44 B3 D8 66 3D 55 1E 2E F8 92 22 A6 33 8C DE 24 -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQEVAwUBLx3pBHmgMs8UcStNAQGuhAf+KGYJnYRPAODHvjF4CmakZ1oZKDo2BiFQ 5nkphXtoTgADVQM8XlYhqW3MTKhKo30dfZ0/GMLdaZ+4SCM85thqNvLW1klaandk tuAF4NBTx8f5rb3ufm9iktXv907smf6aUZAwz0T4cl630+/dVPI2Jkit0jid5KmD TzB/bFr1lM7Vp2ZoR+Y6/geYYNx06OMvLl62hTpGP2KtusFs1erRuYXj8UJkXd77 JNnTLjrPdLhx6iIQ6IxNXAISnmIgXPGFVRto/EEgFa8XWUoOBeuRdBGhT4eoNAjX 45YxdJrmlSPxHZDu8P2rldfays31qPCY7qJtSKBdRzYqqyz8LT+4YQ== =luXm -----END PGP SIGNATURE-----

1 0

Factorisation and Discrete Logs (was Re: EE Times on PRZ)
by dwa＠mirage.svl.trw.com 17 Dec '03

17 Dec '03

strick wrote: > > DH uses "discrete log" as the hard problem, and very straightforward > mathematics. > > RSA uses "factoring" as the hard problem, and a very clever back door. > > How do you decide if one is based on the other? > > # public-key technology with Diffie-Hellman public-key in particular, which > # (as I understand it) is not particularly secure. > > It's still up in the air, isn't it, whether the discrete log or > factoring is the harder to crack. My intuition is they're the > same hard. > > I know of no problem with DH that RSA doesn't have similar problems. > > strick It seems to me that factoring a large number is no harder than finding a discrete logarithm. Assume, for the moment, that an efficient method of computing discrete logs has been discovered, rendering systems like Diffie-Hellman key exchange unusuable. I contend that RSA is now equally unusable. The following variant of the Pollard p-1 method should provide an efficient factorisation method for an RSA modulus, say N. Choose, at random, "a" such that gcd(a,N) = 1. Compute x such that: a^x = 1 (mod N) [ Discrete log time! ] Partially factor x; say x = f * f * f ... where f is not necessarily prime. 1 2 3 n Note that it is usually easy to partially factor a "random" large integer. Simply using trial division up to some limit; or, at worst, pollard rho or pollard p-1 (on x) should suffice. If you're truly unlucky, pick another value for a. Compute: M = a^(10000! * f) (mod N) Where f is some partial factor of x. gcd(M-1,N) should yield a non-trivial factor of N. If it doesn't, another choice of f and/or a should work. I'm by no means a professional mathematician, but it seems that this scheme should work. Comments, anyone? Dana W. Albrecht dwa(a)mirage.svl.trw.com

4 3

Re: Factorisation and Discrete Logs (was Re: EE Times on PRZ)
by Hal 17 Dec '03

17 Dec '03

From: dwa(a)mirage.svl.trw.com (Dana Albrecht) > Choose, at random, "a" such that gcd(a,N) = 1. > > Compute x such that: > > a^x = 1 (mod N) [ Discrete log time! ] DH uses prime moduli, I believe. Solving the DL problem sufficiently to break DH may not let you solve it for composite moduli, not without knowing the factors. Hal

1 0

Re: TEMPEST
by pstemari＠erinet.com 17 Dec '03

17 Dec '03

At 12:32 PM 1/18/95, Timothy C. May wrote: > ... And if nothing is seen with our $10K of equipment, what does this >prove against an attacker who can easily afford to spend 20 or 30 >times that amount to equip a van? > >Cypherpunks have been exploiting technology that is comparatively >_much cheaper_ and which changes the equation. > > ... In general, for reasons many of us have written about here before, and in >particular, because I think spending $10,000 to prove what we already >know--that RF emissions can be detected and demodulated--is a poor use >of money. That $10K would go a long way to getting PGP Phone finished. ... I'm not sure how encryption chages the equation if that van on the street can read the data off your screen and/or printer. Just as cryptanalysis is a necessary component of good cryptography, some sort of VanEck analysis would seem to be required in order to evaluate the utility of methods to block it. What exactly would one need to do to block Van Eck monitoring? I've seen computer rooms that were completely lined with copper sheathing. Is it really necessary to go to that extent? Will aluminum foil and power line conditioning handle it? You can't really tell how effective counter-measures are unless you have some way to see what reduction in emitted signal they provide. --Paul J. Ste. Marie pstemari(a)well.sf.ca.us, pstemari(a)erinet.com

1 0

The Remailer Crisis
by tcmay＠netcom.com 17 Dec '03

17 Dec '03

I don't think I've used "crisis" in a thread I started, so this tells you what I think. The remailers are stagnating, only very slowly adding important features, and the number is not growing...in fact, it's been shrinking. This, as the remailers are under attack. Editorialists are railing against the dangers of anonymity and remailers, and the "Church of Scientology" is threatening lawsuits against remailer operators unless they block certain newsgroups. (Lots of issues, clever workarounds proposed, etc.) Now whether or not the Church of Scientology, or RSADSI (for the RC4 code remailing), or anyone else could legally _win_ such a case is not the issue. Not many remailer operators would be able to mount a defense...they are not, frankly, folks heros as Zimmermann is--and even PRZ is struggling to raise money. Maybe the EFF (ugh) would leap into the fray, but I doubt it. Nope, I fear that the pressures that have driven several remailers to halt operation are just the tip of the iceberg. (Some remailers have quit because of words from their sysadmins, semi-threatening conversations with Brad Templeton--Hi, Brad!--, and so on. Real letters from real lawyers would have a profoundly more chilling effect.) We've debated this stuff many times, but the numbers of remailers are shrinking. Raph Levien summarized the situation nicely in a post to alt.anon-server: "We're down to fifteen remailers now, many of which are unreliable. I would say the remailer net is not in good shape right now. Fortunately, I know of a number of plans to bring new remailers up, as well as restoring old ones." (I hope Raph can share with us soon what some of the plans are, unless he fears revealing this will aid our opponents.) I could see the whole remailer system imploding. A few threatening letters, especially if sent by real lawyers, could demolish the few U.S. sites. And with the Netherlands toying with proposals to ban or limit private encryption, having Holland as the main remaining host for remailers would be rather precarious. We need lots more remailers, in many more countries. What can be done? * We need to get the number of "solid" remailers up from the current dozen or so up into the _hundreds_. * I favor separating the "account that remails" from the "owner of the site," as I have argued in vain in the past. (Example: a willing site gives out or sells many accounts...each is legally separate, and each must be legally challenged. My longer posts dealt with this.) (The owner of the site/machine can take a "hands off" attitude toward what his customers are doing in their accounts. This doesn't stop pressures from being applied, but it slows them down, and (probably) better insulates the owner from legal charges.) * Traffic needs to be increased. Remailers should probably go to constant padding traffic, to do this. Exhorting end-users to use remailers more won't be enough. * More offshore sites are needed. So far, only Canada and the Netherlands are offshore hosts (and Canada is effectively part of the U.S.). Some sites in Russia would be dramatic. Ditto for Asia, South America, etc. * The whole issue of "remailer businesses," with all the usual issues of digital postage, stamps, coupons, etc, has to be resolved. That is, we need to get some movement here. Most remailers are run as projects by students on machines they don't control, or that they can be pressured on, or with a committment to stay in business that will evaporate too easily. A business, with business responsibilities, is usually a more stable solution. I think we need to set some rough targets and brainstorm on how to get to them. For example, the Linux mavens could tell us if Linux-based remailer boxes could be hung on dedicated connections to The Little Garden network, as a specific example. A "black box remailer" such as several folks have suggested (Chaum (in 1981), me, Eric Hughes, others) might be buildable for under a grand. We could ask here for contributions, and might even raise this amount of money. Then each of us who contributed could have "accounts," maybe several of them. Imagine 100 or more "remailers" all on this one machine--I'm deliberately ignoring the security issue for now. Little Garden has stated categorically that they will not tell users what they can or can't do with their machines (though I can think of some cases where they might have to, as with spamming, etc.). Anyway, you can see where I'm headed. My big fear is that the one really major achievement of the Cypherpunks group, the remailers, are not increasing in number and could be knocked out all too easily with some legal attacks. It's time to get cracking on this crisis. --Tim May -- .......................................................................... Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)netcom.com | anonymous networks, digital pseudonyms, zero | 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. Cypherpunks list: majordomo(a)toad.com with body message of only: subscribe cypherpunks. FAQ available at ftp.netcom.com in pub/tc/tcmay

1 0

Internet, spamming, etc.
by rparratt＠london.micrognosis.com 17 Dec '03

17 Dec '03

> From owner-cypherpunks(a)toad.com Wed Jan 18 17:00:14 1995 > Date: Wed, 18 Jan 1995 07:54:00 -0800 > I also think this is the one great flaw in the design of the Internet; > namely, that the sender has all the control over what packets flow > over the net. A receiver can ask for a slowdown or cessation, but > there's no obligation to do so. This will be, if anything, the > limiting factor in scalability of the internet. In theory, yes. However, almost all Internet protocols are TCP/IP based. The receiver of a TCP connection can choose not to accept the connection, or to drop it at any time. The window protocol keeps the sender from transmitting faster than any part of the connection can manage. (How do you think ftp transfers between sites with disparate connection speeds would work otherwise?) One could theoretically have a package that sprayed UDP packets at a particular IP address, or even have a modified TCP that ignored disconnects. I think most service providers would regard using such code as being on a par with running a program that tried to telnet sequentially to all known IP addresses, trying common passwords on each. -- Richard Parratt

2 1

Re: Internet, spamming, etc.
by rparratt＠london.micrognosis.com 17 Dec '03

17 Dec '03

> > One could theoretically have a package that sprayed UDP packets > > at a particular IP address, or even have a modified TCP > > that ignored disconnects. I think most service providers > > would regard using such code as being on a par with running > > a program that tried to telnet sequentially to all known IP > > addresses, trying common passwords on each. > > Code exists to do all of the things that you've mentioned. > > That was the original point of the discussion. No doubt it does. But my assertion is that most service providers would cut you off if you tried running such code, especially the latter case. -- Richard Parratt

1 0