December 2003 - cypherpunks-legacy

SNI_ffs
by John Young 17 Dec '03

17 Dec '03

3-30-96 TWP reports on an Argentinian hacker who penetrated a slew of sensitive networks and how the FBI tracked him by getting a warrant to run a sniffer-filter at Harvard, the POE. The hacker's dad, a Lt. Col., barks, "The Yankees don't have the slightest idea about security. If a kid can enter, they should be ashamed to admit it." The Feds snort, "The case underscores the vulnerability of computer systems worldwide, a preview of a coming era of cyber-sleuthing, what computer crime-fighting will look like in the coming years." And that it shows they can catch tresspassers without violating other users' privacy. Banisar sniffs, "The case shows that there's a need for more encryption." SNI_ffs

2 1

Chaumian ecash without RSA
by David Wagner 17 Dec '03

17 Dec '03

I've always seen Chaum's anonymous ecash system described in terms of RSA. RSA has this ungainly patent which probably will be around for quite some time, yet the Diffie-Hellman patent expires pretty soon. With that motivation, here's a Chaumian anonymous ecash protocol based on Diffie-Hellman. Take a publicly known group G and generator g; breaking Diffie-Hellman and taking discrete logs in this group should be hard. For instance, G might be (Z/pZ)^*, the integers modulo a prime p. The bank picks a secret value k, and publishes g^k. To withdraw a coin, Alice picks an x, sets y = x | hash(x), [ | is concatenation ] chosen so that y is in G. Alice chooses a random secret blinding factor b, sends to the bank A->B: y g^b, and the bank returns B->A: (y g^b)^k, debiting Alice's account. Note that this is a (blinded) Diffie-Hellman key exchange with public exponentials g^k and y g^b; the bank returns the exchanged "secret". Alice unblinds this value, computing z = (y g^b)^k (g^k)^{-b} and now c = (x,z) is a coin in the digital cash system. Note z = y^k. We use the traditional online clearing protocol; to deposit the coin, a shop S sends S->B: x, z. The bank checks to make sure the coin hasn't already been spent, and then computes y = x | MD5(x), checking whether y^k = z. If equality holds, and the coin hasn't already been spent, then the bank credits S's account and adds the coin to the list of spent coins. This is just the same old Chaum anonymous ecash protocol, except that I've replaced the RSA operations by Diffie-Hellman ones. It's a lesser-known fact that you can blind a Diffie-Hellman key exchange just as you can blind a RSA signature. The security of this protocol depends on the intractibility of breaking Diffie-Hellman. In particular, given public exponentials g^k and y = g^m, for k,m unknown, it must be impossible to compute g^{km} = y^k. Furthermore, this protocol depends on the hash function being one-way and possessing no interactions with Diffie-Hellman or modular exponentiation. Comments?

1 0

Re: SNI_ffs
by John Young 17 Dec '03

17 Dec '03

Responding to msg by declan+(a)CMU.EDU ("Declan B. McCullagh") on Sun, 31 Mar 10:42 AM >One gave me the impression the DoJ had to develop >custom hardware and software for this "Internet >wiretap" done without Harvard's direct cooperation. DoJ is probably cutting spying-sensitive Harvard some slack, or slyly crowing about setting up the Crimson butts with a promise for deniability. Is it possible that the CFP chit-chat revved that Janus-spin, practicing for more pervasive cyber-sleuth slathering of wannabe L&O insiders? ---------- The New York Times, March 31, 1996, p. 20. First Internet Wiretap Leads to a Suspect [Excerpts of story not in the TWP] Stephen P. Heymann, a Federal prosecutor in Boston, said investigators had worked with Harvard to determine a method of tracking the suspect that would protect the privacy of legitimate users. He said that the Harvard system had 16,500 accounts and 13,000 users and that about 60,000 E-mail messages each day moved in and out of the area where investigators were looking for the intruder. Mr. Heymann said investigators had used a high-speed computer to check for 10 to 15 key words that matched the intruder's profile. If they were not sure if an electronic communication containing a key word was Mr. Ardita's, the investigators looked at 80 characters on either side of the key word to make that determination. Mr. Heymann said investigators believed that only twice had they read a complete message that was not Mr. Ardita's. [End excerpt]

1 0

Re: [NOISE] Cable-TV-Piracy-Punks
by jim bell 17 Dec '03

17 Dec '03

At 11:59 PM 3/30/96 -0800, Mike Duvos wrote: >A few more hopefully short comments... > > Why not? If the card knows its own key, then someone else > > can probably get the key out by some nasty mechanism. > >There is no physical difference between cards. The key >information is stored in EEPROM, and the links which permit the >EEPROM to be written are burned after programming is complete. >The EEPROM data is then only accessible to intimately associated >circuitry in its vicinity. > >Presumedly the state of the EEPROM cannot be deduced by any >external examination of the card, and any attempt to >incrementally abrade the card down to the relevent circuit >elements should completely obliterate the minute charge >differences which represent the data. > >At least, that's the theory. The Europeans trust this technology >well enough to let it represent real money, so presumedly they do >not consider hacking a possibility. > >Perhaps our resident VLSI and Alpha Particle expert, Timothy C. >May, could give us a guess as to whether Perry's "Nasty >Mechanism" is more or less likely than Maxwell's "Daemon." I don't know what Tim May will tell you, but over 10 years ago a technology was developed which is something like a scanning electron microscope, however with very low beam energies and is designed to be able to scan a chip and quantitatively measure the voltage at various/all points on the chip. It can be thwarted by a thick coating on the chip, but most organic coatings can be removed with a "plasma asher," a chamber designed to remove photoresist coatings on chips. Jim Bell jimbell(a)pacifier.com

1 0

Re: Electronic locksmiths are watching you (Belgium's ban onPGP)
by jf_avon＠citenet.net 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Dave Del Torto <ddt(a)lsd.com> wrote: >At 1:25 am 3/30/96, jim bell wrote: >>We should be particularly suspicious of any hint of a pan-European ban or >>control of encryption, because that is exactly the kind of development that >>could usher in a secretly-negotiated treaty that might be argued to be >>binding on the public. [elided] >Cypherpunks like Jim need to keep doing their homework before they make >such quasi-factual statements. Jim seems to analyse a situation in terms of broad principles, while Dave seems to focus more on specifics that would disprove Jim. I agree with the Dave's last phrase in the sense that our models of reality should indeed be rooted in reality. But considering that in most european country, you are recognized guilty unless proven innocent, it is only non-contradictory to arrive at Jim's conclusions. The basic psycho-epistemology at work there is implying that. Maybe that on the instant, ugly things are not happening, but nevertheless, the basic principles held about man, his nature and therefore, about how he should be treated leads, by their own logic, to ugly things. JFA -----BEGIN PGP SIGNATURE----- Version: 2.6.2i iQEVAwUBMV5p5siycyXFit0NAQGSjAf/abwWK0DoLESROTPjrNgEcX66t+zFJ8X8 c1oMvDqXsUye0WKibkgiO+yQ7Cjt2Zxo1bD9luH9boX0vUPhDxtUclw64HJ7SOGK WVVizbZLYmz5dSoOO8+0JpJUyFFgdI9LOMPjD1wIH84YwC8zjjSx/ZCPoEzt0lux pP2l57+G3aVwvxCXBD0lm/Uwtafn3i35ZsOyuKRq802Is8BhFH1hka+SzUyuU1i2 Tcw7/WbqSklvGSkVkBxKMJot7GbBTEQMNqVH7oLECY1oTKafEnXzAKihKJYef49w AJJpsOr5LQ9hyrKzAL9Mhzpx/DrrnIQ3BxMb7VgtbHSMjaEMJ+rS0g== =3bz4 -----END PGP SIGNATURE----- Public Key at http://w3.citenet.net/users/jf_avon Jean-Francois Avon <jf_avon(a)citenet.net> 2048 bits key ID:C58ADD0D 1996/03/01 fingerprint=52 96 45 E8 20 5A 8A 5E F8 7C C8 6F AE FE F8 91 Unsollicited commercial e-mail will be proofread at a rate of 165 $ U.S. per hours. Any sender of unsollicited commercial e-mail will be considered as to have accepted the above mentionned terms.

1 0

Medusa's Tentacles
by vasudeva＠mindport.net 17 Dec '03

17 Dec '03

(Please forgive the uninformed method of my inquiry.) Would any here happen to know where I might come across a later version of 'Medusa's Tentacles?' I've got the first release, alpha/beta 1, but, as many probably know, it carries a few bugs, such as not allowing purging of fuzzybases... ...vasudeva ___________________________________ 'Hot water is the revolutionist's element. You clean men as you clean milkpails: by scalding them.' -GBS PGP messages encouraged - look at MIT keyserver for public key. ___________________________________

1 0

No Subject
by David Loysen 17 Dec '03

17 Dec '03

> >The Navajo/Na Dene codetalkers (WW II) developed a real arcane jargon, >so I was told. Maybe it was because the conditions of war were >completely different than their language's environment so they >were forced to invent words, or maybe they thought it was a good >idea, or whatever. But I believe transcripts of their transmissions >are often mostly unintelligible to native Navajo speakers who weren't in >the know. > > > It is my understanding that the codetalkers invented very few new words, they simply combined the words they already had to create descriptions of things that the language was never meant for. One of the reasons they were so successful was that more than one phrase could mean the same thing. For example, both "A flock of eagles with fire in their bellies is coming from the rising sun" and "Many birds are flying from the east to rain fire on you" Could reliably be translated to "There are bombers coming from the east" Add some new words for specifics that you needed (like altitudes and compass directions) and the codetalkers presented the bad guys with a language that was completely unrelated to anything they had heard before. Also (I copied this from RSA) "The Navaho language is so difficult to learn and its linguistics are so complex that it is virtually impossible for a non-native speaker to counterfeit its sounds. Furthermore, Navaho seems to have no linguistic connections to any other Asian or European language. Consequently, at any given time, there are only a few thousand people capable of speaking the tongue. For these reasons, the U.S. military made extensive use of hundreds of Native American codetalkers . During World War II, Navaho codetalkers relayed operational orders in the Pacific theater with a level of security that was unattainable by current encryption algorithms. The Japanese signal corps task was further complicated by the codetalkers liberal mix-in of Navaho and military slang resulting in a communications network so secure that it was, in fact, never compromised by Axis powers." In my stone age level of Crypto understanding I would liken this to having a public key that was the Navajo culture and an algorithm to process it that only runs on the human brain. Makes me wonder when somebody will set two AI computers down and tell them to invent a code we can't break. dwl(a)hnc.com David Loysen 619-546-8877 x245

5 4

Re: Crypto CD UpDate
by JR＠ns.cnb.uam.es 17 Dec '03

17 Dec '03

>> Hey, I just got Appl. Crypt. 2nd Ed. yesterday. Meanwhile with >> the 1st Ed. and the code I had in my CDs I could very well manage with >> most things, protocols, algorithms... Of course, the version of Crypto++ >> I have there is outdated now, but the algorithms are still valid, and >> I still have much more algorithms there than the current version has, and >> I can always plug in or adapt a new one should I need to. > >Would you like to suggest some of your favorite algorithms not in the >current version of Crypto++ for inclusion in the next version? > >Wei Dai > Point taken. Well, as it is now, it is good enough for my current needs. And it's a great work you've done. What I was referring to instead was actually what I don't want or expect in a "good" library: simple, extraneous, old, maybe breakable algorithms, things no longer useful... they make a great material for studying and learning, and sometimes to demonstrate some colleagues how simple some approaches are. These materials, that I wouldn't use in a decent application are still usefull for me from time to time, and are nice to have at hand. Personally, I would prefer some algorithms not to be in Crypto++ less someone use them confident that if they are there, they must necessarily be good, although they might be useful for learning purposes. Nevertheless, as soon as I find something I think interesting to add, I'll let you know, or even contribute it myself. And thanks again for this fine piece of code! jr

1 0

Re: Crypto CD UpDate
by JR＠ns.cnb.uam.es 17 Dec '03

17 Dec '03

>If pieces of the source/executable are digitally signed, you have a basis >for some degree of trust. (My pgp came with a detached signature. A bit >self-referental, but at least a start.) > >Regards - Bill Agreed, but it imposes further restrictions: it's OK if you can put PGP digitally signed by prz, but not all packages will be available signed from their authours, especially compiled for various platforms. OTOH, most unknowledgeable people will trust almost anything (they are already doing so when downloading java applets). And it would do a great work to spread knowledge about cryptography. Which is a Good Thing. All in all, I think it is a good idea, but addressing the general public will require quite some work, and the 'connoisseurs' might either do as Tim (only use the net) or just make their own mass-store, removable, thingies. Stil I'd bet many people will be eager to get a mirror of the major sites on CD. jr

1 0

Re: [NOISE] Cable-TV-Piracy-Punks (fwd)
by Jim Choate 17 Dec '03

17 Dec '03

Forwarded message: > > Presumedly the state of the EEPROM cannot be deduced by any > > external examination of the card, and any attempt to > > incrementally abrade the card down to the relevent circuit > > elements should completely obliterate the minute charge > > differences which represent the data. > > They aren't immune to the laws of physics. If it can be put together, > it can be taken apart. I can even surmise HOW it can be taken apart. You wouldn't even have to take it apart. Just subject it to analysis using SQUID's. Using this technology you would not even have to physicaly touch the card, let alone remove any parts of it.

2 1