December 2003 - cypherpunks-legacy

bad news from Georgia (the Republic of)
by Stanton McCandlish 17 Dec '03

17 Dec '03

This might be interesting, though international, to many of you. Seems a certain Internet node is perceived as vital to the Georgians in their struggles. Forwarded message:

2 1

Dangers of skipping...my solution for now
by tcmay＠netcom.com 17 Dec '03

17 Dec '03

Earlier I wrote about the dangers of skipping too many messages, of a kind of "response cluster" when the same question gets answered many times. Somewhat making their point, several people have responded to me privately saying they try to answer questions in e-mail and then let the questioner summarize the responses in public, if he wishes. Advantages of this approach: - reduces repetitive answers (saves list bandwidth) - questioner can synthesize from all answers Disadvantages: - others may want to see the answers and may, under this approach, flood the list with "me too" lines (seen in many newsgroups, when "send answer in e-mail" gets met by "Hey, I'd be interested, too" follow-ups). - others may not know the question has already been answered, resulting in much wasted time (e.g., 30 private e-mail messages all addressing the Navajo code talker question). (Aside: there's a standard game theory example on how often one should help...it works better in game theory than in practice.) Finally, to me the most serious problem with moving more responses to private e-mail is that the discussion group aspect of the list gets lost. So I still don't have an answer. However, as an experiment, I plan to try to move as many of my responses off the list proper as possible. Things that are clearly of interest to a reasonable fraction of the entire list, I'll still post to the list. (Note again, as many of us have described, the Extropians list software allows threads to be ::excluded, as well as authors, so repetitive traffic can be reduced. It takes a minute or so to send off the ::exclude, so many of us prefer to just hit the "D" key promptly. Still, a portent of things to come.) -Tim -- .......................................................................... 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: 2^756839 | Public Key: PGP and MailSafe available. Note: I put time and money into writing this posting. I hope you enjoy it.

1 0

FLAME: breaking DES
by an42035＠anon.penet.fi 17 Dec '03

17 Dec '03

I have been following this argument, and I cannot let this post slide without comment. As near as I can tell, Karl pointed out an attack which shows that double encrypting really doesn't buy you much security over single encrypting. Perry has chosen to ridicule this for impracticality. (Incidentally, I notice a slight error in a followup post by Karl, undoubtedly a typo on his part. The number of chosen plaintexts needed by differential cryptanalysis is 2^47 and not 10^47. While still a large number at more than 100 trillion, it is at least much smaller than 10^47!) This fact seems to have eluded Perry, despite his apparent expertise. There are other meet-in-the-middle attacks, such as the one Chaum and Evertse use against a DES variant (fewer number of rounds), described in "Cryptanalysis of the DES with a reduced number of rounds" at CRYPTO '85. The meet-in-the-middle attack Karl posted is by Merkle and Hellman, from the 1981 paper "On the Security of Multiple Encryption", Communications of the ACM. Strange, but I don't recall Perry asking if Merkle and Hellman were worried that others might think they beleived this. Incidentally, the complexity of factoring a 2000 bit number is on the order of 2.4 X 10^43 steps. >Whenever Perry Metzger defends an opinion of his well, somebody whines. >Whenever Perry Metzger disagrees with someone's arguments, someone else >crys "unfair! He's using his WITS on me!" >Whenever Perry Metzger has to stop and demonstrate something as simple as >arithmetic to end an argument, someone cries "foul! sarcasm! Hey, he's not >playing nice!" Well, if Perry were to have the BALLS to make impracticality arguments against the experts who actually created the protocols and methods, instead of taking cowardly pot-shots, maybe we wouldn't think he's full of shit. The attack is impractical, and if Perry weren't so strung out he could have written a coherent post stating despite the cleverness of the attack, it is out-of-reach with today's technology, with some calculations showing this. In the 1980 paper by Helleman "A Cryptanalytic Time-Space Tradeoff", Helleman suggests it is possible to pre-compute and store 2^56 results, and use this information to lookup keys. Cost: estimated at $5 million. (Although this is for singly encrypted text). Funny, but I don't recall any dissenting papers by Perry on this subject. He didn't ask if Helleman really wanted people to think he beleived this. >Y'all need to develop slightly thicker skins. Perry's efforts could serve to >raise the thought level of other posters to this list, if they'd stop reacting >to each prod with an irritated swipe back at the prodder. First off, Perry Metzger had treated this list to such CYPHERPUNK (?) gems as: 1) his dripping sarcasm 2) posts about theories on government, anarchy, and nazis 3) riduling posts obviously mistakenly sent (one such post with several ^H in it) 4) mailing list vs. newsgroup and how he can handle 300 mails a day and get work done 5) posts on the wide variety of services banks offer in New York 6) posts on the former libertarian orientation of Rep. Dana Rohrbacher 7) other pointless garbage If Perry wanted to raise the thought level of other posters, he would quit posting such bilge. Ah, but then, he wouldn't be posting at all! If he would take his BULLSHIT to private email, this list would improve vastly. If I missed the cypherpunk content in the above examples, point it out. He has a rather inflated ego and is quick to insult others: >You don't know anything about hashing, then. So if Perry feels that "you can lead a horse to water but can't make him think" he could perhaps pull his head out of his ass, wipe the shit from his eyes and ears, buy a fucking clue at the nearest store, and shut up. ------------------------------------------------------------------------- To find out more about the anon service, send mail to help(a)anon.penet.fi. Due to the double-blind, any mail replies to this message will be anonymized, and an anonymous id will be allocated automatically. You have been warned. Please report any problems, inappropriate use etc. to admin(a)anon.penet.fi.

1 0

Re: Spread-spectrum net (vulnerability of)
by George A. Gleason 17 Dec '03

17 Dec '03

RE your item, "a sort of virtual-space map, which would reveal nothing about actual phyical location of the sites or the person you are contacting." I'm not so sure... in cellular systems, cells must know where the handsets are located in order to send incoming calls. Your transmitter has a physical location which could presumably be tracked in the normal manner, and I would expect the overall routing information in a net to be susceptible to traffic analysis in any case. An individual who is using the system to communicate wouldn't be able to find the physical address of another user, but e.g. an intelligence agency which was looking at the entire network would. Even assuming spread-spectrum and various link encryption techniques on top of whatever end-user encryption is supplied; with enough traffic and enough time, it should be possible to do TA. Or have I missed something....? -gg

3 2

Re: FLAME: breaking DES
by lefty＠apple.com 17 Dec '03

17 Dec '03

According to <an42035(a)anon.penet.fi> > >Well, if Perry were to have the BALLS to make impracticality arguments >against the experts who actually created the protocols and methods, >instead of taking cowardly pot-shots, maybe we wouldn't think he's >full of shit. Firstly, these are bold words coming from someone posting through an anonymous remailer. Secondly, what do you mean "we", kimosabe? I can only imagine that you're referring to yourself and your intestinal parasites. -- Lefty (lefty(a)apple.com) C:.M:.C:., D:.O:.D:.

1 0

Midwest Cypherpunk Meeting
by jdblair＠nextsrv.cas.muohio.EDU 17 Dec '03

17 Dec '03

Are there enough Cypherpunks in the Midwest to have a meeting, no matter how small? I've been following the list (rarely posting) for quite a while now, but I'd like to discuss these issues in person, and maybe feel a little less isolated here in Ohio. E-mail me back directly if there is any interest. John Blair jdblair(a)nextsrv.cas.muohio.edu

1 0

!FLAME: breaking DES
by holland＠CS.ColoState.EDU 17 Dec '03

17 Dec '03

OK, I sort of wanted to break up the constant flaming on this thread, so I thought I would introduce my own (admittedly ignorant) way to break DES. 1. Start with a full keysearch of all 2^56 possible DES keys. 2. If you have plaintext to work with, compare each sample decryption with the plaintext. 2a. If you don't have plaintext, I would try washing the decryptions through a pattern searching algorithm that would include a dictionary, patterns from compression programs like PKZip, machine language instructions, and any other possible form of communication that can be represented as a binary string. At least that's how I would do it. If anyone has any suggestions, corrections, etc. I wouldn't mind hearing them. Doug

1 0

Re: Spread Spectrum Nets (Way to do safely)
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

Robert Woodhead writes: > Succinctly, use the purloined letter technique -- hide it in plain sight. > There are RF bands set aside for unlicensed low-power operation; some > new phones use spread-spectrum techniques in these bands. It's more than just purloined-letter issues - it's *legal*. That means that the FCC won't be triangulating around to find and destroy your transmitters, which is good. So you're back to threats from the CIA/DEA/NSA/FBI, pigeons, vandals, etc., and encryption and spread-spectrum make it a lot easier to avoid problems from frustrated cops.

2 1

RSA crap
by Godzilla's Home Boy 17 Dec '03

17 Dec '03

Howzsa battles going. What are the chances of getting caught using pgp out of this country? I have friends in Russia, I'd like to use more than ordinary encryption getting to them.

1 0

DES: breaking, attacking
by Karl Lui Barrus 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- On the method Douglad Holland posted about breaking the DES... an interesting property of DES is that of complimentation. That is, if encryption of plaintext P under key K produces ciphertext T, then encryption of plaintext P' under key K' produces ciphertext T', where the primes (') mean bitwise complement. If: T = DES(P ,K) then T' = DES(P', K') Cryptanalysis can exploit this if two pairs are available such that T' = T or P' = P. What this boils down to is a 50% reduction in work. So now instead of taking 10 billion years it'll just take you 5 billion :-) Complexity wise, it's just 2^55 steps instead of 2^56, so it may not buy you too much. A more interesting attack I read of is a probabilistic attack against DES. Suppose you encrypt sensitive data, and change keys once a month. I can attack you by trying as many decryptions as possible within the month. Sure, I am not guarenteed success, but the point is that if I do this month after month, I have a decent chance of success at some point during the year. And if the information you encrypt is valuable, once success during the year is all I'll need. An upcoming book on cryptanalysis details this attack; the author calculates success rates given various estimates for hardware (speed and number) and frequency of key shifting. It is quite arresting: even if you change keys every week, if I can muster enough computer power to give me a 1% chance of success, the chances are "good" that at some time during the year I'll succeed in decrypting a message. For example, if I can be guarenteed of breaking your DES encrypted message in a year (say I can mount a brute force attack that does 2^56 encryptions and takes a year), there is an 8% chance the key will be recovered in one month. So even if you change your key every month, I have an 8% chance of success. If your key is needed to conduct financial transactions and you are a bank, I can profit greatly from one success. So basically, DES is looking more and more disadvantaged with its relatively small key size (64 bits, of which 8 are predictable parity bits). Karl Barrus <klbarrus(a)owlnet.rice.edu> -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLLzEJ4OA7OpLWtYzAQHE7AP7BpG0k0bOjLbTPJbiaMzEFSZuquo/te0w EeD4cC22jru8jXCEtM6AUHp94UJucAXDzY7szDhNlcMY5uv0OIXaFyoohPxBDEw5 CxKKUt0TxpqFdGwwkW8jpPMoBtZa8rGu/yPEZJ9CmcuHDANbEnVlQ9w8weMTWDHD qqCSlsuziOs= =hS3G -----END PGP SIGNATURE-----

1 0