December 2003 - cypherpunks-legacy

No Subject
by owner-cypherpunks＠toad.com 17 Dec '03

17 Dec '03

Well, it turns out that (according to a nice person at SurfWatch who put up with the rant I mailed in and send back a denial rather quickly) io.com isn't actually blocked. At least, so they say. No comment on whether there's anything like the "25 complaints" policy. I don't own SurfWatch, but if anybody does and finds www.io.com blocked, I'll send in another rant. I wonder whether they've actually considered the liability situation in re: blocking sites that shouldn't be blocked? I mean, sure, they seem nice enough about setting things right (like with the Nynex sites whose url's had "xxx" in the paths), but it seems to this non-lawyer that a case could be made for damages inflicted by being known as a purveyor of filthy indecency for even a short while. ______c_____________________________________________________________________ Mike M Nally * Tiv^H^H^H IBM * Austin TX * pain is inevitable m5(a)tivoli.com * m101(a)io.com * <URL:http://www.io.com/~m101> * suffering is optional

1 0

crypto ++ where?
by Oscar Boykin 17 Dec '03

17 Dec '03

How can I obtain crypto++, I seem to have deleted the email I had reguarding it. any info would be appreciated. thanks. oscar boykin mailto:boykin@pobox.com http://pobox.com/~boykin home: 404-206-0477

2 1

Leahy bill nightmare scenario?
by jim bell 17 Dec '03

17 Dec '03

In case any of you people still think that Leahy bill ostensibly freeing up encryption is "progress," the following scenario is provided for your consideration: "Bob" runs an encrypted remailer. His system forwards mail whose contents he cannot read, even if he wanted to. He cannot know from where the messages originated, or where they ended up. He likes it this way, because nobody can accuse him of complicity with a (encrypted) message that he can't read. One day, Leahy's bill passes, as described by VTW (and quoted by Peterson): "§2804. Unlawful use of encryption to obstruct justice" "Whoever willfully endeavors by means of encryption to obstruct, impede, or prevent the communication of information in furtherance to a felony which may be prosecuted in a court of the United States, to an investigative or law enforcement officer shall-..." I am assuming they fix the obvious error in the phrasing above. "Bob," the operator of the encrypted remailer, receives an email one day which states something like: "Thanks for the use of your nifty anonymous remailer. Under a different name, I intend to use this remailer (along with others) to transmit child pornography, plot terrorism, and do all of my drug deals. You've made my life so much more secure!" Bob, alarmed at this note, tries to cover his ass by sending back a message asking this person to not do anything illegal on his machine, hoping that this will protect himself. The response is "as long as the system operates, it will be used for whatever I want!" What "Bob" doesn't realize is that the message came from an agent for the cops, who now have proof that he is aware that his system will be regularly used for illegal purposes. If "Bob" is smart enough, he will realize his quandary, and he has only two choices: 1. Shut the remailer down to prevent such use. 2. Continue to run the remailer, knowing that it is being used for illegalities. If he should choose the latter, the cops merely have their agent mail some kid some child pornography, and use Bob's remailer as the last link in the chain. At that point, the "investigation" starts. The cops approach "Bob," and insist that he tell them from where the message came. Naturally, however, "Bob" is an honest fellow, and he runs a remailer that doesn't keep records. At that point, Bob is GUILTY of violation of the Leahy bill, because his encrypted anonymous remailer: 1. Uses encryption to thwart message tracing, and thus the "criminal investigation." 2. Bob has already been informed that his system will be used for illegal purposes; the cops have the messages to prove he has been told. He's GUILTY GUILTY GUILTY, he will definitely lose the system and possibly whatever residence it runs in, and will probably have to pay a huge fine to boot. Now, you may not sympathize with Bob. This doesn't affect YOU, right? RIGHT?!? But let's suppose the cops offer him a DEAL! "Spy for us, keep records and forward each and every one of them to us, and we won't prosecute you!" Such a deal! At that point, even an idiot begins to see the problem: Suddenly, you can no longer trust ANY anonymous remailer, because the operator might have been "stung" already, and he's keeping his system up only to keep his house and life savings. One by one, each encrypted anonymous remailer is dealt the same treatment, and pretty soon you can't trust any of them. All the systems run by honest, uncoerced people will go down. Naturally, this treatment will occur in every country that sites anonymous remailers. So maybe the word gets out, occasionally. At that point, usage of anonymous remailers declines, and people willing to risk operating one declines. A few come up which are run by the Feds, which log anyone who attempts to use it... Or am I the only person who can see this, huh? If anybody doubts this scenario, I challenge you to tell me WHY it cannot happen. I am, frankly, astonished at anybody who did not immediately see the potential downside to this portion of the bill! Jim Bell, Pessimist and proud of it. Klaatu Burada Nikto

2 1

Fractals, Cellular Automata, and Encryption
by tcmay＠got.net 17 Dec '03

17 Dec '03

Note: I have changed the thread title from the meaningless "Signature" (meaningless to this context) to something I think is more appropriate. Someone recently wrote to me asking why I so often change thread names, as, in his words, "it screws up the threading in my reader." Well, I think accurately labelled articles are more important that having a reader place an article in its "correct" position when the themes have changed so much. I urge all of you to think about what your article says and what the most accurate name is for it. By all means leave the name the same if you want, or if the Hamming distance is not too great. But if the topic has changed away from the name given by default, then _change_ the name to reflect the topic at hand. On to the actual article: At 11:54 PM 3/6/96, Mark M. wrote: >On Tue, 5 Mar 1996, Charles Choi (SAR) wrote: > >> 1) Is it possible to base a privacy key ( e.g. PGP ) on a fractal >> equation, instead of an algorithm based on two primes? >> This would allow for an eternal level of complexity due >> to infinite field of depth one can find as one 'zooms in' >> closer ( correct me because I'm wrong; I'm not a math major, >> although increasingly I wish I was... ), allowing for near >> unbreakable privacy of information. > >The fact that the private key is based on fractals rather than prime numbers >really doesn't make a difference. Fractals are not random, and do in fact, >have a pattern. The Mandelbrot Set, for instance, can be expressed in a >few bytes of information even though it is infinitely complex. Therefore, >the fractal has extremely low entropy making it a bad choice from which to >obtain random data. Besides these points, something missing from cellular automata-based crypto schemes is this: invertibility with a different key than was used to encrypt. That is, in any fractal or cellular automata-based schemes I have seen, a generator iterates a data set, transforming it into something which appears to have "no resemblance" to the original data set. The problem is that there is no second key, the decryption key (or "private" key) which reverses the process and recovers the original data set. That is, it is certainly possible to get some "messy" output by running a cellular automata (think: "the Game of Life" as an example) on an input. An attacker would be hard-pressed to determine the starting pattern if given the nth generation! So far, so good. The problem is that the _recipient_ would also have a hard time determining the starting pattern! And a more detailed wrinkle is this: at best, the system would be a single-key or symmetric system, losing the advantages of a public key system. At worst, the scrambled message could _never_ be recovered, as no inverse can be found of the CA. (In CA research, finding a starting pattern from some nth generation is known as the "Garden of Eden" problem, for reasons I won't get into here. Clearly, some CAs have no single inverse, as multiple inputs map into the same output--again, think of "Life.") Steven Wolfram had some speculations about using fractal or cellular automata-based systems for a new kind of cipher. His paper is in one of his books ("Theory and Application of Cellular Automata"), but it doesn't really get beyond just speculating. And, I recall that someone proved several years ago that Wolfram's CA-based encryption scheme was formally equivalent to a linear congruential generator. I think I included a few paragraphs on this topic in my Cyphernomicon. --Tim May Boycott "Big Brother Inside" software! We got computers, we're tapping phone lines, we know that that ain't allowed. ---------:---------:---------:---------:---------:---------:---------:---- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Higher Power: 2^756839 - 1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

2 1

surfwatch filtering
by mutant＠compuville.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- someone wrote: >[homepage on IO] > >BTW, did you hear that IO got on the SurfWatch "bad list" because there >were more than 25 complaints about objectionable material in the io.com >domain. (Apparently, when SurfWatch hears more than 25 complaints >against a particular domain, the just deny access to the whole domain >rather than particular URLs.) A quick phone call to the Surf Watch people revealed that they rarely block whole domains, and that they are not currently blocking all of io.com. - -- mutant(a)compuville.com "They that can give up essential liberty to mutant(a)cypher.net obtain a little temporary safety deserve (send mail with Subject: neither liberty nor safety." send-pgp-key for PGP key) -Ben Franklin ~1784 -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBMT9NaH0ndmRPxibdAQFSJwQAtq/s1c+GZlHvgVOhNgM8Dseq0SQ+6L8U bx8FV0l85anNk+fjRnY5hD8RflHFeVzLis2yDmn8BLjYzVnHBEWBXXncpwg3okoQ FOMJJTZZbOec10ESy3pvRPwgie60IaR3qIcHUmUHRqbv5dOjgDg5GhYWeZ8rhCCN 0pisfaQBsVc= =qgSw -----END PGP SIGNATURE-----

2 1

http://www.grfn.org/~blacktar/KilltheNewbies.htm
by uunet!in2.uu.net!newsfeed.internetmci.com!gordius!news.service.uci.edu!usenet＠warwick.com 17 Dec '03

17 Dec '03

http://www.grfn.org/~blacktar/KilltheNewbies.htm

2 1

Re: Looking for code to run an encrypted mailing list
by Bill Stewart 17 Dec '03

17 Dec '03

>John Pettitt <jpp(a)software.net> wrote: >> Poster cryptes mail with pgp using list exploder key. List exploder decrypts >> mail and recrypts with keys for all current list members and then sends the >> mail. At 09:24 PM 3/4/96 -0800, abostick(a)netcom.com (Alan Bostick) suggested having the mailing list handler decrypt the session key using its private key, and re-encrypt with the public keys of the list members, but not do the IDEA decryption and re-encryption N times of the message body. This has the further advantage that the cleartext message body is never sitting around on the server where it might end up in swap space or file system leftover blocks. On the other hand, it really only costs you one IDEA encryption if you want to use the multiple-recipients options to PGP. With the current PGP, this means you don't have to hack your own crypto code; the toolkits in PGP 3.0 will make that easier, though. >> [I don't want all the list members to need to know every other list members >> public key] The multiple-recipients PGP doesn't give away everyone's public keys, only their keyIDs. If you want to remain pseudonymous, just create a public key that you use only for subscribing to the list. (Hmmm - multiple recipient support probably makes stealth-PGP harder to implement...)

3 2

Re: Bootable disks
by JonWienke＠aol.com 17 Dec '03

17 Dec '03

I recently installed an IDE version of the Syquest EZ 135 hard drive for a client. Since it is a fully functional hard drive, it is bootable, unlike the parallel port model. The biggest advantage to this type of drive is that the cartridge (and any potentially incriminating data) can be quickly removed and destroyed (burn it, smash it with a large heavy object, insert it into a bulk eraser) or taken to a secure location for storage. Also, this type of drive provides an excellent place to put swap files, which have an annoying tendency to pick up sensitive data. That is less of a problem if you can remove the cartridge and store it in a secure location when not processing sensitive data. Jonathan Wienke

2 1

Re: Jump Start ecash With IPhone
by mccoy＠communities.com 17 Dec '03

17 Dec '03

eggplant writes: >How about...only paying for the initial investment of the phone program and >hardware. No more costs to pay, get rid of long distance charges >all-togther. Two problems: 1- Chaum did not have the clue to pick up on this when two cypherpunks (who shall remain nameless...) pitched this and several other ideas to involve ISPs in jump-starting ecash almost two years ago, so why would he figure it out now? 2- The phone companies (actually ACTA, the telco lobby) petitioned the FCC on Monday to regulate the Internet phone software companies. IP phone systems themselves may need to move underground just to escape regulation. OTOH, there is nothing to prevent people from building their own systems once some free software is out there... On another related note the IRS is not getting someone what concerned about the opportunities the Internet offers people to avoid taxes according to the Treasury department's international-tax counsel. The government will "maintain toll booths on the information superhighway." (WSJ, 3/6/96) jim

2 1

What's anyone know about Teledyne Electronic Technologies?
by Mark Aldrich 17 Dec '03

17 Dec '03

Teledyne (TET) is out marketing a "new" crypto system as a competitor against DES and what TET calls "linear" cryptosystems. They are claiming a non-linear approach in which (as best I can tell) they are permuting the s-boxes and using 4 bit/16 entry substitution tables. They are also asserting that these "key generated substitution tables and inter-round permutations" are supported by "nonlinear orthomorphic mappings generated from arbitrary key" (and that this is patented) and "row-complete Latin Squares generated from an arbitrary key" (and that this is also patented). The claimed benefits include being invulnerable to differential and "linear" cryptanalysis, no linear key/data interaction to hide, that this results in the tables and permutations being "transient/secret," and that fewer rounds are needed (fewer than what, I don't know) to attain "resistance to cryptanalysis." If there's any c'punks who'd care to comment on these notions, I'd very much appreciate the feedback. Also, if anyone's actually seen the algorithm, reviewed it, or knows of published material about these techniques, I'd also appreciate seeing those posted. Of course, any feedback from existing customers of TET would also be great. Thanks! ------------------------------------------------------------------------- |So, I went walking through the street. |Mark Aldrich | |I saw you strung up in a tree. |GRCI INFOSEC Engineering | |A woman knelt there, said to me, |maldrich(a)grci.com | |Hold your tongue, man, hold your tongue. |MAldrich(a)dockmaster.ncsc.mil | |_______________________________________________________________________| |The author is PGP Empowered. Public key at: finger maldrich(a)grci.com | | The opinions expressed herein are strictly those of the author | | and my employer gets no credit for them whatsoever. | -------------------------------------------------------------------------

2 1