December 2003 - cypherpunks-legacy

Re: (Fwd) Australian "ITAR" regulations
by cjh＠osa.com.au 17 Dec '03

17 Dec '03

Peter Trei writes: > Forwarded from the cypherpunks list... > http://www.austlii.edu.au/cgi-bin/sinodisp.pl/au/legis/cth/consol_reg/cer43… Thanks for posting this URL Peter. Of particular note is the heading: > 43. Other goods as follows: An anonymous opinion from inside the Defence Dept holds that electronic bits on a wire do not constitute goods, and as a result if you ship electronically, you are not subject to the regulations. If you ship a CD or floppy or other physical media containing software, you violate the regulations. I'll add this to the SSLeay legality FAQ - standard disclaimers apply of course, obtain professional legal advice before exporting SSLeay. ------------------------------------------------------------ Clifford Heath cjh(a)osa.com.au Open Software Associates Limited 29 Ringwood Street / P O Box 401 Phone +613 9871 1694 Ringwood VIC 3134 AUSTRALIA Fax +613 9871 1711 ------------------------------------------------------------ Deploy Applications across the Internet and Intranets! Visit our Web site at http://www.osa.com

3 2

Re: Snake-Oil FAQ
by steven ryan 17 Dec '03

17 Dec '03

At 05:27 AM 9/22/96 -0700, you wrote: >My view is that people interested in buying and using crypto are either >bright enough to learn, or are not. A "Snake Oil FAQ" is largely >unnecessary, for either category. For the first, because they're bright. >For the second, because they're not. My view is that there is a large third group of people who are bright enough to learn, but don't have the time or inclination to read books or do extensive research on the subject. There are a lot of people using PGP for the wrong reason, not because they read the books or did the research. Nor do they even understand how it works as opposed to how it is used. They are using it because they cruised the net and read good things about it or heard it was cool. A Snake Oil Faq could help prevent these people from choosing wrong products. It would also be very helpful to have all the arguments in one place in one concise faq. Before I joined this list and read Applied Cryptography I was in a discussion in a previous job about securing one of our products. The programmer wanted to protect the key with a convoluted series of transpositions. I knew it was dumb but couldn't successfully argue the point why. A faq would have been helpful. There a lot of people with a casual interest in crypto who will remember that there is a faq on bad crypto. When the time comes they may be able to use those arguments to help avoid the use of bad crypto. Steven ------------------------------------ Steven Ryan - Reading Access - sryan(a)reading.com PGP Fingerprint: E8 A2 C5 A2 7A C4 77 93 0A 1B 1D C6 B9 2F 36 9B Finger me for my PGP public key

1 0

provably hard PK cryptosystems
by Adam Back 17 Dec '03

17 Dec '03

extracting from part of another thread... Dimitri Vulis <dlv(a)bwalk.dm.com> writes: > Adam Back writes: > > [...] I think I remember that you posted some time ago a way to have > > two plausible decryptions for one ciphertext, to enable things like > > duress keys, in terms of RSA. The problem with this, however is > > that RSA is currently very slow to use in its pure form for > > messages. > > I don't recall this one... My feeling about RSA is that one of these > days there will be a breakthrough allowing much faster factorization > (either through a better algorithm on a conventional computer, or by > a practical quantum computer) and then all the codes based on > factorization will become essentially plaintext. It's time to start > looking for other hard problem to base PKC on. I'm not sure about quantum computers, some people who know much more about particle physics than I do seemed initially sceptical, and didn't think it was doable. However I have read some optimistic sounding news clippings (on the list) which sounded as if things are progressing well, with techniques being found using redundancy to get around what were earlier problems of reliability. Is this accurate reporting (thinking of garbled stories by over enthusiastic journalists)? I'd be interested to hear opinions from anyone who does know about particle physics about the likihood of practical quantum computers being practical in the next 20 years or so. RSA has always had problems with two pitfalls in any proof of its hardness: no proof that RSA is as hard as factoring, and no proof that factoring is hard. Rabin and variants of it at least have the proof of being as hard as factoring. You mentioned your opinion that a vastly better factoring algorithm may yet be found. What about the possibility of finding proof of a useful lower bound for the complexity of factoring, do you have any thoughts on the likelihood of this being achieved? That would be a very significant result, and in the absence of a quantum attack would be as good as it gets. It would allow you to make realistic calculations about physical limits of attacking it, in a similar manner to the entropy calculations possible for brute force of symmetric ciphers. (Heat death of the sun before a solution even if every particle was part of hardware key cracking computer type arguments). One other area that did sound promising was some kind of mapping problem in n dimensional tiling that Tim was discussing at a physical meet while I was over in the US. A news clipping posted to cypherpunks a short time ago was talking about a researcher who had constructed a crypto system related to the problem of finding paths in a tiling problem (sounded like the same problem). The interesting part was that the researcher was reported as having a proof of hardness. Similarly is this being reported accurately? > E.g., does anyone know of any progress made on public-key > cryptosystems based on the word problem in semigroups, described in > Neal Wagner and Marianne Magyarik, _A public key cryptosystem based > on the word problem_, Advances in Cryptology: Proceedings of Crypto > '84, G. R. Blakley and D. Chaum, eds., Lecture Notes in Computer > Sciences #196, Springer Verlag, 1985, and also mentioned in Wayne > Patterson, _Mathematical Cryptology for Computer Scientists and > Mathematicians_, Rowman and Littlefield, 1987? Not familiar with the problem. Is it something you could explain briefly, or are there any on-line papers on the subject? > >From what I neard, NSA tried very hard to implement it and failed, > and the Soviets actually built a cryptosystem similar to what they > described. I tried to duplicate what the Soviets supposedly did, but > without success. Adam -- #!/bin/perl -sp0777i<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<j]dsj $/=unpack('H*',$_);$_=`echo 16dio\U$k"SK$/SM$n\EsN0p[lN*1 lK[d2%Sa2/d0$^Ixp"|dc`;s/\W//g;$_=pack('H*',/((..)*)$/)

1 0

Re: Macintosh Mixmaster port... Who's doing it?
by Greg Broiles 17 Dec '03

17 Dec '03

At 03:16 PM 9/22/96 -0500, Adam Shostack wrote: > A real Mac port of Mixmaster, that integrated with Claris >EMailer and Eudora would be a huge boon to the millions of Mac users >out there. I have no doubt that Vinnie's mac crypto conference talked >a lot about this sort of thing. Actually, the Mac crypto conference didn't spend any time at all on Mixmaster - which is not intended as a criticism of either the conference or of Mixmaster, but it just didn't happen. I gave a very short talk and said that I thought the Mac needed three apps, for people who wanted to jump in a write something useful to the cause of privacy on the net and didn't want to reinvent any wheels: a remailer client with a good user interface, a Mac-native remailer, and an implementation of DC-nets. Mixmaster would, of course, address two of those three. Lucky tells me that there is already a Mac implementation of DC-nets, but it doesn't seem to be very well known. My impression of the demographics of the conference was that it was folks who are mostly working developers who aren't necessarily up-to-the minute on crypto and ecommerce stuff, but are interested enough to at least think about including it in their applications. I'm pottering around with a Java-based remailer that acts like a POP client so it can run on a client machine, not a Unix box; but other people should take that as a challenge to see if they can finish one before/better than me, not a reason to avoid writing one. Hal Finney has already done some very nice work with Java and mailing; see his home page (the address of which I don't have immediately at hand) for more details. For what it's worth, I think future remailer/Mixmaster development might do well in Java. I'm not especially sold on or trusting of the alleged security or trustability features of Java (sorry, no offense) but I *do* think it's a neat tool for building non-machine specific network aware applications. Ignore the fact that people use it to build silly animations or that downloadaded applets may or may not be secure - it's still useful as a development tool. -- Greg Broiles | "We pretend to be their friends, gbroiles(a)netbox.com | but they fuck with our heads." http://www.io.com/~gbroiles | |

4 3

Re: Internet File System?
by Scottauge＠aol.com 17 Dec '03

17 Dec '03

This is actually a good idea. There is something in the UNIX world called NFS (network file system). It is based on TCP/IP sockets. I do not know if you know anything about UNIX, but the file system used does not have logical drives (like a: or b: etc), just one huge tree that can span across many harddrives on many computers, thus making them all look like one. NFS helps to connect the other computer's file systems to the network tree. I think it is pretty cool... There is an RFC for it, though I do not remember the number. I read it once cuz I was planning a windows 95 version so I can hook my laptop up to my UNIX computers, but I found software to do it with and was like, why should I kill time on it when I could be using it to make money! It is very complete and designed to be implemented in C. Since Java is so close to C, a port could probably be made fairly easily (provided ya know something about writing file systems/operating systems ;) )

2 1

LD
by blanc 17 Dec '03

17 Dec '03

Dr. Dimitri Vulius, you made this remark to someone: ...You really should get to know L.D. better before you repeat the lies Timmy May spreads about him. ......................................................................................... May I recommend that you read the content within these urls: http://www.csn.net/~ldetweil/ and especially: ftp://crvax.sri.com/risks/15/risks-15.27 (and 28) These contain a history which will clarify for you some of the comments made about L.D.'s relationship to the list, Tim May, etc. .. Blanc

1 0

Transforming variable-length to fixed keys
by pgut001＠cs.auckland.ac.nz 17 Dec '03

17 Dec '03

I posted this to sci.crypt recently but the response to it was rather underwhelming, so I thought I'd repost it here to see if anyone has any comments on it. What it is is a scheme for transforming arbitrary user keys (typically a long passphrase) into a fixed-length key for a particular algorithm. This has the following properties: 1. The user key 'userKey' is transformed into an algorithm-specific key 'key' using a number of 'iterations' of a hash algorithm 'hash()'. 2. The transformation is strongly serialized so that any form of attack involving parallelization or precomputation isn't possible. 3. The transformation is non-reversible, so that recovering the transformed key won't recover the original key. 4. The result of the transformation is algorithm-dependant, so that if an attacker recovers a transformed key for one algorithm they can't recover the transformed key (from the same user key) for another algorithm. 5. The transformation can be iterated as often as required to make password-guessing attacks difficult. 6. The transformation process is algorithm-independant and can use any type of hash algorithm and original and transformed key size. The transformation algorithm (which was designed with the help of John Kelsey) is as follows: key[] = { 0 }; state = hash( algorithm, mode, parameters, userKey ); for count = 1 to iterations for length = 1 to keyLength (in hash_output_size blocks) state = hash( state ); key[ length ] = hash( state, userKey ); The state acts as an RNG which ensures that the key hashing is serialized. The initial state depends on all encryption parameters, not just the user key. If we hashed the user key directly and then used it for a number of algorithms then someone who could recover the transformed key for one algorithm could compromise it if used for other algorithms (for example recovering a DES key would also recover half an IDEA key). Hashing all algorithm-related parameters means that a successful attack one an algorithm, mode, or configuration won't allow the key for any other algorithm, mode, or configuration to be recovered. The code which implements the iterated hashing is: /* Hash the variable-length input to a fixed-length output */ memset( key, 0, keyLength ); for( count = 0; count < iterations; count++ ) { for( keyIndex = 0; keyIndex < keyLength; keyIndex += hashOutputSize ) { /* state = hash( state ); key[ n ] = hash( state, userKey ) */ hash( state, state, hashOutputSize, HASH_ALL ); hash( NULL, state, hashOutputSize, HASH_START ); hash( temp, userKey, userKeyLength, HASH_END ); | | | output input input size /* Copy as much of the hashed data as required to the output */ length = ( keyLength - keyIndex ) % hashOutputSize; for( i = 0; i < length; i++ ) key[ i ] ^= temp[ i ]; } } Peter.

2 1

RE: Where to write crypto?
by Will rodger 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- On Sat, 21 Sep 1996, Dan McGuirk wrote: >>| > If I want to go to a country outside the United States to write >>| > cryptographic code that will be freely distributable, what's the best >>| > place to go? >>| >>| Or, on the other hand, is there no way that a U.S. citizen can legally do >>| this? >> To which Adam Shostack replied: >> Thats not clear. The ITARs seem, on their face, to create a >>prior restraint on speech based on its content, and forbid Americans >>the right to leave the country to pursue their livlihoods. The odds >>of geting persecuted seem pretty low. Could be. But prominent crypto atty. Ken Bass told me he thought such actions would be vigorously prosecuted - quicker than you could say "global proliferation." Thing to notice here: no prominent cryptographer in the US has come forth and said "yes, I helped write strong crypto abroad." Which isn't to say most of them aren't traveling abroad frequently these days - take, e.g. the presence of M. Blaze, W. Diffie et al. at the OECD in Paris this week. >> As to the (predictable) comment that I'm not doing this, I'd >>be happy to entertain offers of crypto work that are not in the US, >>possibly leading to me being a test case. >> Call before you do, Adam; I'd kill to have that story first... Will Rodger Washington Bureau Chief Inter@ctive Week -----BEGIN PGP SIGNATURE----- Version: 2.9 iQEVAgUBMkXvQ0cByjT5n+LZAQGLyQf8Ctlf2JVeYI8Ws47YovJ0fhF3nzf9ihLd nHzZNqGpEVGVcOXYw2u84jV68y9Y7M3vdo/BDZCA3G0T8zZopmpiNZHgblh15ndP jPdYRkz4+7NnEn/Tz/1LQS6SVp/LA1G/qoCCNqGuYneSVkZxxsoeoEp9ZdCMhDE9 iq7IEI+pyY44vnSWvo65YNQZXt4thf94E94pspoNWC+DNwRDj8BkK28MpM3vduqD i/cViVUcKOLVg/b31HTVrq0rwWClfD0+J4a4F8AcGcf4SwedGCVaUAdSWRw0FQTJ 6AioCSd5Cs/g9k+xxaqRMzExJcBphwnlCZvA+qpgjY+KdagaEqJwGA== =vyTv -----END PGP SIGNATURE-----

1 0

Re: Bernstein hearing: The Press Release
by jim bell 17 Dec '03

17 Dec '03

At 05:15 PM 9/22/96 -0700, Rich Burroughs wrote: >At 04:39 PM 9/21/96 -0800, Jim Bell wrote: >>At 02:03 PM 9/18/96 -0700, John Gilmore wrote: > >[snip] >>>* Any legal framework that allows a government bureaucrat to >>>censor speech before it happens is an unconstitutional prior restraint. >>>The government is not allowed to set up such a drastic scheme >>>unless they can prove that publication of such information will >>>"surely result in direct, immediate, and irreparable damage to our >>>Nation or its people" and that the regulation at issue is necessary >>>to prevent this damage. >> >>At the risk of being a devil's advocate, let me suggest that you are >>conceding too much even with the preceding paragraph. The 1st amendment >>says nothing about preventing speech which (even admittedly) would result >>in "direct, immediate, and irreparable damage to our nation or its people." >[snip] > >Jim, that's a quote of Potter Stewart from the Pentagon Papers case, if I'm >not mistaken. It's not written in the Constitution, but Supreme Court >precedent is the next best thing as far as con law goes. Under the circumstances, I think it would be wise to actually EXPAND our protections. A lot has happened in the last 25 years. If a reasonably extensive list were made of situations and circumstances where the wool was pulled over the eyes of Americans by government...and how subsequent developments showed that these things were best revealed THEN and not later...it could be reasonably argued that there is no or almost no area where the "direct, immediate, and irreparable damage" exception is likely to realistically exist. You should point out that the Pentagon Papers case was decided notwithstanding an extraordinarily generous set of assumptions common in Cold-War America about the power and authority of government, many if not most of which would be no longer considered valid. Put the opposition on the spot: Insist that they provide a substantial number of examples where speech was expected, was attempted to be restricted by gov't under such an exception, where that speech actually occurred (either by refusal of the court to intervene or by refusal of a party to the case to restrict his speech), and show how this actually caused any "direct, immediate, and irreparable damage." In the absense of such examples, it is reasonable to presume that, generally, they don't exist or are so rare as to be ignorable. (For example, in the Progressive case from the middle 70's, the gov't dropped its case. What harm occurred?) >The government can restrict speech in time, place or manner, according to >the courts. Restricting content is more difficult, and places a much >heavier burden on the gov. > >To expect people writing about and arguing the case to completely ignore >the existing case law is foolish, IMHO. I am _not_ suggesting that case law be "ignored." Rather, point out that the case law, even though it goes in your favor, was still decided in a thoroughly biased atmosphere and therefore further reductions in government authority are appropriate. Jim Bell jimbell(a)pacifier.com

1 0

Paradox db passwords/encryption
by David K. Merriman 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- To: cypherpunks(a)toad.com Date: Mon Sep 23 12:52:32 1996 Sorry to disturb everyone, but I've got a questions that is (gasp!) immediately related to crypto: Does anyone have any info on the encryption used in the Paradox (4.5 for Win) database? I've got someone that has an encrypted (well, password-protected :-) Paradox database that needs some maintenance, and the person that knew the password is now - shall we say - 'no longer with them'. Right now, doesn't matter if it's something clever, or a brute-force hack; they just need to get into the tables and such. Dave Merriman - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- PGP Email welcome, encouraged, and PREFERRED. Visit my web site at http://www.shellback.com/p/merriman for my PGP key and fingerprint "What is the sound of one hand clapping in a forest with no one there to hear it?" I use Pronto Secure (tm) PGP-fluent Email software for Windows -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBMkYXlcVrTvyYOzAZAQGcxgQAleY6gXoM1n0AbYSUogW9Zo4brlHgsEHW vs3aAh+DQEaWJKc9ltXlDK94su9uJbAo3gb4cvG0EnBDifpqlS/bya7fG9KlcM6H k6REe2Ui6xBLtbjJTa5fmAjmeLYPSKnF5z6Eql8VpdZPoS0hqKZKgdyogMrai4Dx esOt90XBqVk= =hibj -----END PGP SIGNATURE-----

2 1

A daily word of caution in reference Tim C[unt] May
by Dustbin Freedom Remailer 17 Dec '03

17 Dec '03

Tim C[unt] May is not only as queer as a three dollar bill, but he is also into having sex with children.

2 1

Re: Snake-Oil FAQ
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 1:41 AM 9/23/96, steven ryan wrote: >At 05:27 AM 9/22/96 -0700, you wrote: >>My view is that people interested in buying and using crypto are either >>bright enough to learn, or are not. A "Snake Oil FAQ" is largely >>unnecessary, for either category. For the first, because they're bright. >>For the second, because they're not. > >My view is that there is a large third group of people who are bright >enough to learn, but don't have the time or inclination to read books or do >extensive research on the subject. There are a lot of people using PGP for >the wrong reason, not because they read the books or did the research. Nor >do they even understand how it works as opposed to how it is used. They are >using it because they cruised the net and read good things about it or >heard it was cool. Well, there are a bunch of books out on PGP, which they can read. And there are already some good FAQs out on the basics of cryptography--surely concise enough and yet detailed enough to warn folks away from some basically flawed programs. But just how far can one go? Some people just won't be taught, despite the several very-accessible books on PGP and crypto. So? And I don't really think there's a problem. Just how many of these "Snake Oil" crypto programs are people really _buying_? And does it matter if they buy a reasonably-competent program (*) like "DiskLock" instead of using 3DES or one of the good disk encryption programs? (* By "reasonably competent" I mean not "snake oil," and roughly able to do the job for which it was intended. Many people just want casual-grade crypto, to stop casual attempts to look at what they've written. We may disagree with them, but, hey, it's their choice. I maintain that these people are unlikely to read something called "The Snake Oil FAQ.") To coin a phrase, you can lead a person to strong crypto, but you can't make him drink. >A Snake Oil Faq could help prevent these people from choosing wrong >products. It would also be very helpful to have all the arguments in one >place in one concise faq. Before I joined this list and read Applied At some point this become YACB (Yet Another Crypto Book). If you and others want to donate time to help educate the (small, I think) class of users who won't read the PGP books, or the PGP articles in the magazines, and yet who you think are smart enough...blah blah...well, go ahead and write such a thing. (BTW, Schneier has a book out on "Security for the Macintosh," a kind of watered-down intro to crypto and security....he makes the points a "Snake Oil FAQ" might make...again, I think this is an overcrowded market.) >Cryptography I was in a discussion in a previous job about securing one of >our products. The programmer wanted to protect the key with a convoluted >series of transpositions. I knew it was dumb but couldn't successfully >argue the point why. A faq would have been helpful. Wouldn't arguments out of the standard textbooks have been just as effective, and perhaps even more "credentialled" than words from a FAQ? I hope you are not expecting that a FAQ would have the precise magic words dealing with your programmer friend's situation? At best, it would contain seom reworded arguments out of the well-known textbooks. I just don't see the point. But if it keeps folks busy, and happy, I guess it's harmless. (:-}) --Tim May We got computers, we're tapping phone lines, I 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: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

1 0

RE: Where to write crypto?
by Remo Pini 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit To: cypherpunks(a)toad.com Date: Mon Sep 23 08:31:11 1996 > > > On Sat, 21 Sep 1996, Dan McGuirk wrote: > > If I want to go to a country outside the United States to write > cryptographic code that will be freely distributable, what's the > best place to go? Switzerland has NO crypto laws (yet)... - ------< fate favors the prepared mind >------ Remo Pini rp(a)rpini.com PGP: http://www.rpini.com/crypto/crypto.html - ----< words are what reality is made of >---- -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: noconv iQEVAwUBMkYusBFhy5sz+bTpAQHziQf/eGRk90lC1/0NNMj9Qco54BHZhElwPCnK CsH8UJr90Ue68K5GA5VSBcWrBw/7j/AxaM5RiwnNzeuyw1GwYWlSKyfKzczqU6Z/ 2DuCGz/cRkDuUYjK5PrWbh59Pr6PH8flqnFEap9cWU1Fl8BjdCBX+6IYTjVh1Ph4 Z6M7kYB3rDlTYECB8CtZhAELq89+nD4Rk1SI/lIckGNOS/tSESBQ+5L/t+JDFmUd e6w07eRchN22+XDkVxBJ6UNPGR+3IiaNduOMXDyAgsHetXkfN51TaGOSpTlJltOz Wd7K0vjhGPDNhSgAa9dQ5J3B8ckHGsELHloXvjYfzfzW/tDOtbdjmg== =Whww -----END PGP SIGNATURE-----

2 1

Public Schools
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 4:31 AM 9/23/96, Lucky Green wrote: >On Sun, 22 Sep 1996, snow wrote: >> I would agree that parents can do as good or better at _most_ subjects >> thru about the 3rd or 4th grade, and I do agree that most of todays schools >> are shit, however there is one area--social skills--that homeschooling >> simply can't compete. Children need to learn how to interact with one another >> in groups larger than a family unit. I don't think that homeschooling can >> accomplish this nearly as well as the public (or private) schools could. > >I understand that many parents that homeschool belong to organizations >that provide for meetings twice a week in which the children so educated >in a certain area get together. Homeschooling does not have to stand in >the way of a normal socialization process. Also--and I mean this point completely seriously!--many parents are not altogether convinced that the "public school socialization" is all that beneficial. Do kids really _need_ to learn to wear gang colors, smoke to be cool, get pregnant at age 14, and so on? An awful lot of accomplished persons grew up in isolated areas, on small farms, and were educated in very small classrooms (or at home). I think the "social skills" Snow talks about above are actually the _worst_ part of public schools in fin-de-Siecle America. If I had a kid, I wouldn't want him or her in the local public schools. (And my area is more remote than over in San Jose--San Francisco, which is even worse. And LA is even worse.) --Tim May We got computers, we're tapping phone lines, I 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: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

5 5

Kicking the Furriners out of our Classrooms
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 6:19 AM 9/23/96, Matthew Gream wrote: >Hi there, > >> An anonymous opinion from inside the Defence Dept holds that electronic >> bits on a wire do not constitute goods, and as a result if you ship >> electronically, you are not subject to the regulations. If you ship a >> CD or floppy or other physical media containing software, you violate the >> regulations. > >Watch out for those anonymous opinions; I received exactly the opposite >opinion when I spoke to the Defence Signals Directorate about the issue >(back in 1994) -- after specifically asking about a few hypothetical >cases. Of course, either opinion may be correct, which is the real >problem! Cindy Cohn made the excellent point at the Bernstein hearing that the ITARS are so vague and overbroad that a professor cannot be sure if his lecture is violating the law because foreigners are in the audience (a la the Junger case). As she notes, nearly all college and postgraduate classes are heavily populated by non-U.S. citizens, and the ITARs specifically make illegal the propagation of certain items to non-U.S. citizens. (There is no exemption in the ITARs for university professors teaching their classes. If foreigners are in the classroom, and cryptographic or weapons-related knowledge is imparted, an ITAR violation has probably occurred.) Personally, I'd relish the opportunity to say to my class: "Now, the International Trafficking in Arms Regulations, the ITARs, make it a felony for me to disclose certain methods or techniques to non-U.S. citizens. Accordingly, in this class, I must insist that all non-U.S. citizens, or suspected Israelis, ragheads, Papists, or Marxists, illegal Mexicans, etc., leave the lecture hall immediately. All those remaining must present at least five forms of identification, dating back at least to 1978. Foreign-looking persons should provide at least seven forms of identification, including documentation that they are not in the U.S. illegally. Jew-looking or Jew-sounding names will be subjected to special scrutiny to ensure that they are not dual-citizens with the ITAR-restricted Zionist Entity. "Oh, and since classroom participation counts for 65% of your grade, those I exclude under the ITAR restrictions had better score 155% on the exams I let you into the classroom to take, else you'll flunk. Good luck, you fucking foreigners!" (Cindy Cohn pointed out to the court that their might also be 14th Amendment problems with enforcement of the ITARs with regard to teaching in universities.) --Tim May We got computers, we're tapping phone lines, I 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: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

1 0

Re: provably hard PK cryptosystems
by tcmay＠got.net 17 Dec '03

17 Dec '03

I want to say something about "tiling problems," an area I find very exciting. I ordinarily would not have commented on "quantum computers," but will make just a couple of comments, since I want to comment later on tilings. At 9:58 PM 9/22/96, Adam Back wrote: ... >I'm not sure about quantum computers, some people who know much more >about particle physics than I do seemed initially sceptical, and >didn't think it was doable. However I have read some optimistic >sounding news clippings (on the list) which sounded as if things are >progressing well, with techniques being found using redundancy to get >around what were earlier problems of reliability. Is this accurate >reporting (thinking of garbled stories by over enthusiastic >journalists)? I'd be interested to hear opinions from anyone who does >know about particle physics about the likihood of practical quantum >computers being practical in the next 20 years or so. Caveat: I'm a skeptic on quantum computers. I've read a couple of the early papers, but am not current. And I certainly am not an expert. For what it's worth: * I think it's nearly certain that no significant results will be gotten in the next 20 years, probably not in the next 50 years. (I personally am skeptical that significant results will be gotten in 200 years, and probably never, but this is a harder point to make.) * Sure, there may be demonstrations of a "collapsed quantum state" which is isomorphic to factoring a number like "42," but not 50-digit numbers anytime soon. And, I feel, a 300-digit number is probably many, many orders of magnitude harder. * Experts may point to partial successes, and maybe they're right. But I'm skeptical. The recent post by someone outlining a bunch of "promising approaches" has got me thinking of spending some more time looking at recent results, though. I doubt anything will happen in the next 20 years...that just isn't much time in the high-tech business (which may sound crazy, but it's true). (A way to "creatively visualize" this point is this: for those of you who are now 25, when you are 45 the world will probably _not_ be radically different. Sure, computers will be faster and cheaper, but most things will look _mostly_ the way they look today. Not a compelling argument, perhaps, but one which makes sense to me. When I was 25 I suppose I expected dramatic advances to come...largely, they haven't. I can elaborate on this if there's sufficient interest.) >One other area that did sound promising was some kind of mapping >problem in n dimensional tiling that Tim was discussing at a physical >meet while I was over in the US. A news clipping posted to One of the most interesting books I've read is David Harel's "Algorithmics: The Spirit of Computing." (Warning: The book came out in a second edition, which as near as I can tell took out some parts and synopsized the book a bit. My reading was of the First Edition.) Harel described Huang's work on "tiling problems." These are easier to describe with pictures, and I don't have the time to try to generate ASCII representations of tiles. So, I refer people to Harel's book. Briefly, imagine a grid in a plane. Imagine a set of "dominoes" or "tiles" with different edge properties, e.g., some edges are blue, some red, some green, etc. (or the edges can be numbered, have symbols on them, etc.). Suppose one has an unlimited supply of, for example, N different type of tiles. Suppose a tile is placed at some place on the grid, and another tile (possibly a different tile, possibly the same type of tile) is placed some distance away on the grid. The problem is this: Can a "domino snake" be found which reaches from the first tile to the second tile, with the constraint that edges must match up on all tiles? (And all tiles must be in normal grid locations, of course) If the grid area is, say, a finite space, then clearly an exhaustive search of all legal domino snakes would answer the question. (However, even a relatively small grid of, say, 8 x 8, generates a truly enormous number of possible snakes, each of which must be tested. Of course, if a domino snake is "guessed" (a la the "nondeterministic" language one encounters in complexity theory), verification that it is answer is nearly instantaneous (one glances at the solution and verifies it). If the grid is unlimited, then even if the two initial tiles are located close together, there are of course an infinite number of possible snakes.... (Oddly, to me, the domino snake problem is "more undecidable" in the infinite half-plane that in the full infinite plane....) Huang proved that the domino tiling problem is formally equivalent to the standard set of NP problems (consult Harel, or Huang, for more details and more precise language...I'm just going from recollection here!). I like this approach because I can easily visualize the domino tiling problem and can make some points about expansion of knowledge into uncharted areas (research and tool-building is a kind of domino snake expansion, with local laws of physics, etc., the constraints on stepping stones (hand wave inserted here)). I find this easier to understand than more "algebraic" and "logical" versions, such as the representation problem and the word problem. (These other NP problems are also described in Harel (and in standard complexity books, such as Garey and Johnson, Papadimitrou, etc.). It has always been a great hope that some of the provably-hard problems be adapted for use in crypto. And yet factoring remains the core of popular modern crypto programs (discrete logs, too, for D-H). I don't think this is a major cause for worry, though. It's not as if factoring is suddenly going to become "easy." (I think the consensus is that factoring _is_ hard. Whether RSA is really equivalent to factoring also hasn't been proved, as folks have noted, but many think it is, loosely speaking.) --Tim May We got computers, we're tapping phone lines, I 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: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

6 5

Re: Mercenaries
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 4:45 AM 9/23/96, Brian Davis wrote: >On Wed, 18 Sep 1996, Timothy C. May wrote: >> Brian Davis, our former Prosecutor, can tell us how likely it is that any >> person would be charged and brought to trial for being a paid mercenary for >> some small country in the Third World. The CIA is often behind such >> mercenaries, so national security issues could make the issue murky. > >Exceedingly unlikely. A variety of practical problems. If you came up >to a U.S. Attorney to show him your picture on the cover of The Zaire Daily >News as mercenary of the week and spit in his face, you'd get prosecuted. > >For spitting in the prosecutor's face. > >On a slightly more serious note: you'd only get prosecuted in someone at >Main Justice (i.e. in Washington) wanted you prosecuted. This of course matches my expectation. I still think if prosecution is so unlikely, the law ought to actually come to terms with this and remove the "official illegality" of such things. _Almost_ more important to me than "libertarian" ideals are "consistency" ideals: namely, that there oughtn't to be laws which are not enforced, or which are too expensive to enforce, or which can be selectively enforced. And since I know that the full suite of laws, all 25,000 or 45,000 of them (on all 27 linear feet of bookshelf space) cannot possibly be consistently enforced, I favor a "minimalist" or "fallback" position of having relatively few laws, covering mostly "crimes" which are more easily detected and prosecuted (with draconian punishments). >> But the real reason such prosecutions are rare is that the government >> realizes how Orwellian it sounds to say: >> >> "You are being prosecuted because you were a mercenary for Oceania in its I was probably overstating things to say this was "the real reason." The real reason (or the more important reason) is that resources are finite, and, as Brian noted, no prosecutor is much interested in someone being in Zaire's mercenary army. It doesn't register on the public's concern meter, and it probably doesn't get one promotions in the prosecutorial community. But the law should still be changed to reflect this reality. --Tim May We got computers, we're tapping phone lines, I 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: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

2 1

Re: Bernstein hearing: The Press Release
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 5:10 AM 9/23/96, Brian Davis wrote: ... >There isn't such a clause. The allowed restrictions were developed in >case law. > >Constitutional literalists take note: the First Amendment says nothing >about what the executive branch or the states can do .... Which is why President Jefferson was able to say: "While the Constitution says "Congress shall make no law...," I am hereby outlawing all discussion of the following subjects:.... And I have had it with the Catholics and Jews in this country. While Congress has no power to make laws respecting the establishment of religion or the exercise thereof, I am under no such restrictions. Therefore, I am ordering the immediate arrest and summary execution of all Papists and Jews...." Seriously, does not the Constitution and Bill of Rights define what states may do? And Presidents? The Executive is under various restrictions, and cannot behave unconstitutionally. After all, if Alabama, for example, reinstituted slavery, would not the 14th Amendment trump this? If California were to, say, ban speech critical of women's or homosexual's rights, would not the First Amendment trump this attempt? --Tim May We got computers, we're tapping phone lines, I 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: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

4 3

RE: Where to write crypto?
by Duncan Frissell 17 Dec '03

17 Dec '03

At 08:33 AM 9/23/96 +0200, Remo Pini wrote: >Switzerland has NO crypto laws (yet)... > However they are cooperating with the Gang of Seven (G7) Net restriction efforts. See: http://www.admin.ch/ejpd/d/bj/internet/indexe.htm DCF

2 1

Re: A daily warning regarding Tim ... [edited]
by DMiskell＠envirolink.org 17 Dec '03

17 Dec '03

>On Fri, 20 Sep 1996, John Anonymous MacDonald wrote: > >> Tim C[*] * studied yoga back-streching exercises for >> five years so he could * * (*). > >Likely the NSA has nothing better to do these days. They got bored with >just listening, so now they're spamming. Can we lock this fucker out of >the list? Sure, he'll probably pick another nym to post from... > Can't lock him out of the list, but you can either set up a kill file, or find something like the 3.0b version of Eudora, which has filters; you can set it up where, if an incoming message has a certain From: string, it is transferred directly into the Trash mail folder, which is, as a rule, cleaned up before quitting. Daniel. -- If in fact we are the only intelligent life on this planet, why the fuck are we in this goddamn mess? -- Find my public key on the World Wide Web -- point your browser at: http://bs.mit.edu:8001/pks-toplev.html

2 1

Snooping ISP admin??
by Michael E. Carboy 17 Dec '03

17 Dec '03

Greetings All, Question for the group: I have encountered a situation that causes me to believe an ISP is snoopingthrough encrytped mail. It seems that PGP'd mail has aroused the curiosity of an ISP (not hooked.net) I have encountered "POP3 account in use by another user" several times in the past few days and I am the only user... wondering if that "in use" messsage is the result of a clumsy sysadmin being caught with his hand in the cookie jar. Any thoughts from the group??? If those more knowledgeable than I deem these NOISE... my sincere apologies. regards, Michael E. Carboy carboy(a)hooked.net carboy(a)carboy.com

7 6

Re: (Fwd) Australian "ITAR" regulations
by pgut001＠cs.auckland.ac.nz 17 Dec '03

17 Dec '03

>>An anonymous opinion from inside the Defence Dept holds that electronic >>bits on a wire do not constitute goods, and as a result if you ship >>electronically, you are not subject to the regulations. If you ship a >>CD or floppy or other physical media containing software, you violate the >>regulations. > >Watch out for those anonymous opinions; I received exactly the opposite >opinion when I spoke to the Defence Signals Directorate about the issue >(back in 1994) -- after specifically asking about a few hypothetical >cases. Of course, either opinion may be correct, which is the real >problem! If the DSD is anything like our GCSB then what they'll have told you is what they'd like to be the case, not what's really the case. They will tell you what it pleases them to tell you, which doesn't necessarily have anything to do with the facts. The easiest way to check the real situation is to look at your customs act, the NZ Customs Act of 1995 (which in the relevant area is almost identical to the 1966 one) covers forms of export in excruciating detail. Doesn't mention anything about computer networks in there. In practice it'd have to be decided in the courts, but I don't think the DSD will take action because there's a very good chance they'd get a ruling against them, which is also why the NSA is so reluctant to enforce the ITAR in court. Peter.

1 0

RE: USA_exe
by Brian D Williams 17 Dec '03

17 Dec '03

John (jya(a)pipeline.com) posts: >The Washington Post reports at length today on the Defense >Department's disclosure yesterday of heretofore classified >training manuals used in the School of the Americas to instruct >Latino troops on killing a wide range of civilians for political >purposes. This is not particularly new news, I have a copy at home of one of these manuals, it was put out by a humans rights group that got their hands on it during all the El Salvador business. It's as nasty as it sounds... Further info on request. Brian "Semiautomatic's and Boundless Love."

1 0

Bank for International Settlement's Security of Digital CashReport now on the web...
by Robert Hettinga 17 Dec '03

17 Dec '03

--- begin forwarded text Sender: e$(a)thumper.vmeng.com Reply-To: Ian Grigg <iang(a)systemics.com> Precedence: Bulk Date: Mon, 23 Sep 1996 12:22:59 +0200 From: Ian Grigg <iang(a)systemics.com> To: Multiple recipients of <e$(a)thumper.vmeng.com> Subject: BIS_rep Thanks to all the help I got last week, I have located the BIS report on Smart Card Security. I have C&P'd and squeezed it into HTML format. There are still many errors in it, notably many double words that C&P contributed, and I haven't touched the Annexes (messy footnotes, no graphics or headings). However, for those desparate to get their hands on it, go to our page: http://www.systemics.com/docs/papers/ and type (anti-bot protocol format): BIS_smart_security.html at the end of the URL. The real (no errors :-) document is located at: http://www.bis.org/publ/cpss18.htm but only the forward, exec summary and intro are there at the moment. When they are all there, I will drop my version. I am not really that keen on repairing the errors in my version, as I gather that BIS will make the original available some day. But last I looked, they didn't have an email address, so who can say? -- iang iang(a)systemics.com --- end forwarded text ----------------- Robert Hettinga (rah(a)shipwright.com) e$, 44 Farquhar Street, Boston, MA 02131 USA "'Bart Bucks' are not legal tender." -- Punishment, 100 times on a chalkboard, for Bart Simpson The e$ Home Page: http://www.vmeng.com/rah/

1 0

Re: Public Schools
by Duncan Frissell 17 Dec '03

17 Dec '03

On Sun, 22 Sep 1996, snow wrote: >I would agree that parents can do as good or better at _most_ >subjects thru about the 3rd or 4th grade, and I do agree that >most of todays schools are shit, however there is one area-- >social skills--that homeschooling simply can't compete. >From John Holt's "Teach Your Own" "If there were no other reason to homeschool your children, protecting them from the 'valuable social atmosphere' of the schools would be sufficient." John was a commie liberal BTW. He felt that the schools had a very nasty and artificial social environment with rigid age segregation that bore no resemblance to real life where there are people of wider age ranges. Certainly, most people suffer worse mistreatment from their "peers" at school than they do later in life. As a libertarian, I would add that the social atmosphere of a Stalinist "brain factory" is not exactly the socialization I would choose for my children. I would choose a more market oriented model. DCF "Even if they manage to teach nothing else, the very existence of government schools teaches State supremacy."

7 10