December 2003 - cypherpunks-legacy

University logging mail to anon.penet
by Jeff Simmons 17 Dec '03

17 Dec '03

This just came up locally, and I'd like to have some comments on it, especially from people who understand the law a lot better than I do: Our local University apparently has been logging ALL mail to anon.penet, including faculty, students, and off-campus users. They maintain such weak security that someone was able to "obtain" the logs and post them to a local usenet group, thus compromising everyone's "anonymous" identities. -- Jeff Simmons jsimmons(a)goblin.punk.net

6 7

IDEAL FOR BEGINNERS WORK-AT-HOME
by rewards＠t-1net.com 17 Dec '03

17 Dec '03

/////////////////////////////////////////////////////////////////////////////// This Message was Composed using Extractor Pro Bulk E- Mail Software. If you wish to be removed from this advertiser's future mailings, please reply with the subject "Remove" and this software will automatically block you from their future mailings. //////////////////////////////////////////////////////////////////////////////// I have been a business owner for over 30 years, involved in various things, and I have found different strategies for making money, working at home.... I thought you might like to know, this really works, if you will only try it... For more information send e-mail to max(a)answerme.com

1 0

Bernstein decision: image files upgraded
by John Gilmore 17 Dec '03

17 Dec '03

If you couldn't read the decision because the original image files were too fuzzy, try them now. They're wider and longer, and a lot easier to read. They're almost as small as the previous ones, because we encoded them better. We're still working on scanning it in to plain text. http://www.eff.org/pub/Legal/Cases/Bernstein_v_DoS/Legal/970825_decision.im… John Gilmore

1 0

Re: What is truth?
by James A. Donald 17 Dec '03

17 Dec '03

Crypto relevance: Absolutely none: At 10:57 PM 9/6/95 -0700, Timothy C. May wrote: > If you want more information on my views about truth, check out > the work on "evolutionary epistemology," especially the writings > of William Bartley and Karl Popper. Popper went off the deep end because he tried to justify science without relying on the principle of induction, or its equivalent, Bayesian probability. A hopeless endeavor, like the attempt of the behaviorists to describe behavior without reference to intention, desire, and knowledge. His reasoning leads logically to the polylogism of the fascists, an outcome he imagines he avoided, but in fact he merely rephrased in language that superficially sounds more favorable to science. Bayesian probability leads us to the conclusion that some scientific theories have a probability of truth that is exponentially close to unity. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.jim.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the state. | jamesd(a)echeque.com

1 0

High Society Magazine Clicks
by email＠exoticwhispers.com 17 Dec '03

17 Dec '03

Dear Fellow Webmasters, I would like to take this opportunity to introduce you to the highest paying and most diversified Click Through Program on the web for Adult Webmasters, simply named "Serge Cash". http://sexia.com/~knight/RAW/TIM/ This is the real deal. No more NO NAME site Click Throughs Programs with low visitor to click ratios by fly by night companies. Get signed up with such International Adult Magazines such as Playgirl, High Society, Climax ..etc. Compare for yourself ! Here is a list of our current programs and their payouts: ** You choose how you want to be paid! RAW Clicks or UNIQUE Clicks ? ** Available Programs Payout Per Raw Click Payout Per Unique Click ------------------------------------------------------------------------------------------------------------------------- PINK BITS $.02875 (2.875 Cents) $.055 (5.5 Cents) PLAYGIRL MAGAZINE $.0325 $.055 HIGH SOCIETY MAGAZINE $.02875 $.055 CLIMAX MAGAZINE $.02875 $.055 AMATEUR AVENUE $.02875 $.055 XXSEX $.02875 $.055 ASS $.02875 $.055 SEXY CLIPS $.02875 $.055 We also pay our advertisers every two weeks and our minimum check amount is only 10.00. Give us a try for two weeks and see for yourself. Make sure our check arrives, and see how much more it is compared to your current sponsors. Complete information and Online Signup to get you started right away at: http://sexia.com/~knight/RAW/TIM/ Tim Goyetche Representing Serge's Cash

1 0

Le Monde: "Militant Cypherpunks liberate crypto software"
by Lucky Green 17 Dec '03

17 Dec '03

I am not making this up. Le Monde, a major French newspaper, discovered that "militant" Cypherpunks aid in the spread of crypto :-) In all fairness, the article is rather favorable. [Available in French only]. http://www.lemonde.fr/multimedia/sem3397/textes/act33971.html --Lucky Green <shamrock(a)netcom.com> PGP encrypted mail preferred. DES is dead! Please join in breaking RC5-56. http://rc5.distributed.net/

2 1

Re: Growth of actions definded as crime. Which math formula?
by James A. Donald 17 Dec '03

17 Dec '03

At 12:29 PM 9/6/95 -0700, Timothy C. May wrote: > I've seen figures on the "linear feet" of regulations, and how they are > growing exponentially, but I don't recall the numbers. Something like the > total number of laws doubling every 10 years or so, but don't quote me on > this one. In addition one should also consider that most of these new crimes are deliberately written to be sweeping and vague so that they can be enforced in a selective and capricious manner: For example wetlands are deliberately defined in a way to give them considerable elbow room, indeed so much elbow room that much of death valley technically qualifies as wetlands. The wetlands law was written to make it easy for bureaucrats to win lawsuits, not to reflect common sense or basic sanity. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.jim.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the state. | jamesd(a)echeque.com

1 0

Cryptography Technical Report Server (CTRS) needs submissions!
by David J. Bianco 17 Dec '03

17 Dec '03

Earlier this week I posted a note bemoaning the lack of a good searchable repository for cryptography related technical reports. In short, I volunteered to develop, host and maintain such a system as a service to the rest of the Internet. The system is ready, now the fun part begins... The Cryptographic Technical Report Service needs *YOU*. Specifically, it needs your technical reports (or other technical documentation). In order for CTRS to become popular, it needs to have a useful amount of data to search, so I'm currently soliciting submissions for the database. Contributing to CTRS is pretty simple. Detailed information can be found at <http://porthos.itribe.net/CTRS/FAQ.html#contribute>, but basically all that's required is to send <ctrs-submissions(a)itribe.net> a refer-format bibliographic entry which contains a URL pointing to the paper's real Net location. I'll index the entries (prettyprinting them, of course), and CTRS users will follow that URL if they want to retrieve the paper. If you don't have a URL, I'm willing to host as many papers as I have resources for. Full details can be found at the URL above, but please note that I'm only able to accept papers submitted by the authors (or copyright holder). If you'd like to know more about CTRS, you can check out it's beta version at <http://www.itribe.net/CTRS/> (<https://www.itribe.net/CTRS/> for the SSL version). The database doesn't have much (anything) in it right now, so I'm relying totally on submissions in order to build the collection. I firmly believe that CTRS can provide a useful service to the cryptographic community, so I hope you will all consider contributing your reports to CTRS. If you have questions, please check out the CTRS FAQ <http://www.itribe.net/CTRS/FAQ.html>. If you still have questions, feel free to contact me at the email address below. ========================================================================== David J. Bianco | Web Wonders, Online Oddities, Cool Stuff iTribe, Inc. | Suite 1700, World Trade Center | email: <bianco(a)itribe.net> Norfolk, VA 23510 | URL : http://www.itribe.net/~bianco/

1 0

PGPlib-0.3 is available
by Robert Hettinga 17 Dec '03

17 Dec '03

--- begin forwarded text To: pgplib-announce(a)cryp.to Subject: PGPlib-0.3 is available Date: Wed, 27 Aug 1997 11:10:03 +0200 From: Tage Stabell-Kulo <tage(a)ACM.org> Reply-To: tage(a)acm.org Sender: pgplib-announce-owner(a)cryp.to Precedence: list -----BEGIN PGP SIGNED MESSAGE----- This mail is sent on the list pgplib-announce(a)cryp.to. It is a (very) low voulme list where announcments are made when new versions of PGPlib becomes available. Please find below a copy of the file README distributed with this new release. Instructions on how to obtain your own copy of PGPlib is included. === Welcome to version 0.3 of PGPlib. This code has not yet matured enough to reach version 1.0. However, this is probably the latest development version before 1.0. Detailed instructions for how to obtain this fine code can be found at the end. Experienced users will know that ftp://dslab1.cs.uit.no/pub/PGPlib.tar.gz is the place to look. What is here for you ===================== PGPlib is a library that lets you generate (and manipulate) PGP packets without having to run PGP. In particular there is code to generate and understand the following types of PGP packets: - Data can be signed with a private key; - Data can be encrypted with a public key; - Data encrypted with your public key can be decrypted; - You can verify signatures on public keys and on buffers (files); - Convential encrypted (IDEA with Zimmermann's context sensitive feedback). The library can both read (decrypt) and write (encrypt) convential packets (in PGP format); - Armor. You can (de)armor a buffer or a file into a buffer or a file; - UserID packets are read and written in a variety of formats; - Literal with filename, mode, etc. You can create literate packages from files, or from buffers, and create files from literate packets; - Keys can be obtained from a database (which is provided) or by parsing keyrings. Keys can be kept in buffers or on files; - You can maintain a PGP public-key database (I use this library to maintain a database with ~40.000 keys). There is code to use DBM as supplied from Berkeley. A copy of DBM is included for your convinience. In general, PGPlib operates on buffers in order for you to use it in your applications. You will find included a small program that will (de)armor anything, a parser to parse PGP files (including decryption and so on), a shell to manipulate a keydatabase, a keyserver to run on top of such a database, a program to verify signatures on keys and/or files, a program to split keyrings in smaller parts and a program that will sign files for you. You will find all these (and more) in the PGPlib/applications/ directory. None of these uses PGP, the library provides all the functionality you need. What is not there for you ========================= - Sufficient documentation. I have written this library because I need it. In order to make life less hard for my students, I will have to provide some documentation. Possibly even on-line. However, quite a few manual pages are included. What you need ============= - Unix (for some definition of "Unix"). - You must have the SSLeay library as I have not implemented any cryptographic functionality; I hope SSLeay is a good choice. I did not major in mathematics and can thus not judge the quality of their work, although it looks solid (to me). I link with their version 0.8.1. I rely on SSLeay for many things, in particular their BIGNUMs, RSA encryption and IDEA. You can obtain your copy of SSLeay from: * ftp://ftp.psy.uq.oz.au/pub/Crypto/SSL/ - SSLeay source * ftp://ftp.psy.uq.oz.au/pub/Crypto/SSLapps/ - SSL applications * http://www.psy.uq.oz.au/~ftp/Crypto/ssl.html - SSLeay Programmer Reference SSLeay is quite large and I only use a fraction of it. other hand, SSLeay seems to be well maintained and their crypto-library might very well shrink or become more modular as time passes. Also, work is in progress to strip out from SSLeay only thise parts that PGPlib uses. Porting ======= PGPlib has been developed on NetBSD 1.2. It should compile smoothly on FreeBSD, HP-UX and Linux. If you port PGPlib to some other platform, I would be greatful if you would send me patches. In particular, PGPlib does not compile on SunOS 4 and 5, your assistance would be valuable. Regarding PGP ============= - Work is underway to ensure that PGPlib is compliant with RFC1991. - The name PGPlib might not be final as I respect Zimmermann's "ownership" of the acronym PGP and he might not like that this library uses the name PGPlib. Thus, this library might become libRFC1991.a or something. - PGP-5.0 is available. It has several new features, such as a multitude of encryption methods. PGPlib might evolve to support them all. Your Feedback ============ Your feedback is solicited. Peter Simons <Peter.Simons(a)gmd.de> has created a mailinglist for PGPlib. To subscribe, send an e-mail to the address pgplib-dev-request(a)crypt.to and write the command SUBSCRIBE into the BODY. If you want to be subscribed under a different address than the one you're mailing from, you can also use SUBSCRIBE yourname(a)somewhere.else to do the trick. To post to the list, send an e-mail to pgplib-dev(a)cryp.to as usual. There is also (very low volume) list pgplib-announce(a)cryp.to. Subscribe by sending an e-mail to pgplib-announce-request(a)cryp.to with SUBSCRIBE in the body. Both lists are archived at URL:http://www.cryp.to/ If you write a nice application based on this library (the ultimate feedback :-), please feel free to send it to me and I will include it in the next release; Where to get PGPlib =================== ftp://dslab1.cs.uit.no/pub/PGPlib.tar.gz This file is a link to the latest version. If you obtain the file PGPlib-0.3.tar.gz instead, note that a name conflict will occur when you unpack the file (see below). What to do ========== First, uncompress PGPlib, which should give you the file PGPlib.tar. When you unpack PGPlib.tar you will get three files in your CURRENT directory. These files are README : This file PGPlib-0.3.tar : The library PGPlib-0.3.tar.sig : My signature on the above. Verify the signature by running pgp PGPlib-0.3.tar.sig and verify that PGP prints Good signature from user "Tage Stabell-Kulo <tage(a)cs.uit.no>". When you unpack PGPlib-0.3.tar, tar will create the directory PGPlib-0.3 and place all the files within it. Unless you run SunOS, running the script configure should produce Makefiles that you can use to compile everything. To verify that everything compiled correctly, you should run the script applications/test.sh. Before you do so, you must edit away the code that asks you to read this file. COPYRIGHT (yum, yum) ========= The library and included applications are all available under "Berkeley style" copyright terms. Basically, this means that PGPlib is FREE for commercial and non-comercial use, and that you can do almost anything with the code. The only thing you can not do is to say that you wrote it. See the file COPYING for details. //// Tage Stabell-Kuloe | e-mail: tage(a)ACM.org //// /// Department of Computer Science/IMR | Phone : +47-776-44032 /// // 9037 University of Tromsoe, Norway | Fax : +47-776-44580 // / "'oe' is '\o' in TeX" | URL:http://www.cs.uit.no/~tage/ -----BEGIN PGP SIGNATURE----- Version: 2.6.3ia Charset: latin1 Comment: Processed by Mailcrypt 3.2, an Emacs/PGP interface iQCVAwUBNAPu4wMzZ6tx+9RpAQEM2AP/Sp7REQHWSRGhscblUpX7CPoin6h9Lo5Y D0CzO/7cm9kqXqB3oQ3xPhchEcXzJEyABz8mehoAm6+D3Bqus1aI1JkcD6mVmoUE yQ3QU0X0jXdN84xcjTn6cyQx3iuu17MQ5/WKoLh2Ftq5DzBGATGlmGN6O6TRUbAW SZrCOQDN54M= =XjX4 -----END PGP SIGNATURE----- ------------------------------------------------------------------------------ PGP Library Announcements --- end forwarded text ----------------- Robert Hettinga (rah(a)shipwright.com) Philodox e$, 44 Farquhar Street, Boston, MA 02131 USA "... however it may deserve respect for its usefulness and antiquity, [predicting the end of the world] has not been found agreeable to experience." -- Edward Gibbon, 'Decline and Fall of the Roman Empire' The e$ Home Page: http://www.shipwright.com/

1 0

Re: fast modular reduction (proof?)
by danisch＠ira.uka.de 17 Dec '03

17 Dec '03

> In the following pseudocode, B is the radix in which the numbers are > represented (2^32 for a 32-bit machine), n is the length of modulus in > blocks, U is B^(n+1) mod the modulus, X is the number to be reduced, k+1 > is the length of X, and Y is the result. > > 1. Y = X > 2. For i from k down to n+1, repeat steps 3 and 4 > 3. Y = Y - Y[i] * B^i + Y[i] * U * B^(i-n-1) > 4. If Y >= B^i, then Y = Y - B^i + U * B^(i-n-1) To do a proof I rewrite the algorithm: n = len(modulus) // modulus < B^n Y = X // obviously Y = X mod modulus K = B ^ (n+1) - U // U = B ^ (n+1) mod modulus, // therefore K = 0 mod modulus // furthermore K > 0 for (i=len(Y)-1 ; i>n ; i--) { F = B ^ (i-n-1) * K // F > 0 // F = 0 mod modulus Y -= Y[i] * F // Y shrinking, but // Y still the same mod modulus if ( Y >= B^i ) Y -= F // again shrinking, // still the same mod modulus } This shows that Y was shrinking, but is still equal to X mod modulus. To see whether Y really shrinks enough: Y = sum(i=0..len(Y)-1) Y[i] * B^i In the step Y = Y - Y[i] * B^i the highest block of Y is deleted (what could be done fast by reducing the length of Y). Now Y < B^i Afterwards the same value mod modulus is added to keep Y constant: Y = Y + Y[i] * U * B^(i-n-1). Y[i]<B -> Y[i]+1 <= B U < modulus < B^n , therefore U < B^n -> (Y[i]+1) * U < B * B^n = B^(n+1) -> Y[i] * U < B^(n+1) - U -> Y[i] * U * B^ (i-n-1 ) < F Therefore after doing the addition Y < B^i + F Check of the last step: 0 <= U < B^n therefore B^n < B^(n+1) - U <= B^(n+1) Therefore in every loop B^(i-1) < F <= B^i -> Y-F < B^i Partial Correctness: Y = X [ Y = X mod Y < B^len(X) ] K = B ^ (n+1) - U [ K = 0 mod B^n < K <= B^(n+1) ] for (i=len(X)-1 ; i>n ; i--) { [ Y = X mod , Y < B^(i+1) ] F = B ^ (i-n-1) * K [ F = 0 mod , B^(i-1) < F <= B^i ] [ 0 <= Y[i] < B ] [ Y[i] * F = 0 mod , 0 <= Y[i] * F < B^(i+1) ] [ Y >= Y[i] * B^i -> Y >= Y[i] * F ] Y -= Y[i] * F [ Y = X mod , Y < B^i + F (reason see above) , Y >= 0 ] if ( Y >= B^i ) Y -= F // again shrinking, // still the same mod modulus [ Y = X mod , Y >= 0 , Y < B^i ] } Last i was n+1, therefore Y = X mod , Y >= 0 , Y < B^(n+1) This is not enough, Y < B^n is requested. The loop can't be done once more because i-n-1 would become negative. k+1 was the length of X, and n the length of the modulus. You walk down from k to n+1 . In every loop you remove one block of the number. This means you have to do len(X)-len(modulus) loops. In the pseudocode you do only len(X)-len(modulus)-1 loops. One loop seems to be missing. This may be a result of confusion whether your Y starts with Y[0] or Y[1]. I do understand the algorithm as: n = len(modulus) U = B^n mod modulus K = B^n - U // = 0 mod modulus, 0 < K < B^n Y = X for(i=len(X)-1 ; i>= n ; i--) // squeeze Block i in Number Y { // Y < B ^ (i+1) F = B ^ ( i-n ) * K // F = 0 mod modulus Y -= Y[i] * F // subtract Y[i] * B^i, now Y < B ^ i // add the equivalent Y[i] * B^(i-n)*U <= F // now Y < B^i + F if ( Y >= B[i] ) Y -= F // now Y < B^i } Last i was n, therefore Y < B^n , Y = X mod modulus , but perhaps still Y >= modulus. Ok, algorithm understood and agreed (after modifying the loop counter). Any more agreement or disagreements? Hadmut

1 0