December 2003 - cypherpunks-legacy

Re: Why ecash is traceable
by Donald M. Kitchen 17 Dec '03

17 Dec '03

Re: ecash being non-tracable if anon accounts are allowed... And if not, a market will probably develop to launder ecash. Don

1 0

Re: DD, pedaphiles, and Terrorists, oh my
by shamrock＠netcom.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- In article <9509141640.AA30602(a)zorch.w3.org>, hallam(a)w3.org wrote: >We never used to see kiddie porn on the internet. The net would go balistic if a >picture of a teen age nude was posted. Recently there has been a flood of hard >core paedophile material. There has always been nude teens on the net. The kind of pictures the most casual observer can take at any Mediteranian beach at any given day in the Summer. There is no real kiddy porn readily available on the net. I looked long and hard. Its a red herring. ANYBODY on this list seen some real kiddy porn on the net? I am talking intercourse, etc., not scans of six year olds from the Sears catalog, or naked boys building sandcastles on the beach. - -- - -- Lucky Green <mailto:shamrock@netcom.com> PGP encrypted mail preferred. - --- [This message has been signed by an auto-signing service. A valid signature means only that it has been received at the address corresponding to the signature and forwarded.] -----BEGIN PGP SIGNATURE----- Version: 2.6.2 Comment: Gratis auto-signing service iQBFAwUBMFjDkioZzwIn1bdtAQEB0gF+M67Ih6WLAl8J9bxPZOBmyjxKsxUczpqv th5x7ZuILyuf/+bFTtcLKwbOnzZkIfpJ =pjLL -----END PGP SIGNATURE-----

3 2

PGP BUG/FEATURE: multi-platform keys
by nobody＠alumni.cco.caltech.edu 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Recently I have received two keys that were unusable without modification (from wonderer and Sameer) because they were not prepared using canonical text. I had to replace all the carriage returns by hand (this is the workaround for mac users who receive PC keys). Why is non-canonical text even an option in pgp? Who would use it except by mistake? As long as I'm at it, here are some more macPGP2.3 bugs, for those who care. Trying to decrypt a message signed by a key I don't have causes a serious crash requiring reboot. Trying to select cancel in a signature dialogue just raises the same dialogue box instead of cancelling. I'm running on a powerbook 145 under system 7.0.1. Eternal!Optimist(a)anon.penet.fi (copyright 1993, Eternal Optimist [Ha Ha Ha]) 0) 0) =:()]-< <na26436(a)anon.penet.fi> -----BEGIN PGP SIGNATURE----- Version: 2.3 iQCVAgUBLNe7R4jvfLxJbYYtAQFFHQP7BCqn1JKaI09wBrWnjOF73CPz7GetaC/y XL8zqhmzDdrGdLbWB/vBDgXW7z+2EJHazbvqaUhq1GQw8bq+opC2fe4mXcw2x3Y+ wliLy1CWDtfl24L8ah//nLQMtttfG4kXjiB8JqUnS7US+W3vvH3AXxi+wZb5W2qi 7gPXMErUCUA= =Ew/6 -----END PGP SIGNATURE-----

1 0

Re: Digital Fingerprinting
by James Caldwell 17 Dec '03

17 Dec '03

On 12 Sep 95 at 11:27, Andrew Loewenstern wrote: > Such technology would be very useful in business, especially the > high-tech industry. Think of how many non-disclosure agreements > are signed every day relating to new products developed for the > software industry alone. Many companies are very paranoid and > already 'fingerprint' information by using unique code-names for > projects, for instance. i.e. the spec sheet on their new GAK > crypto product they give to Alice may be code-named 'project foobar' > but the one they give to Bob may be code-named 'project burris'... > Then, when the information leaks out they check which person they > gave the document with that code-name and they know who to sue (or > at least not give any more trade-secrets to). It's very simplistic > but it has been know to work in the past. > > Most of the real technology for doing this is much better, of > course... However, what stops you from printing out a > fingerprinted document and scanning it back in, for instance? Well, there is selective wording, mispellings, punctuation and formatting. These can be corrected easily if allowed to be transported as a text or common file type. Another way is to place the document in a PItA proprietary graphical format for transport and viewing only, stego identifier imbedded if you chose * , so that every portion of the document has some indentifier imbedded in it. Many obvious and many devious. Electronic drawings with a harmless and useless circuit(s) added on , software with do nothing code (by design!:) ). Difficult and time consuming to do, but for megabuck items, no prob. Automated for an additional fee of course. Start a service industry for such, make money, pay me back by running a fast, reliable remailer. Idea is to make the thief go to some major effort and if the scanning option is used to make the deletions as obvious and telling as the former identifier. > andrew note* Makes a neat way of putting copyright & source information in picture and sound files, somewhat useless but every trip up helps.

1 0

e$: Progress and Freedom Conference this Summer
by rah＠shipwright.com 17 Dec '03

17 Dec '03

There was a pointer here to an article in the NYT or WSJ about the conference this summer of the Progress and Freedom Foundation, with muchos net.luminarios (Huber, Myhrvold, Barlow, Brand, Kelly, Dyson, Keyworth, etc.) in attendence. Topics were mostly about e$ and the causes/consequences thereof. I just saw this thing on CSPAN, and I thought it was way cool. Here's how to get it from CSPAN in case they don't run it again. Ask for Tape number 57759-63 CSPAN prices their tapes by the hour, at $35/hr. This tape is $105.00 CSPAN Department 53 Washington, DC 20055 In case I botched something above, or you want to put this on your favorite book-entry transaction account ;-), CSPAN's phone number is 202-737-3220 and ask for Viewer's Services. Cheers, Bob Hettinga ----------------- Robert Hettinga (rah(a)shipwright.com) Shipwright Development Corporation, 44 Farquhar Street, Boston, MA 02131 USA (617) 323-7923 "Reality is not optional." --Thomas Sowell >>>>Phree Phil: Email: zldf(a)clark.net http://www.netresponse.com/zldf <<<<<

3 2

Hazard of encrypt to self?
by nobody＠eli-remailer 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I always encrypt stuff when I send it to the remailers. I always encrypt to myself along with the remailer, in case I need to look at the message again. But since trying to decrypt a message reveals the id's of the keys that can decrypt it, this reveals information about the sender of the message that wouldn't be available if I only encrypted to the remailer. A partial solution - I generated a new key pair to use for the encrypt-to-self function (a nice short one). I put some meaningless nonsense in the id, so anyone who observes my message on its way to the remailer will no know it is from Eternal!Optimist(a)anon.penet.fi (copyright 1993, Eternal Optimist [Ha Ha Ha]) 0) 0) =:()]-< <na26436(a)anon.penet.fi> -----BEGIN PGP SIGNATURE----- Version: 2.3 iQCVAgUBLObHdojvfLxJbYYtAQEZAgP8Db7qpgqy17wmP91/zl5ZioF/53IIugpW eFjJVtT/UC0wEB/wdvQSBuG2xoXFPWxwWraEtWUCvsAP580juKiGSvmpZARxnm30 yfHVuqv+cfo8FYm7KJpWGdSIFmoqvG8h/a4wiGWMg/Dnai5wOZtRt0mWDyftj+gt EzgFJSwTtIc= =xEZA -----END PGP SIGNATURE-----

1 0

Re: laptop passwords
by Alan Olsen 17 Dec '03

17 Dec '03

At 01:22 PM 9/15/95 -0800, you wrote: >I've got a dilema. > 2 laptops that I just purchased at a police auction have passwords, >from what I can tell they are passworded at the cmos level. I would normaly >just pull out the battery and clear the cmos but these are laptops and I'm >not to comfortable about doing that. Any ideas? > >laptop1 - Toshiba Satelite Pro T2400CT >laptop2 - Epson Action Note 500 c I am not familiar with those models, but if they work like a regular AMI bios, try hitting ctrl-alt-delete at the password prompt. On many passworded bios it will get you right past it. That will at least give you enough info to see if there is anything worth saving. (I have no problem giving out that trick since I have found most bios passwords pretty useless unless used against the most clueless of users.) | Minister of Forced Caffinization in the DNRC | alano(a)teleport.com | |"The moral PGP Diffie taught Zimmerman unites all| Disclaimer: | | mankind free in one-key-steganography-privacy!" | Ignore the man | | -- PGP 2.6.2 key available on request -- | behind the keyboard.| | http://www.teleport.com/~alano | <fnord> |

1 0

Re: Digital Fingerprinting
by Alan Olsen 17 Dec '03

17 Dec '03

At 12:28 PM 9/15/95 +5, you wrote: >Another way is to place the document in a PItA proprietary graphical >format for transport and viewing only, stego identifier imbedded if >you chose * , so that every portion of the document has some >indentifier imbedded in it. Many obvious and many devious. Electronic >drawings with a harmless and useless circuit(s) added on , software >with do nothing code (by design!:) ). Difficult and time consuming >to do, but for megabuck items, no prob. Automated for an additional >fee of course. Start a service industry for such, make money, pay me >back by running a fast, reliable remailer. I seem to remember a software piracy case from a number of years back. The case was settled by Apple showing that their code had been copied by the copyright "easteregg" hidden in the roms. Made for a quick and effective demonstration in court. (The story may be apocryphal, but it makes for an example as to how such things do have a (supposed) valid use in protecting code from being snagged by other companies.) | Minister of Forced Caffinization in the DNRC | alano(a)teleport.com | |"The moral PGP Diffie taught Zimmerman unites all| Disclaimer: | | mankind free in one-key-steganography-privacy!" | Ignore the man | | -- PGP 2.6.2 key available on request -- | behind the keyboard.| | http://www.teleport.com/~alano | <fnord> |

1 0

PRZ is still at large
by Philip Zimmermann 17 Dec '03

17 Dec '03

I have received several phone calls today from people who read some sort of April-fools posting on some newsgroup that I had been arrested for PGP-related stuff. Well, it isn't true. I'm still at large. I'm still unindicted. And still not sued. Someone had an idea that this would make a funny April fools joke. I hope this clears things up. Feel free to repost this message to all the relevant newsgroups, because I can't do it because the newsreader at my sight is having problems. I hope this cuts down on the phone calls. Philip Zimmermann prz(a)acm.org

2 1

SecureDrive News - Win95 yes, Iomega zip drive no
by edgar＠highnrg.sbay.org 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- SecureDrive Users, I have seen inquires here and received direct e-mail enquirinq about SecureDrive and WIN95, and the Iomega zip disk. I have recently heard from users that SecureDrive 1.3d does work on Windows 95, although it does force it into "DOS Compatibility Mode", which is a performance hit on disk access. I've also heard that LOGIN does not work from a WIN95 DOS window, so has to be issued outside of Windows. I haven't heard anything specific about SecureDrive 1.4, but I believe it should work the same as 1.3d on WIN95. I'm sorry to report that SecureDrive does not support the Iomega ZIP drive. I believe that's because the TSR driver supplied with the hardware does not support a DiskBIOS interface. I had thought that Secure Device, another real-time encryption system, would support this drive, since it works through a device driver mapped to a dos file. But an actual trial proved otherwise. The problem is that the DOS file must be present when the CONFIG.SYS DEVICE= statement for the SECDEV.SYS is processed. But the zip drive driver is a TSR that doesn't get loaded until after all DEVICE statements have been processed. There's no particular reason that Iomega couldn't have written their driver as an MSDOS device driver rather than a TSR, in which case SecureDevice would probably work, but they didn't, so it doesn't. Hope this info is helpful. If any of you discover any interesting new facts using SecureDrive, please let me know. Edgar W. Swank <edgar(a)HighNRG.sbay.org> -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBMFirU94nNf3ah8DHAQEEHAQAg4IB1XDQdFIxHto1iTPpORjuubUdhRzD t8MT350E0KZVCMSj+Vh4y4wEPw2dmms/QY1iMxuwVV/lja+l9yHaRPuxpi5EjQdv Mi8Sk73M/gcL35TTjUli5dnjzHsJ6uYt5Q51j7s0EspJ1qprQlSlgmNvvOTqDgyf /he5BfEfAU4= =GXYT -----END PGP SIGNATURE----- --- edgar(a)HighNRG.sbay.org Keep Freestyle Alive!

1 0

number theory number theory
by catalyst-remailer＠netcom.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- All right, a number theory discussion! >The integer N is prime if: > 2^N - 2 > --------- > N is an integer. Well, this is false. The above formula is derived from Fermat's Little Theorem, or Euler's Generalization of Fermat's Little Theorem. a^(n-1) = 1 mod n, n prime, gcd(a,n) = 1 ==> a^n = n mod n a^n - n = kn, for k some integer (a^n - n)/n = k, for k some integer now sub in a = 2. However, the converse of this is not true (n isn't necessarily prime if it satifies the formula). Composities that satisfy this are called pseudoprimes. For example, for a = 2, n = 341 satisfies the relation, so 341 is a pseudoprime base 2. Now it works "most" of the time, and in fact one method of testing large integers for primality is to choose a whole bunch of a's and plug in n. If a^(n-1) mod n != 1, the number is composite and can be rejected. But, if a^(n-1) mod n == 1, you can only be 50% sure n is prime. (Roughly speaking; Phil Karn notes that the PGP docs indicate a 50%, I've seen proofs that this pseudoprime test fails 50% of the time, etc. But these are upper bounds; the real percentage seems much lower and I haven't seen a tighter bound on it). There is a "strong psuedoprime" test, in which failure occurs for at least 25% of integers in the range, thus the probability that a composite will pass is at most 25%. Even better is Lucas' test, but it runs a bit slow. However, you can be unlucky and pick a Carmichael number, which will pass the pseudoprime test for all bases relatively prime to n (for all a such that gcd(a,n) = 1). Ray Cromwell advises to choose n = 561, the smallest Carmichael number (an excellent choice!) Carmichael numbers exist, they are relatively rare, formulas exists for generating some of them... Eric Hughes mentions that 1729 is the next Carmichael number... not quite true. 1105 is the next Carmichael number. (But congrats Eric for even remembering the third one!) ;) Now, some other topics: > For any number n which is prime, (2^n)-1 is also prime (Mersenne's > theorem). Hm... some confusion here. A Mersenne prime is of this form (2^n) - 1 where n is prime, but not all number this formula generates are primes. Mersenne primes are related to perfect numbers. An example of a composite of this form: for n = 11, 2^11 - 1 = 2047 = 23 * 89 > For any number n (2^(2^n))+1 is prime. (I might have that wrong, I > don't remember exactly) Well, no. These number are Fermat numbers, and while the first 5 (n=0 to n=4) but Euler showed that the Fermat number for n=5 is composite. As an aside, Fermat numbers satisfy the pseudoprime test. > For any number n, if the square root of (n!)+1 is an integer, it is > also prime. (This is interesting, but rather useless in practice) A couple of issues here: I think you may be remembering a different theorem, a consequence of Wilson's theorem. Wilson's theorem says: for any prime p, (p-1)! = -1 mod p The theorem I think you are referring to is: if P is the product of the remainders relatively prime to m, then P = +/- 1 mod m; +/- = plus or minus The congruence is +1 except in three cases: 1) m = 4 2) m = p^b (m is a power of an odd prime) 3) m = 2p^b (m is twice the power of an odd prime) I'm still trying to either prove or disprove your claim! Two followups relating the the original formula posted: > For any number a, 1<a<=n, n! mod a == 0; therefore, n!+1 mod a == 1. > n!+1 is prime. Prime numbers don't have integral square roots. Good analysis, except for the "n! + 1 is prime" part. The only thing you can say is n!+1 has no factors <= n. For example, n = 4, n!+1 = 25 = 5 * 5. > Well, it was quoted from memory, so it's possible that I made an > error, but it seems to work as stated... > For example : > (4!+1)^(1/2)=5 > (5!+1)^(1/2)=11 > (7!+1)^(1/2)=71 > I can't find a value which produces a result that is a non-prime > integer. (Of course that doesn't prove that there isn't one.) Still working on this... ;) -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLanNNYOA7OpLWtYzAQF10wP9GExbaoloiXqFe7AtXb/UzUHXhW3VDC1b mfD0RhgK2i0Dr05RW5FCvj/9i7Jxhrd3E26hTe5g4WckvIcvp+GWhE/5fkdtVMA9 THutX1ukGO/5qCxSRT4hVCeXStAz7tunkF3fcEQjPe8pSSvKxN8tw/wIZzclRDRx JDE4HYRhAz0= =OW8h -----END PGP SIGNATURE-----

4 4

Web page
by banisar＠epic.org 17 Dec '03

17 Dec '03

Hi bruce, Hows it going? I'm back from DK and the UK. I understand you're just coming back from JP (wow what world travelers we are!). have you heard from Sutherland? Do we have a title? page length? It seems to me that we should take out some of the clipper stuff and put in more on CKE since thats pretty hopt at the moment and will further distinguish us from Hoffman's book. We have also moved upto 56kb FR here in the office. Are you still interested in a home page here? Or do you have one we can link to? Dave David Banisar (Banisar(a)epic.org) * 202-544-9240 (tel) Electronic Privacy Information Center * 202-547-5482 (fax) 666 Pennsylvania Ave, SE, Suite 301 * ftp/gopher/wais cpsr.org Washington, DC 20003 * HTTP://epic.digicash.com/epic

1 0

more number theory more number theory
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- > What estimates exist for the density of large Carmichael numbers, > say 1000 bits long? I'm not sure off hand - maybe Ray can try to check the source of his formula. Carmichael numbers must be square free and the product of at least three primes... I seem to remember a formula for the distribution of square free integers, but can't quite remember it... > test? Are other probability tests like Miller-Rabin any more > provably likely to detect these? Well Phil, you are in luck! Miller-Rabin isn't fooled by Carmichael numbers. There still is a chance for failure, but it doesn't depend on the input (i.e. there are no bad inputs for Miller-Rabin like there are for pseudoprime testing). Failure depends on how many iterations you perform (n iterations = 2^-n chance of failure) and the values of the base you choose. For example, in Miller-Rabin, the Carmichael number 561 is exposed to be composite by choosing a base of 7. I'm familiar with two other primality testing algorithms (I'm no number theory wiz so there are probably more): Lucas' and Lehmer's. Well, Lehmer's method is a modification of Lucas' method. They both are slow, but have the advantage of being true. -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLaoM/IOA7OpLWtYzAQEXPQQAy1110rgCUzLtKoaTsWvGCujq3fWD7Ppz A+/2b4NmR9+YmqHl63kb9zKU1/KOfDVXsmE7o0beyRQzSNGzj2I5yEUrnz0IzBLt cy4ooiE3ED/jBBc01MBYhm5v3s9dIMJNXbsw7mBSBasqzEvHHpjH8dnGZA8QXhYT fKTlU7rKa0o= =XgrZ -----END PGP SIGNATURE-----

2 1

More on ECheques
by John Hemming CEO MarketNet 17 Dec '03

17 Dec '03

1. I was not suggesting that you put your real bank account number in the program. A test number is perfectly adequate. 2. In the UK people advertise their Bank Account numbers so that people can pay them. 3. If you do use your real bank account number, I will not pay you anything nonetheless.

2 1

HRT_web
by John Young 17 Dec '03

17 Dec '03

9-16-95. NYPaper Page Oner: "Computer Stings Gain Favor As Arrests for Smut Increase." The F.B.I. has plans for its own web to fight computer crimes. As arrests continued in a high-tech drive against the peddling of child pornography on the nation's largest computer network, the Federal authorities said today that they expected to apply the same surveillance techniques to fight many other computer-related crimes, from consumer and securities fraud to money laundering. One way of combating these crimes, they said, will be for Federal agents to go undercover and appear as prey for unwitting criminals. Timothy McNally of the F.B.I.said that while the traditional form of surveillance included activities like waiting in cars and watching suspects for hours, "the playing field has now changed" and now requires agents to become more proficient in using computers to thwart crimes. But some civil liberties experts expressed alarm at the latest surveillance technique, saying it would have a chilling effect on what is now a freewheeling medium. HRT_web (9 kb)

1 0

Chained Remailing Strategy and Tactics
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

-------------------------- In order to preserve anonymity and thwart traffic analysis in chained remailings, it would seem useful to include a very BUSY remailer in the chain, and try to ensure that the message arrives at the busiest time of the day for that remailer, from a traffic standpoint. Hitting a remailer at a slack time when, let's say, only one message arrives over a period of several hours would seem most unwise. Can some of the major remailer operators make available some "sanitized" traffic stats of average traffic by hour and day of the week? The vox.hacktic.nl remailer sounds useful in this regard, since it apparently uses a UUCP link, and batches up accumulated messages, both incoming and outgoing. When are the "best" times for chained traffic to arrive there? Can someone familiar with remailer software answer something? When a message is encrypted, using the "Encrypted: PGP" header, will everything after the end of the encrypted message itself be ignored? I ask, because this seems like a good place to introduce "padding" into the message length to thwart detection of identical messages, assuming that such extraneous material wouldn't screw something up. What's the best strategy for utilizing a given group of remailers in a chain? Which ones would be most advantageous as the FIRST link in the chain, since this is the one link that has direct address to the originator's address. How would "someone", hypothetically, follow the chain backwards? Let's say that a message traveled down the chain A -> B -> C. Couldn't someone with enough clout ask "C" where a certain message (based on header data) originated, find out it was relayed by "B", ask "B" for the source, etc. and trace it all the way back to the source? What, if anything, would prevent that? For the sake of argument, let's assume a worst-case scenario: a chained message to "president(a)whitehouse.gov" containing a seemingly credible threat to harm the President of the United States, or perhaps a chained message, ultimately posted to Usenet via a mail-to-news gateway, containing the first part, with more installments threatened, of certain highly classified U.S. military secrets. IOW, a scenario where powerful agencies are motivated enough to invest considerable resources in tracking the culprit down. While we might agree that in those two cases, the persons deserve to be caught, what's to prevent a President or other highly placed federal bureaucrat from MISusing those same resources on something less critical, such as tracking down and persecuting someone who anonymously posts "Clinton is a prick" or "Clipper sucks"? ----------------------------

3 2

Re: "Use implies consent to monitoring"
by Rishab Aiyer Ghosh 17 Dec '03

17 Dec '03

Rich Salz <rsalz(a)osf.org>: > > Please be advised that use constitutes consent to monitoring > > Use of what? Obviously of something that can be monitored - as this is apparently from the ECPA I would assume use of "electronic communications." Of course I haven't verified that this quote is actually from the ECPA, that's what the InterNIC claims it is. ---------------------------------------------------------------------- The Indian Techonomist - newsletter on India's information industry http://dxm.org/techonomist/ rishab(a)dxm.org Editor and publisher: Rishab Aiyer Ghosh rishab(a)arbornet.org Vox +91 11 6853410; 3760335; H 34 C Saket, New Delhi 110017, INDIA

1 0

Remailer chaining results
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I've done some calculations on the mixing properties of Chaum-style networks and gotten some interesting results. Recall that in a Chaum-type remailer network users use nested encryption and remailing instructions to set up a chain or "cascade" of remailers. Each remailer strips off the encryption envelope and sees the address of the next remailer in the chain or, for the final remailer, the ultimate destination. All messages are the same size and carry no distinguishing features. We assume that the opponent is monitoring all messages traffic into and out of all remailers on the net but can't see what is happening within each remailer. Let's take a concrete example and suppose there are four remailers. Everyone sets up a chain of 2 remailers, chosen at random from these four. A batch of messages is received by each remailer, which strips off the envelope and sends them on to the next remailer in the chain, where they are mixed with the other messages which chose that remailer as the 2nd in the chain, then sent out to their ultimate destinations. This model is a little artificial in that we are assuming a certain amount of synchrony of the operation of the various remailers for simplicity. (Note that for this four-node network there are twelve possible two-node chains where the nodes are different.) There are three measures that I am interested in: bandwidth used (the less the better); message mixing (the more the better); and immunity to subversion (the more the better). For bandwidth we can measure the flow through the remailer. Due to the symmetry of the situation, the inflow and outflow are equal and the same for all remailers. Message flows per remailer are the sum of the flow into the remailer from outside (the user messages), plus all flows into the remailer from the other remailers. Mixing can be measured by a probability distribution over the outgoing messages which represents how likely they are to be a given incoming message. For simplicity this can be expressed simply as the number M of messages which are equally likely to be the original (in an earlier message I used entropy which is a log measure of the same thing). I am thinking of measuring immunity to subversion in terms of how much mixing is lost by a certain number of "failed" (that is, subverted) nodes. Some networks are vulnerable to "single point failures", where a single subverted node destroys all the anonymity. A more robust network would require multiple failures for this to happen. However, it turns out that even in a multiple-failure network a single-point failure may reveal some information about the messages, which we can express as a loss in mixing. Let the total message bandwidth into the network be N packets per time unit. Due to symmetry, each node will receive N/4 packets. With the chains as defined above, the other three nodes will all be equally likely to be the 2nd in the chain, so N/12 packets are sent to each of them. Simultaneously, N/12 packets come to this node from each of the others. This is a total internode bandwidth of N/4 in each direction per node, or N total per direction. Add this internode bandwidth to the user-link bandwidth of N per direction and we get 2N total, or N/2 per node. At the beginning of each chain, we have N/4 packets come in and get mixed as each node. As the packets go out, they are sent to the other three remailers, and when they leave they may be any of the output of those three. Thus they are equally likely to be any of 3N/4 of the packets, and this is the amount of mixing we have. If one of the two nodes in your remailer chain is compromised, it provides no effective mixing. This means that your message is only mixed at one node, where it is combined as part of a batch of N/4, so that is the degree of mixing you have with a single failure. If both remailers are compromised then of course you have no mixing, which we would write as a factor of 1 in uncertainty increase. This can also be expressed in terms of a percentage compromise of the network. If 1 node is compromised, which can be represented as p=.25, then the six of the twelve remailer paths which use that node will have single-point failures with the comcomitant reduction in mixing. In other words, half of the messages will have the full 3N/4 mixing while half have N/4. With p=.50, two nodes are compromised. Two paths are safe, eight have single failures, and two have double-failures. So we have 1/6 of the messages with 3N/4, 2/3 with N/4, and 1/6 with only 1 mixing. With p=.75, three nodes compromised, there are no safe paths; half have single failures and half have double. So 1/2 the messages have mixing of N/4 and half have 1. And of course with p=1 all messages are compromised with mixing factor 1. Let me just go on and extend this analysis in one way. In the discussion of the chains, we have assumed that the two nodes in the chain would be different. Logically though one could have chains where both nodes were the same. Let us compare this network with the one we just did. There are now 16 possible chains. Total bandwidth is somewhat less (since we don't count the messages which stay in one remailer). Now only 3/4 of the messages from each node need to get exchanged. Per node, there will be N/4 messages to users and 3N/16 messages to other nodes, for a total of 7N/16 per node or 7N/4 total (above the 7's were 8's). Mixing is actually improved; there is no limitation on which input messages might map to which output ones, so we have full N-fold mixing (compared to 3N/4 above). With single-point failure mixing is again N/4 as above. The failure behavior is quite different. With p=.25, 1 of the 16 paths is totally compromised, 6 of the 16 have single failures for N/4 mixing, and 9 of the 16 have no failures for N mixing. With p=.50, 4/16 of the paths have mixing 1, 8/16 have mixing N/4, and 4/16 have mixing N. With p=.75, 9/16 have mixing 1, 6/16 have N/4, and 1/16 have N. It's not clear what measure is useful to compare these failure situations. A double-point failure seems much worse than a single one. I wonder whether taking a geometric mean (which would be equivalent to taking the arithmetic mean of the entropies) would be valid. If we did that for the p=.25 case, we get average mixing of .59N^(15/16) for the self-chain network, and .27N for the network where all chains are two different nodes. For N less than about 250,000 packets per (network-wide) batch the self-chain network provides superior average mixing in the p=.25 case by this measure. Sparing the math, for p=.50 the self-chain network is superior for batch sizes smaller than 29 packets, and for p=.75 the self-chain network is only superior for batch sizes smaller than 16 packets. This suggests that if the network is likely to be mostly safe then the extra mixing allowed by same-node chains is worth the small increased risk of exposure. But as the chance of encountering bad nodes rises it becomes unwise to take this chance. Here is a quick summary of the extension of these results to larger numbers of remailers and longer chains. Let there be R remailers and let the chain length be K. Let the number of message packets per batch (network wide) again be N. (I will neglect the differences between same-node chains and different-node chains as they are generally small effects on the order of 1/R.) Bandwidth per node is approximately KN/R. Network wide it is therefore KN. Adding remailer hops increases network bandwidth loads directly in proportion to the number of hops. Mixing is approximately N for K=2 and up, which is the maximum possible. For K=1 mixing is N/R. Fault tolerance is interesting. A K-length cascade is invulnerable to up to K-2 failures! At K-1 the mixing decreases from N to N/R, a significant decrease. And with K failures of course the mixing drops to 1. I was surprised how robust these networks are. The reason is that with even K-2 compromised remailers in a K-length cascade there still remains a safe length 2 cascade, and as we saw above that provides N-fold mixing. This provides some guidelines on the choice of K. First, K should clearly be at least 2. The increase from K=1 to K=2 increases mixing from N/R to N, a considerable increase. Secondly, K should probably be at least 3. This will provide full mixing even if you are unlucky enough to choose a compromised remailer. Beyond this, you can calculate that with a chain length of K and probability p of a compromised node, the expected number of compromised nodes in your chain is Kp. This suggests that you should choose K large enough that Kp is well below K-2. If you estimate p=.50, for example, you might choose K=8. The binomial theorem states that the chance of x failures out of k nodes where the probability of each failure is p is (p^x)*((1-p)^(k-x))*k!/x!(k-x)!. In this example, the chance of 7 failures out of 8 is about 3% and the chances of 8/8 is about .5% for a total risk of 3.5% that you won't be fully protected. Now, how many people read this far? ;-) Hal -----BEGIN PGP SIGNATURE----- Version: 2.1e (yikes, where'd I find this old version!) iQCVAgUBLkbYB6gTA69YIUw3AQHligP+PBRC1pmZ6+T10WCQ91SZ2GdYX4/iEsKQ eMfCLlQ0PFbPEWZ5TaDwbOLCCUSBAbb6OO3Y2U8SHF/zZKJLrHI09/Ssl/ZeQ3st 9G9JrncU9Wo7Z9N1zMPJuQy21qFKNOkAwVQHxThObMSxQWh+TWem8lDKzm6ea0VH sejMQG+nVyo= =BWsP -----END PGP SIGNATURE-----

1 0

Wash Post coverage of NIST Key escrow-export
by pfarrell＠netcom.com 17 Dec '03

17 Dec '03

Today's washington post has some coverage (finally) of the encryption/key-escrow/FIPS standards issues. Headline is Feuding again erupts over encryption exports I typed in the text. url: http://www.isse.gmu.edu/~pfarrell/nist/post.html Pat Pat Farrell grad student http://www.isse.gmu.edu/students/pfarrell Infor. Systems and Software Engineering, George Mason University, Fairfax, VA PGP key available via finger or request #include standard.disclaimer

1 0

Re: Linking = Showing = Transferring?
by shamrock＠netcom.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- In article <v02130501ac7e53505548(a)[18.157.1.107]>, joseph(a)genome.wi.mit.edu (Joseph Sokol-Margolis) wrote: >Here's my question, and a thought to ponder: If it is/becomes illegal to >have links on a homepage to pornography because it is ruled as the same as >having the pornography there; then can you have links to pages with links >(the same as having it) to pornography? This repeats, so would it be legal >to links at all? The answer is trivial. If it pisses of the fascists in power enough, you go to jail or get killed. - -- - -- Lucky Green <mailto:shamrock@netcom.com> PGP encrypted mail preferred. - --- [This message has been signed by an auto-signing service. A valid signature means only that it has been received at the address corresponding to the signature and forwarded.] -----BEGIN PGP SIGNATURE----- Version: 2.6.2 Comment: Gratis auto-signing service iQBFAwUBMFjNIyoZzwIn1bdtAQFtXQGAkgf19PR9xYU91knoFXfYCR5NaPyWCcPz BfpmmPksdMUFCk73R4rMJIc+LruHQGb9 =2aYL -----END PGP SIGNATURE-----

4 3

(Noise) X-Files anarchist
by dsc＠swcp.com 17 Dec '03

17 Dec '03

I hardly watch TV ...blah, blah..., but I saw X-files last night. A very minor character was a cryptohacker who--to his surprise--brute-forced a military key for some files. He was described as an anarchist. To _my_ surprise, this anarchist was not associated with violence or lunatic ravings. He was shown as reading a book entitled something like Survey of Modern Conspiracy Theories. I see this as an improvement in the stereotype of computer-associated anarchists. And this I see as a good thing, since I think there is a lot that could be learned from anarchists. I think the character was killed, but I may have missed something. Dar =========================================================== Dar Scott Home phone: +1 505 299 9497 Dar Scott Consulting Voice: +1 505 299 5790 8637 Horacio Place NE Email: darscott(a)aol.com Albuquerque, NM 87111 dsc(a)swcp.com Fax: +1 505 898 6525 http://www.swcp.com/~correspo/DSC/DarScott.html ===========================================================

2 1

Re: CYPHERPUNK considered harmful
by David C. Lambert 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Peter Trei <trei(a)process.com> wrote: > Terms like 'cypherpunk' and 'cryptoanarchy' tend to pigeonhole us as > nutcases for many people, and are a barrier to getting our ideas > across. I'm not saying this pigeonholing is correct - in fact I despise > people who judge a book by it's cover, but so many people DO make such > snap judgements that we need to take this into account when talking to > the general public. Two of the responses to the "An opportunity..." post unapologetically admitted that they refused to read the text of the message due to the presence of the word "cypherpunk" in the first sentence. Shows you what we're up against. BTW, I'm not for changing the name of the list, but I do see the logic in a more establishment-friendly name to use when lobbying the public. David C. Lambert dcl(a)panix.com (finger for PGP public key) -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBMFneAqpplsfgM88VAQE+5wP9GZrIoComyFDeQ6brTLJwZ2oamry3IfC7 XhlqMlAZnC2b1w521nm085nZKtO9/Ru9Fw3BllPinG8nGcBggr9zkH4Ba+Zbezh0 FKK29d7wFGQz1d5JfyYCZhl4dTBdbpnU4jd5Rb27XtvP9livAeR/HIwNAJMccSCX qO/kxlVRRZg= =1Lxr -----END PGP SIGNATURE-----

3 2

More on remailer chaining
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I have one small addition to the analysis I did yesterday of remailer chaining effects. Previously I was assuming that there was a uniform distribution of messages across remailers, so that all saw the same number of packets. How does this change if some remailers are used more than others? Again I will sneak up on the problem by taking a concrete example. Suppose there are two remailers and that we are using two-remailer chains which include the possibility of using the same remailer twice. Suppose one of the remailers is used much more than the other. Take an extreme case, where remailer 1 is used 90% of the time and remailer 2 is used 10% of the time. If we suppose that these probabilities affect both the choice of the first and second remailer in the chain, then the four possible chains have the following frequencies of use: 1,1 81% 1,2 9% 2,1 9% 2,2 1% Notice that this also implies that 90% of the messages enter the net at remailer 1 and 10% at 2, and also that 90% leave via 1 and 10% leave via 2. Now, ignoring for a moment the fact that there may be some reason people are shunning 2 (they suspect it is compromised, or it is unreliable, or something - but maybe it's just new and a lot of people haven't heard about it yet), what is the safest way to use this network? The key, I believe, is to imitate the observed statistics in your own choice of a chain, at least for the 2nd hop. 90% of the messages coming out of the first stage of either remailer will go to remailer 1. If you want your message to be lost most effectively among the others, you should choose remailer 1 as your own 2nd hop 90% of the time. This way your message will be 9 times more likely to go to 1 than 2, but since there is 9 times the traffic going to 1 than to 2 it will be perfectly masked. The result will be that your message is equally likely to be any of the N messages coming out of the remailer. Your statistics will match all of the others. Therefore, you get a full factor of N mixing with such an unbalanced network, just as much as you get with a perfectly symmetrical network - as long as you imitate the network statistics. The choice of the first remailer in your chain does not appear to be critical. We assume the opponent can see which remailer you have chosen (by tracking your message from your site to the remailer) so there is no advantage to choosing 1 over 2 as far as secrecy. You will get full N-fold mixing in either case. This is a bit counter-intuitive; it might seem that choosing 1 is superior to choosing 2 in terms of mixing. But look at a specific example: Suppose 100 messages enter the network, 90 at 1 and 10 at 2. After the first step, 9 messages go from 1 to 2 (10% of the 90) and 9 messages go from 2 to 1 (90% of the 10). Then 90 messages are sent from 1 and 10 from 2. Now, if your message entered at 2, but had a 90% probability of going to 1 at the second hop, then there is a 90% chance that it ended up as one of the 90 messages leaving 1, and a 10% chance that it ended up as one of the 10 messages leaving 2. This tells observers exactly nothing about where your message is. So choosing 2 as the first hop is just as good as choosing 1. Although I have not yet extended these results to longer hops and larger numbers of remailers, my guess is that the same general rule will apply there as well. This suggests that it will be useful and important to have accurate information about the usage levels of the various remailers so that you can accurately mimic those probabilities. How bad is it if you don't have accurate usage information? According to my calculations, in the case of two remailers, if the actual probabilities of the two remailers being used are p and 1-p, and the probabilities you use are q and 1-q, the mixing level you get decreases from N to N * (p/q)^q * ((1-p)/(1-q))^(1-q). If q=p and you have accurate information there is no reduction. In the example above, with p=.9, if you didn't know this and used q=.5, your mixing level reduces to N*.6. This is not a huge reduction even for this rather extreme case, but I can't guess how this will extend to larger networks and chains. Assuming these results do hold true, though, it suggests some interesting "market" dynamics. Patterns of usage of the remailers may tend to be stable since anyone who departs from the current usage pattern will stand out and hence lose security. It may be difficult for new remailers to become established since their initial usage level will be low, making it risky to use them to any significant degree. These considerations are somewhat similar to situations where there are competing but incompatible standards (e.g. Beta vs VHS VCR's) in terms of the barriers to entry. There may also be considerable misinformation about usage levels. It will be to the advantage of a site to exaggerate the number of messages they are handling. Especially if noise messages are used (a strategy I haven't tried to analyze yet) it would be easy to generate bogus statistics. Maybe some organization could collect statistics by polling remailer users about their practices rather than believing the operators, and make that information available. Another point is that, assuming that remailer operation is actually going to be profitable some day, there will be advantages to being one of the first to market. Getting your remailer widely known and used in the early days could establish market leadership which will have considerable staying power just from the inherent properties of how these networks work. Heavily-used remailers could charge premium prices while the "little guys" have to be cut rate in order to grow, compensating users for the loss of security they will experience. Maybe this will encourage people to make the investment to become what Tim May has called "Mom and Pop" remailers. This might be the golden opportunity to get in on the ground floor. For more information, send $10 in digital cash for our investment kit: "How you can make a fortune running anonymous remailers!" Please include an anonymous return address. :-) Hal -----BEGIN PGP SIGNATURE----- Version: 3.14159 iQCVAgUBLkghT6gTA69YIUw3AQFaJgP/e7RRWrEowQDQ9RdN+w9wC5zQ3Zod2w5n oeZLFlMJFzEjer2gxjh0yt+a0CPJA1p33W1BvxNODI2nmPHiFeVcD24L9oNzoyf9 QBrUMAJiuR09QQCPz8MjBwXdIXD1hU25hMiCN/drrJuRCgsFpp1wPlmWU2EnHK4g uoiDsWb4Wg4= =l7nS -----END PGP SIGNATURE-----

1 0

Re: CYPHERPUNK considered harmful
by aba＠atlas.ex.ac.uk 17 Dec '03

17 Dec '03

Dimitri Vulis <dlv(a)bwalk.dm.com> writes: > Anything posted to the main cypherpunks mailing list and the > spun-off mailing lists (steganogrpahy, remailers, nym servers, etc) > could be posted to the newsgroup by maiking one of the mail2news > gateways a subscriber. Not sure that would be a good idea, then the news group would *be* cypherpunks, and the cypherpunks list would have effectively become a USENET newsgroup (albiet with some mail2news stuff behind). For me part of the fun of cypherpunks is that things get discussed here which don't get discussed anywhere else, things happen here 1st, rc4 leak, that kind of thing, if it's happening crypto wise, it's on cypherpunks. You might lose some of the community feel to it if it was gated straight to a USENET group, I mean at that point the majordomo list server would have become a news2mail gateway. I was thinking more that you would have a separate newsgroup, but perhaps my feelings are more of the elitist stuff Peter was referring to, still I see no need to change a good thing, can't the two co-exist separately, Peter's stated aims didn't co-incide with my perception of the cypherpunks lists goals. ie the list (modulo cpunks write code wars) seems to me to be about people who share the common goal of free crypto for everyone, discussing how to go about this, writing code to help it happen, and analysis of what the government is up to where it infringes on cypherpunk goals. I think Peter's goals sounded more amenable to a forum which was a) easily accessible, and b) had a stated aim of a kind of cypherpunk technology transfer forum. Perhaps I am being eliteist, so shoot me, but I like cypherpunks the list the way it is! For your convenience reading cpunks there are a couple of solutions to the deluge of mail problem: there used to be a nntp server at nntp.hks.net which nntp served cypherpunks as a newsgroup (I haven't used this in a while as it appeared to be down or empty or somthing for a while), there are filters which you can set up to put all cpunks traffic in a separate mbox, and there are a couple of digest forms of cpunks around also. The filter won't help your down load time, the digest (I think Hal offers / was offering an encrypted digest which would also be compressed) and the nntp server might. > > Reckon cypherpunks as a group has enough readers to hmm, push through > > a vote for group creation, if the majority thought it was a useful > > exercise. A group soley for what? cypherpunks technology, social > > No vote is needed to create an alt.group: something like > 'alt.security.cypherpunks' or 'alt.privacy.cypherpunks'. Just post a proposal > to alt.config, post many articles seconding the proposal, let it be discussed > for a week, then issue a newgroup. Me, I quite like the cypherpunk name, I reckon it would be best from a distribution pov to get a big8 group, as you discuss. > To create a newsgroup in the 'big 8' (comp., sci., etc) one needs to > deal with unpleasant control freaks like group-advice, news.groups, > and David Lawrence. It takes up to 6 months. On the other hand, > many sites that have Usenet have comp.* but not alt.*. Their users > would still have to use the mailing lists or find another site. To > create something like comp.security.cypherpunks (I think this would > be the most appropriate place, since there's already c.s.announce, > c.s.misc, and c.s.firewalls), talk to the group-advice cabal. I'll let other people argue about newsgroups names. Adam

1 0

Re: CYPHERPUNK considered harmful.
by Peter Trei 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Tim writes: >Indeed, some people here have been calling themselves by various names >("CypherGeek," "Crypto Rebel," etc.) for a long time. Personally, I despise >the terms "geek," "dweeb," and "nerd," and think anyone who calls themselves >by these insults is not "reclaiming and deconstructing the labels of the >oppressor class," as the slogan goes, but is merely insulting themselves. >Blacks who call themselves "niggers," homosexuals who call themselves >"queers," and computer programmers who call themselves "geeks" and "dorks" >are all playing the same game. I fail to see how you can write this, and then claim it's OK to call yourself a 'punk'. >A better approach is for the "conservatives" who want a "more conservative" >name to simply do what I suggested: form a new group and name it what they >like. They wouldn't have to quit this list, they would just be able to >cleanly recruit for their new list. >Seems simple to me. And honest. And less devisive than trying to change the >name of a group with a long history (by modern standards) and with several >well-known achievements. Who is trying to be dishonest, and what are they trying to be dishonest about? I'm really confused by this. Tim, I don't want to take your toy away, or minimize the acheivements of the cypherpunks (among whom I somewhat presumptively include myself). OK. Let's NOT drop the term 'cypherpunk'. However, I do think we need to have available another term for 'people with our interests' if we're going to have maximal effectiveness. What follows is my original followup, written last night. - --------------------------------------------------------------- I've gotten a number of responses to my post 'Cypherpunk considered harmful." It's pretty clear that a lot of people did not really understand what I was getting at. This is my fault. I'll try to clarify. What is the situation on the ground, here in mid September 1995? It's utterly clear that the US and other governments, are dead set against the widespread use of strong, unencumbered crypto. 'They' are using many different (and weak) pretexts to delay or prevent it's incorporation into commercial software. If there is a hidden state agenda, 'our' belief is it is that the state wishes to retain it's current ability to spy unseen on the citizenry, wiretapping with or without warrents, and examine stored information without the cooperation of the owner. The general 'cypherpunk viewpoint', if I may generalize, is diametrically opposite: It's not really a hidden agenda, since frequent rants make it pretty clear. Since we're not a unified command and control system like the state, different cypherpunks have different goals, but I hazard that most would agree with: 'Strong, uencumbered cryptography is an empowerment tool for the electronic age; it strengthens the individual in relation to the state, allowing him (or her) to communicate privately, and store his data as securely as if it resided in his own head. The widespread use of strong cryptography will lead to social changes that we regard as desirable.' If you accept this arguably paranoid vision of our situation, then it's clear that we are in a battle for the hearts and minds of the people: The State wants to convince people that their safety and prosperity lies in trusting Big Brother to watch over and take care of them, and nothing but danger lies outside of BB's line of sight. The cypherpunks take a much more libertarian viewpoint; that the state is already too strong and intrusive, and that cryptography will not only allow us to engage in commerce and protect ourselves against crime, it will also allow us to act outside of improper state control. But time is running out. All the state has to do is convince people that strong unescrowed encryption is needed only by criminals - if you trust the state, there is no reason to object to it being able to listen in - after all, all laws are reasonable, and the state won't listen in without a warrant. 'We' have a harder task. We need to convince people that they need encryption - it will protect us against criminals, snoops, and hackers, and the state's proposals will not do this. It's a much more subtle thesis, without the easy emotional hooks that can be exploited by the state to promote it's position. I don't think we can turn people into libertarians, and then get them to adopt cryptography to protect themselves from the statists. At least, not enough people to count in the short term, and the short term is where we need to focus at the moment. I *do* think we can persuade people that they need cryptography to protect themselves against criminals, and that the governments proposed standards for key length limits and key escrow will make the crypto so weak and insecure that it will cause more crime than it will prevent. That's a critical point - that the individual can, without difficulty, use cryptography to protect him or herself against criminals more effectively than he could by relying on the state. The memes we want to propagate are 1. "I need strong unescrowed cryptography to protect myself against criminals." If we can convince a large portion of the population of this (and we have truth on our side), we'll be well on our way. 2. "The government's initiatives on encryption, while they may be well-intentioned, are worse than useless, and will endanger me and my loved ones." If we can get this through the public's collective head as well, we'll have won the battle. This goal is where the 'cypherpunk considered harmful' title comes from. I want to propagate these ideas. If I go up to an average person and tell them "The State is working to ensure that it can spy on your every communication, and can't be trusted to follow it's own rules for doing so. Use cryptography to help promote cryptoanarchy!", he will usually dismiss me as a nutcase. If I say "Criminals and hackers can tap your internet communications, defraud your cell phone account, eavesdrop on your calls, impersonate you, steal your credit card numbers, and spy on your business secrets. I can tell you how you can easily prevent this", then he is interested, and wants to know more. Terms like 'cypherpunk' and 'cryptoanarchy' tend to pigeonhole us as nutcases for many people, and are a barrier to getting our ideas across. I'm not saying this pigeonholing is correct - in fact I despise people who judge a book by it's cover, but so many people DO make such snap judgements that we need to take this into account when talking to the general public. For those who have responded, *I* do not have a problem with the term cypherpunk. However, I know that there are many people who do, people we want on our side. I propose that in communicating with the cryptographic laity, we should be emphasizing the anti-crime aspects of crypto, as well as it's enabling impact on commerce and the American software industry. We'll get a lot more converts with this approach than we would by emphasizing the political aspects of cypherpunk, no matter how important we think they are. Think about it. Which future would you rather see: 1. Lots of people demanding strong unescrowed crypto for what you consider rather peripheral and politically unaware reasons, which you have educated them in. 2. Unescrowed crypto banned, with those who promote considered aiders and abeters of the four horsemen, (but boy are they ideologically pure!). Peter Trei I probably won't see any responses till Monday. I'm going for a weekend at a resort on Lake Winnipesaukee. If you need me, I'll be in the hot tub. :-) -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBMFmmK1QxhKXe9s8VAQE+tgP+NcKdGvBeJpa8AHGr/RhKCHudZP2qmRNP J699UCCvN006ywIjurjuis48VPj26uxBkcRbeuzNxGlKmZqXIkS0vEcFjrrZhBsK V9dLFFGTZ6JG3nK++mleW1wQB/F0azXXvXKJWa6R+Tnj1oo2ADUxGNQMs9IVmgR3 zsyK3fej+IQ= =YqDH -----END PGP SIGNATURE----- Peter Trei Senior Software Engineer Purveyor Development Team Process Software Corporation http://www.process.com trei(a)process.com

2 1