- cypherpunks-legacy - lists.cpunks.org

Re: broadcast encryption
by snyderra＠dunx1.ocs.drexel.edu 17 Dec '03

17 Dec '03

At 3:37 PM 8/3/94, Brian Lane wrote: >and my Linux system. Maybe they'll changes the rules if enough 'new' hams >could get together. I say 'new' because there are alot of old hams who >like things the way they are - they even fought the change in rules that >allow no-code users to get a license without having to take morse code(a >waste of time in my opinion). This may be blasphomey on this list, but I don't have a problem with the restriction on obscuring the meaning of transmissions on the amateur bands. My reading of the rules (and I'd probably double-check with the appropriate persons at the American Radio Relay League before doing so) is that cryptology that doesn't obscure the message, like digital signature, is permitted. After all, the encrypted portion of the message is easily computed or decoded, so its meaning would not be obscured; it just couldn't be generated by anyone else. This could be a solution to a problem I've always had with the digital amateur radio operations: the ease with which someone could be spoof. It's very easy to configure your system to transmit with the callsign of another, or even with a non-existant callsign. This is possible with wired networks as well, but requires access to the machines or cables on the network. With wireless, you can do it from anywhere, and as long as you keep connections short, not get tracked down. With digital signatures, it is possible to authenticate every message, or even better, every packet, and ignore unsigned message/packets. Getting a ham radio license (within the US) is fairly simple. You take a test the questions come from a pool which is public knowledge, pay the volunteer examiners ~$5 to cover the cost of the test(s), and wait for the FCC to send you your 10 year license. No fee for the license itself, and no renewal fee. Given the public nature of amateur traffic, I don't see a problem with the restrictions. There is no such thing as "private communications" on the ham bands. Never has been. If you want such privacy, use one of the commercial bands where it's allowed. Bob -- Bob Snyder N2KGO MIME, RIPEM mail accepted snyderra(a)dunx1.ocs.drexel.edu finger for RIPEM public key When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.

2 1

Ann: PGS v0.99e
by ER CRAMER 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Just another release of Pretty Good PGP Shell: PGS v0.99e is the bug fix for the public beta version of PGS v0.99d. PGS is a very good shell for PGP. PGS has an advanced keyring management system and reads keyring it selves!!! PGS is very easy to use! PGS supports PGP versions 2.3a 2.6MIT 2.6ui. And ViaCrypt PGP versions 2.4 and 2.7. Special requirements: 80286 or up. Changes: A few minor bugs where fixed. The major memory leak in the key information section has been fixed. Because some users did have some problem with PGS changing the color pallet a new color mode has build in that uses standard (STD) colors. It is now possible to save the color mode (PGS, STD (default) or MONO) in the configuration file. PGS v0.99e is available for download at the following sites: Internet: (Right now) wuarchive.wustl.edu:/pub/msdos_uploads/pgs/pgs099ee.zip 128.252.135.4:/pub/msdos_uploads/pgs/pgs099ee.zip (notice that the filename is pgs099eE.zip and not pgs099e.zip, pgs099e.zip was not uploaded correctly). Fidonet: 2:282/317 Request: PGS099E.* 2:280/202 Request: PGS099E.* - -- ... If you outlaw Privacy, only Outlaws will have Privacy! Eelco Cramer <crame001(a)hio.tem.nhl.nl> ------ - -------------------------------------------------- -----BEGIN PGP SIGNATURE----- Version: 2.6 iQCVAgUBLj7UH4DAdPKe9hHLAQFowQP/Qm2AYxxRGBCf8tMFUE5KJGPd97GsXOkZ 9fnG/ofYHkpVgTRNc/jiTWd7502zQdcI95DM0WZoDirnHVjw1Iqxq8HXaxJ9h37X N+d/ClHTfWao9BTXN4FiO0urY/383kFWSlanZYSTQxatHeiIC/9BRVWoIbj8DndX sRMrKVSfrTY= =csjx -----END PGP SIGNATURE-----

1 0

Re: Latency vs. Reordering (Was: Remailer ideas (Was: Re: Latency vs. Reordering))
by jdd＠aiki.demon.co.uk 17 Dec '03

17 Dec '03

In message <9408051716.AA14773(a)ah.com> Eric Hughes writes: > Back to the start, I guess. > > > Specifically cryptographic elements are easily added to the system > > * packets can be delayed for random intervals > > Let me repeat: > > REORDERING IS OF PRIMARY IMPORTANCE FOR REMAILER SECURITY. > > ADDING LATENCY IS NOT. No need to shout, we heard you the first time. ;-) In a system that is carrying continuous traffic, random packet delay is functionally identical to packet reordering. If messages are fragmented, random delays on sending packets out is functionally identical to reordering. More importantly, RemailerNet as described defeats traffic analysis by more significant techniques than reordering. Reordering is a weak technique. The introduction of noise, 'MIRV'ing of messages, fragmentation of messages, random choice of packet routes, and encyphering of all traffic are stronger techniques. -- Jim Dixon -- +-----------------------------------+--------------------------------------+ | Jim Dixon<jdd(a)aiki.demon.co.uk> | Compuserve: 100114,1027 | |AIKI Parallel Systems Ltd + parallel processing hardware & software design| | voice +44 272 291 316 | fax +44 272 272 015 | +-----------------------------------+--------------------------------------+

4 4

Chaum ecash
by Joseph Block 17 Dec '03

17 Dec '03

Sorry to clutter the list but: I had sent email about beta-testing Chaum's ecash scheme. I've since lost my copy of the original announcement (had a drive crash - joy!) and have never received a reply. Could some kind soul send me a copy of the announcement? thanks jpb(a)gate.net

1 0

Re: Voluntary Governments?
by snyderra＠dunx1.ocs.drexel.edu 17 Dec '03

17 Dec '03

At 2:15 PM 8/4/94, Timothy C. May wrote: >"Opening the list up"? I've been on the list since the beginning, and >anarcho-capitalist, politico-cryptologic themes have _always_ been >with us. The role and nature of government in the presence of strong >cryptography is a recurring, and important, theme. I have no problem with politico-cryptologic themes or discussions on the role and nature of government in the presence of strong cryptography appearing in Cypherpunks. They certainly beat "PGP good, Sternlight/Detweiller bad" messages. I only question generic discussions of forms of government without any reference to cryptography or even privacy. I don't see that as any more appropriate for the group than abortion or health care debates, if it doesn't have a cryptological theme. There's no question you've been involved longer than I. I've only been on the mailing list for 3 or 4 months. In fact, I tend to make a concerted effort to read messages from you, because they tend to be concise and reasonable. I'm just not sure that political commentary without a reference to cryptography fits the purpose of the cypherpunks mailing list. If you disagree with me, fine, we'll disagree, and I'll mentally filter out another subject line. :-) It's not all that hard, which is why I originally responded via direct email rather than sending it to the list. Bob -- Bob Snyder N2KGO MIME, RIPEM mail accepted snyderra(a)dunx1.ocs.drexel.edu finger for RIPEM public key When cryptography is outlawed, bayl bhgynjf jvyy unir cevinpl.

3 2

Re: Anonymous code name allocated. // penet hack
by L. Todd Masco 17 Dec '03

17 Dec '03

Doug Cutrell writes: > >So, in between my check and yours, Todd posted the cypherpunks hack > >telnet 25 and the anonymous id disappeared. Hmmmmm. I wonder how > >that happened? :-) > > > >Thanks Todd! > > > > Rick > > I'd like to understand what Todd's "hack" means... I assume that he's > talking about telnetting to the sendmail port. But I thought that anyone > could unsubscribe anyone from cypherpunks by simply sending a message with: > > unsubscribe cypherpunks obnoxious(a)jerk.com > > It isn't even necessary to forge the return address, because majordomo > doesn't check. I just pulled majordomo's help file. It's appended below. In my experience, listservers will clear any commands that don't come from the person affected by passing them on for processing by the list maintainer as a security precaution. I had assumed majordomo did this, but I'm not certain. -- Todd

2 1

RE: 'Anon' Fake
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- > The mechanism employed was obvious and simple -- someone subscribed an > anXXX address to the list. Anyone looking at the subscription list can > tell that, on their own. This technique has been used before. Is there any evidence to link this anxxxx person to the "tuna fish" spam? Given the fact that by doing a "who cypherpunks" will yield anyone, including various TLAs a list of potential "troublemakers" (by their standards), might this subscriber merely be trying to protect his real identity from prying eyes? Actually, I had thought of doing so myself, but didn't want to over-burden Julf's server with such a high volume of mail, plus having the mesages get cut off whenever a dashed line was encountered. FWIW, other list servers have a "set conceal" option which, upon request, keeps a person's address from becoming public through the "who" command. - -- Diogenes -----BEGIN PGP SIGNATURE----- Version: 2.6 iQCVAgUBLj70d+Rsd2rRFQ1JAQFSlwP+NkXJSaKlUKuFnLhzHWjGxd6X+prUlPiV NsKGBXON6ATKGTvcCE8IG+A17MwkxNi4PxnOvmdCyyI+940Rz9uDmZo8qSW5EWip 7oQ/mEFGnxRY7wkw+99QHpASxBE/9nJSvuCM0AwGfZ5/0rMSUE1t2M52PAfZcELa 9G+cEM9GiPM= =tzW7 -----END PGP SIGNATURE-----

1 0

Re: Art Gallery on internet needs PGP signatures
by tcmay＠netcom.com 17 Dec '03

17 Dec '03

Arsen Ray A. wrote: > Now the question I have is what if the forgery is done so well, would it > fool the scanner and pass the signature? ie: What if you take a picture > of the painting or part or whatever using high resolution film, etc. > Would it pass the signature? I don't know (but I'll make some speculative/engineering guesses), as I never saw the system...like I said, I heard a verbal description of it some years back (from Jim Omura, of Cylink, now famous for suing RSADSI). First, photo-reproduced art is almost never confused with the real thing. Surface texture, brush strokes, light at different angles, etc. Second, I could imagine the scanner system using light at two or more angles specifically to provide better protection against forgeries. (In the intended main applicaiton, that of detecting machine part forgeries, the pattern of natural scratches and abrasions, and grains/regions in the metal, would be essentially impossible to mechanically reproduce. Not completely impossible, but very difficult, and hence not cost-effective.) Anyone interested in pursuing this may be find out if Light Signatures still exists. Omura may know, and someone at RSADSI probably would have records of their licensing arrangements. --Tim May -- .......................................................................... Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)netcom.com | anonymous networks, digital pseudonyms, zero 408-688-5409 | knowledge, reputations, information markets, W.A.S.T.E.: Aptos, CA | black markets, collapse of governments. Higher Power: 2^859433 | Public Key: PGP and MailSafe available. "National borders are just speed bumps on the information superhighway."

1 0

Re: How do I choose constants suitable for Diffe-Hellman?
by Matthew Ghio 17 Dec '03

17 Dec '03

0x7CF5048D wrote: > How do I choose constants suitable for Diffe-Hellman? > According to _Applied Cryptography_ n should be prime, > also (n-1)/2 should also be prime. g should be a primitive > root of unity mod n. n should be 512 or 1024 bits long. > Are there any other requirements? > > How can I choose such numbers? Are such numbers published > anywhere? Yes, Phil Karn posted a list of such numbers to the list last May, and the program used to generate them. Since some people have expressed their distaste for large files re-posted/forwarded to the list, I won't send it, but you can get it from ftp cs.cmu.edu: /afs/andrew.cmu.edu/usr12/mg5n/public/Karn.DH.generator

2 1

How to find a primitive root of unity, for Diffe-Hellman?
by 7CF5048D＠nowhere 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- > >How do I choose constants suitable for Diffe-Hellman? >According to _Applied Cryptography_ n should be prime, >also (n-1)/2 should also be prime. g should be a primitive >root of unity mod n. n should be 512 or 1024 bits long. >Are there any other requirements? > >How can I choose such numbers? Are such numbers published >anywhere? > Ok let me take a stab at finding g assuming n has been choosen to meet the above requirements. (I hope my math is still good.) Let Zn be the field defined by the prime n. Let G be the multiplicitive group defined in Zn. So |G| = n-1. Now n is large so 1 is not equal to -1 in Zn. Let N be { 1, -1} in G. It is a subgroup. Zn is abielian so it is Normal. We can consider the canoical map: G ---> G/N The order of G/N will be (n - 1)/2 which we are assuming to be prime. G/N is a cyclic group with no non trivial subgroups. Every element not = 1 is a generator. Pulling back to G we find that if g is not a root of unity, then the other member of its co-set = -g is! So take any g and raise to (n-1)/2 power. The result will be equal to 1 or -1. g raised to any lower power will not be equal to 1 or -1. Since (n-1)/2 is a large prime, it is odd. So if g to the (n-1)/2 is = to 1, then - -g to the (n-1)/2 = -1. So we can find a g which raised to the order (n-1)/2 power is = to -1. So g to the (n-1) power is =1 and g is a primitive root of unity. Have I made any errors? Did I get it right? -----BEGIN PGP SIGNATURE----- Version: 2.6 iQCVAgUBLmt8dw2Gnhl89QSNAQHzmAP9GUGAmFcbgMyWxKtrzEvQYJS33FXGoGmr w4rXblv14lkwJX32hpoRKmicm3bdND2OPGgmM4EefGYggj+iCI+NU+l6II+MxhjY C4Rk3Xjn59H81FhNdfcNqOU9AirjwMBSqKzYtNCfbedB6HuQDCTeLSU5pjI5PSEQ wvFP7F3i5rY= =0r8J -----END PGP SIGNATURE-----

2 1