- cypherpunks-legacy - lists.cpunks.org

RE: Magic Money attack
by Hal 17 Dec '03

17 Dec '03

>From Pr0duct Cypher: > [ only 18 bytes are checked ] > > Easy enough to fix. Will code this. I just sent new PGP Tools and Magic > Money updates to MPJ. He must be getting tired of me sending him new code > all the time. :-) The latest version does protect against garbling of the > message from client to server. I think it's great that you are able to fix these things so quickly. It's natural that there will be a lot of shaking out in any initial release. > How many small primes would it take? How would he know what numbers to > multiply to get the coins? Just create random coins and look for one which > is made of all small factors? I should try this and see if I can find one. > Not being an expert in the math, would most coins have a large factor, or > would there be a fair number with only small factors? Knuth has some discussion of this in Seminumerical Algorithms. The term for numbers which have only small factors is that they are "smooth". He has some formulas for what fraction of numbers are smooth based on the size of the largest allowed prime and the size of the numbers. Unfortunately I won't have access to my copy until Tuesday. Perhaps someone else can look it up. > >I don't know how this attack can be prevented. > > I can think of one way. Redefine the coin format so the last 2 bytes or so > can be anything you want. Now when the user generates a coin, he sets these > last two bytes to 0001 and then tests for primality. He keeps adding 2 and > checking until he finds a coin which is prime, or at least doesn't have any > small factors. Clever idea. If only it wouldn't be so slow. > The small-factor sieve is fast, and with the proper #defines, it checks > all primes below 8192 decimal. The slowtest() PGP uses is slow even for the > 512-bit primes used to make 1024 bit PGP keys. It would be useless for a > full 1024-bit number. Would eliminating coins with factors below 8192 be > enough? Or how could one more quickly check the coin for primality? The 8192 cutoff might work. We would have to check it out, but it could be that finding 1024-bit numbers in a relatively narrow range of +/- 2^64 which are composed solely of factors in the range, say, 8192 to 16384 would be infeasible. I don't recall whether Knuth considers the problem in this form. This would be a great save if it works. Hal

1 0

PGPTOOLS and Magic Money
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I've got the code written to check the whole coin, and I found another subtle bug caused by precision setting. Since setting precision does not seem to affect the speed of the decryption (I think the mpi library sets it internally during modexp) I'm just going to fix it at MAX_UNIT_PRECISION and leave it there. Tomorrow I will strip out all of these damn things. Pr0duct Cypher -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLVYNRcGoFIWXVYodAQHdCAP/WZwBMm5NFUzYHaYXhE+d3OAXSlNKpGxD ttHtNJCI1gIZGBc2chDrMxdAa7/3xx+WdAAQ20pM/MLF44S2JVHcxnlum7oSsC9r O04uzdNGprZ1v/K/rZtc8o/xkUAUjctVY0qPGO5hK+Cyl9lABtwBeBPRslUCPYgv A1DjN0E6QNc= =HR0H -----END PGP SIGNATURE-----

2 1

Re: Clipper "Above the Fold"
by dmandl＠lehman.com 17 Dec '03

17 Dec '03

> From: Duncan Frissell <frissell(a)panix.com> > > Clipper and the Admin decision to adopt same is reported in a front page > (above the fold) article in the Saturday New York Times. > > Usual errors about how the "backdoor" would work and about how warrants > would be required to get the keys. > > All the usual suspects. Good placement though. > > DCF Unfortunately, though, it was pretty soft on Clipper. Significantly, the piece was not written by John Markoff, who's been covering cypherpunk- and crypto-related issues for the Times for a while now. Markoff has been very friendly to "our side." This other guy (sorry, name escapes me) seemed to swallow the USG's line much more uncritically. I wonder why Markoff didn't write Saturday's piece? I'm not subtly suggesting conspiracy theories here, though I'm certainly open-minded about them. Mainly, I'm noting the difference between the two guys' approaches and how strongly they affect the coverage. I did a mini- rant about the piece on my radio show Saturday. --Dave.

2 1

TEMPEST - Electronic eavesdropping
by R.O.Jackson-SE1＠computer-science.birmingham.ac.uk 17 Dec '03

17 Dec '03

Transient Electromagnetic Pulse Emanation Standard (TEMPEST) is the US standard defining the amount of electromagnetic radiation that a device may emit without compromising the information it is processing. In the US it not illegal to posess TEMPEST-surveillance equipment but it is illegal to take appropriate counter-measures to prevent surveillance. The US government has refused to release details of its TEMPEST research and has restricted the dissemination of independent research by classifying it. The US Drug Enforcement Agency (DEA) makes use of TEMPEST secured electronics and computers as they believe that the drug cartels may possess surveillance equipment. I am interested in gathering comments on the social, legal, ethical, and technical aspects of use of TEMPEST surveillance equipment in the US and Europe with the aim of including it in a discussion of the threats to computer/digital systems. Please reply by E-mail. I will provide a summary to anybody who requests one. thanks, - Rob Jackson (more information on TEMPEST can be found in the paper "Eavesdropping On the Electromagnetic Emanations of Digital Equipment: The Laws of Canada, England, and the US" by Cristopher Seline - available on FTP from csrc.ncsl.nist.gov)

4 4

Re: A serious question of ethics
by Tom Allard 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- nobody(a)pmantis.berkeley.edu wrote: > Ok, I'm in a bit of a quandry. While surfing the net last week, I > happened across an address addached to a machine that belongs the the > federal reserve. No big deal. I telnetted there on a lark, and entered > 'guest' for the account. It dropped me into a shell. It didn't ask for > a password. Intrigued, I did a little looking around. Nothing special, > a CDRom and about 80 accounts. But(!!), /etc/passwd was there and > available and not using shadows. No, I didn't snatch a copy. - ------- Forwarded Message Date: Mon, 07 Feb 94 11:10:05 -0500 From: m1rcd00 To: m1tca00 Subject: Cypherpunk... Guest login was denied this morning... Well, since someone seems to be home now at Minneapolis, if you wanted to send something back to that list, I suppose it would be OK. If you happened to mention in such a missive that the technical contact here at the Board has no responsibility for or involvement with the Bank machine or network involved, did not fuck up, and was not amused, the technical contact would probably not mind. - - --Bob - ------- End of Forwarded Message -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLVZFT6AudFplx0TNAQGZqgP/f8NOdlitIfBV/pAVTBviJ6IOvBArS42L Ntq1+hiXkUbavx3FOdoQCjiQ7IGPHOsH053nY+7YnwECU/Wyatfle2d0JHVNDyxZ ZX1DIKBT+Pkck9fa1xVkdXp86ZTJofNfbykOou+vNqENanTtDeglU9ytzNTA1/fP 1ptoUYFmoGM= =ppC+ -----END PGP SIGNATURE-----

1 0

Re: DOJ procedures relating to Clipper Chips and key escrow
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

Mike Godwin writes: > One of the interesting passages comes at the end of the DOJ memo > about obtaining Clipper keys pursuant to an interception: > > "These procedures do not create, and are not intended to create, > any substantive rights for individuals intercepted through > electronic surveillance, and noncompliance with these procedures > shall not provide the basis for any motion to suppress or other > objection to the introduction of electronic surveillance evidence > lawfully acquired." > > What this means, apparently, is that keys or communications obtained > through noncompliance with these procedures are nevertheless considered > to be "lawfully acquired." No suppression of evidence. No civil suit. > > In other words, "if we break our rules, tough." I thought that was particularly amusing as well. On the other hand, the mere fact that it says it doesn't mean it invalidates any other privacy laws or rules about illegal surveillance or exclusion of evidence, though it does mean you need to argue a lot harder to get a judge to agree. # Bill Stewart AT&T Global Information Solutions, aka NCR Corp # 6870 Koll Center Parkway, Pleasanton CA, 94566 Phone 1-510-484-6204 # email bill.stewart(a)pleasantonca.ncr.com billstewart(a)attmail.com # ViaCrypt PGP Key IDs 384/C2AFCD 1024/9D6465

1 0

Re: reno_key_escrow.statement (fwd)
by Mike Godwin 17 Dec '03

17 Dec '03

Trotter writes: > Thanks to Mike Godwin for forwarding the announcement about the > Clipper chip stuff. I am not a Constitutional law person or > criminal preceedure person, but if I understand this proposal > correctly, it does not require a member of the judiciary to be > involved. Not at the key-escrow phase, no. But you have to have a valid search warrant or authorization order in hand before you can go to the escrow agencies and request the partial keys. Here's the relevant language: > > ATTORNEY GENERAL MAKES KEY ESCROW ENCRYPTION ANNOUNCEMENTS > > > > When an authorized government agency encounters suspected key- > > escrow encryption, a written request will have to be submitted to > > the two escrow agents. The request will, among other things, have > > to identify the responsible agency and the individuals involved; > > certify that the agency is involved in a lawfully authorized ^^^^^^^^^^^^^^^^^^^^^ > > wiretap; specify the wiretap's source of authorization and its ^^^^^^^^^^^^^^^^^^^^^^^ > > duration; and specify the serial number of the key-escrow > > encryption chip being used. In every case, an attorney involved in > > the investigation will have to provide the escrow agents assurance > > that a validly authorized wiretap is being conducted. The reason that Reno doesn't just say "a court-ordered wiretap" is that there are some emergency circumstances under which wiretap authorization can be gotten in advance of approval by a neutral magistrate. Both the Wiretap Act and the Foreign Intelligence Surveillance Act make provisions for such emergencies. Eventually, such emergency wiretaps do have to be reviewed by a magistrate, however. In the Wiretap Act, and, I believe, in FISA, the time limit is 48 hours. --Mike

2 2

Clipper Qs
by Colin Orion Chandler 17 Dec '03

17 Dec '03

Hurm...I have had a couple of thoughts, no dowbt simple ones, but maybe you can help: If I bought a a ClipperFone and switched chips with my neighbors chips (Clipper Chip, that is), could the .gov tell what was going on? Also, can these chips be re programmed? ;) I'd like a cracker... ___________________________________________________________________________ |---===================================--| /\ | | \ |_ _\ \ / | |---Colin Titus Orion Xavier Chandler----| \\ \ | | . | | > < | |---===================================--| \ \\ / \__/ _|\_|___|_/\_\ | | _____ | / \/ / / | |/\ __ \ __ "What year is it?" | / / \//\ "If it's not a | |\ \ \/\ \ _ __ /\_\ ___ ___ | \//\ / / Sun, it's not a | | \ \ \ \ \/\`'__\/\ \ / __`\ /' _ `\ | / / /\ / computer." | | \ \ \_\ \ \ \/ \ \ \/\ \L\ \/\ \/\ \ | / \\ \ .__ __ | | \ \_____\ \_\ \ \_\ \____/\ \_\ \_\ | \ \\ |_. | | |\ | -| | | \/_____/\/_/ \/_/\/___/ \/_/\/_/ | \/ __| I_| | \| __|/160| +________________________________________+_______________________________+ | Colin Chandler |"It can only be accountable to *human* error."-HAL9000| | (415) 388-8055 | orion(a)crl.com, wizard @ BayMOO (mud.crl.com 8888) | |________________________________________________________________________|

1 0

Re: Nate's Remailer Shutdown.
by qwerty-remailer＠netcom.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Anonymous said, "The reasons the Post Office gets more slack are that 1) They're the government, or at least used to be 2) They can randomly open mail when they feel like it, see 1)" So what's your point? Talk is cheap. The situation remains the same. The message in my post remains valid. The reason why it is so doesn't matter to someone desiring privacy. The internet still sucks and always will, due to the From and Received by e-mail headers as well as many other Unix system problems like sendmail logs, and the fact that you can't trust a wire 'cause you can't see it. -=Xenon=- -----BEGIN PGP SIGNATURE----- Version: 2.3 iQCVAgUBLVbo8wSzG6zrQn1RAQEZqgP+LOHqzsOR+mbHjagehpv12qvihvJl9SSm f1Rz/iVtyKhPVpvsmwhIm3S/F6AmAikQwuO7Kt90BFpS8Q2tfV+iL4mRr1009xKi LovMs+oeydinlH6uOvKGvS4vtaju3dd7+SXQIa0sR46cN8r7O0BiVA6K+9AZ91Cx 6oONCh2Wpfo= =7yq9 -----END PGP SIGNATURE-----

1 0

Magic Money -> Chaum Cash
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Ok, let's try this one more time... Based on Eric's long and mathematical explanation, which I did not fully understand and was therefore convinced by, I have changed the program to use full Chaum cash. It takes the 16-byte random number, takes its MD5, and stores the MD5 in the coin. The coin is now a triple (id,e,mpi) and the bank never sees id when blind-signing the coin, thus preserving anonymity. I sent this new version to csn.org as mgmny10c.zip. I haven't had a chance to update the manual or the comments in the code, but it does seem to work. At least, I was able to mint coins and cycle them through the server a few times, so the basic coin cycle seems to work. Please check it out, on machines of both endians, and let me know what happens. Pr0duct Cypher -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLVdfXcGoFIWXVYodAQExBQQAlAOtfHApmQlmj1bk2kdBEg+Rst0I4CcB vIoxQ/iXiAS5c9fGdl5WNWpBk5TYCQSHm3jyzAoYaeLwJ4XsgnH5WbvB+UeRzwJX VatnTUK7x7wZMtIBAAaaPGX2woosns83bnXMa5voKkiYeESFFLgU5Dw5zw24xFas 1fkwlBSnyRA= =L9Ei -----END PGP SIGNATURE-----

2 1