December 2003 - cypherpunks-legacy

Re: low-quality posts
by Cyber City 17 Dec '03

17 Dec '03

Derek Upham <upham(a)cs.ubc.ca> wrote: > There are two ways to filter for content: filter at the newsgroup > source via moderators, or filter at the newsgroup destinations via > killfiles. Moderating does seem to get rid of most of the cruft, but > the moderators are required to read every post that comes through, > and, worse, make judgements with some degree of impartiality (which is > not always possible). There's a simpler solution. Using the majordomo hack I posted earlier, mail from known abusers would be bounced to the moderator for his approval or rejection. This would be a small volume, which should be within the capabilities of one moderator. It's important to note that even abusive posters have their moments of lucidity. For example, one or two of Detwelier's posts were actually worth reading. So a wholesale and automatic filtering would be wrong. As for the marginal stuff, it should be passed for redistribution, and the end-users should be educated on means to do their own filtering. For example, here's a simple scheme to employ the filter distributed with ELM: .forward file: "|/path/to/filter -vo $HOME/.elm/filter-errors" .elm/filter-rules file: if (from = "lassie") then save "~/mail/nal" if (to contains "cypherpunks") then save "~/mail/Cypher" -- Alex Brock

1 0

The Wisdom of Stuart A. Baker
by perry＠imsi.com 17 Dec '03

17 Dec '03

Says Stewart A. Baker, Chief Counsel for the NSA, writes in Wired: MYTH NUMBER TWO: Unreadable encryption is the key to our future liberty. Of course there are people who aren't prepared to trust the escrow agents, or the courts that issue warrants, or the officials who oversee the system, or anybody else for that matter. Rather than rely on laws to protect us, they say, let's make wiretapping impossible; then we'll be safe no matter who gets elected. This sort of reasoning is the long-delayed revenge of people who couldn't go to Woodstock because they had too much trig homework. It reflects a wide -- and kind of endearing -- streak of romantic high-tech anarchism that crops up throughout the computer world. Don't you just love the finely tuned reasoning here? The absense of ad hominem attacks? This is obviously a rapier sharp logician we have here. I, for one, doubt I could ever produce any counterarguments. We must implement a police state, ladies and gentlemen, because its opponents were more interested in studying than in goofing off at a famous rock festival. I see no possible counterargument. Perry Metzger Who's bothered to read his history books instead of making fun of people who know how to read.

1 0

more from RISKS
by dat＠ebt.com 17 Dec '03

17 Dec '03

More from Risks... Amos Shapir's point below is well taken -- if indeed computer capacity is growing exponentially, which from all accounts it seems to be, then any code can be broken in linear time! ------------------------------ RISKS-LIST: RISKS-FORUM Digest Weds 11 May 1994 Volume 16 : Issue 05 ------------------------------ Date: Tue, 10 May 1994 15:37:05 -0400 From: pcw(a)access.digex.net (Peter Wayner) Subject: Re: Elevators, Car bumpers and Cryptography... I once talked to a major elevator company about doing just what the Schindler Elevator Corp. is accused of doing by the Toronto government. (RISKS-16.04). The company told me that they were in the habit of selling the elevators at a loss so they could make up the money in service contracts. Then they found themselves battling independent service companies who undercut their prices. They hoped to use cryptography to lock out any other service provider without the right key. Of course, this loss-lead approach is common in many businesses. Car companies often sell their cars at a low price and hope to make it up selling spare parts later. That is why I discovered that a spare bumper for my car cost over $500. The difference is that other companies are now making duplicate parts. The major automakers can try and discourage them, but they can't lock them out of the business. Cryptographic locks, though, are a different story. They probably can't be broken in a reasonable amount of time. (See also 16-04) I'm not sure of the case law on this, but I would suspect that it might fall under questionable or illegal trade practices. At least in the US. ------------------------------ Date: Tue, 10 May 94 19:33:36 PDT From: Fredrick B. Cohen <fc(a)Jupiter.SAIC.Com> Subject: Re: Bellcore cracks 129-digit RSA encryption code (RISKS-16.04) I think a lot of people are missing the real point about the RSA. On my pocket PC, I can create a code that requires 5,000 MIP years to break in a matter of seconds. If I am willing to use several more seconds, I can make a code that takes 10^25 MIPS years to break. Compare this to any other encryption scheme, and you will find that the workload amplification of the RSA is quite good. And Shannon told us in 1949 that any non-perfect information transform can be broken with enough cyphertext - and developed the concept of workload for evaluating cryptosystems. If we want perfect cryptosystems we know how to get them, but it requires secure distribution. On the other hand, the RSA provides any degree of complexity we wish to generate (finite) and a fantastic complexity amplification factor, and the advantages of a dual public key system that can be used for both encryption and authentication. The point is that the RSA has not been broken, rather it has shown just how much of a David is required to defeat a given Goliath. After all, in terms of that story, David would have been a MIP second and Goliath 5,000 MIP years in relative sizes for a break-even fight. I'll take that David any day. FC ------------------------------ Date: 11 May 1994 15:19:01 GMT From: amos(a)CS.HUJI.AC.IL (Amos Shapir) Subject: Re: Bellcore cracks 129-digit RSA encryption code (RISKS-16.04) > So where does the 40 quadrillion figure come from? It comes from this very table. 10^9 is a billion, not a trillion, in the US system, and 40 quadrillion is 4 x 10^16, which is even less than what I get by interpolating to 425 bits (can anyone who has access to the original RSA article verify this?). There seems to be an interesting risk here: most encryption methods rely on "hard" problems, i.e. problems whose "brute force" solutions require computation resources which are an exponential function of the key length. But in a world in which computing power grows exponentially, such problems can be solved in polynomial (or even linear) time! Amos Shapir, The Hebrew Univ. of Jerusalem, Dept. of Comp. Science. Givat-Ram, Jerusalem 91904, Israel +972 2 585706,586950 amos(a)cs.huji.ac.il ------------------------------

1 0

Encryption metaphor in Monty Python
by Phil Karn 17 Dec '03

17 Dec '03

Recently I watched Monty Python and the Holy Grail again for the Nth time. At the very end of the movie, as the police arrest King Arthur and Sir Bedevere and break up the attack against the French castle, one of the officers accosts a knight. Grabbing the knight's shield, the officer says "Hey, that's an offensive weapon, that is!" I'd say that pretty much sums up the government's attitude toward cryptography, doesn't it? :-) Phil

1 0

Case law re ITAR and export control
by greg＠ideath.goldenbear.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I've been reading more about export controls, and found these cases which may be of interest: US v Elder, 579 F.2d 516 (9th Cir, 1978). Prohibition on export of technical data includes provision of assistance to foreign manufacturers of items which would be on the Munitions List if manufactured domestically. This prohibition does not interfere with First Amend. free speech protection when narrowly interpreted to apply only to tech. data "significantly and directly related" to items on the Munitions List. Conviction under 22 USCA 1934 (since repealed, but similar to 22 USCA 2778, which applies now) requires that where tech data has both military and nonmilitary uses, prosecution must establish accused knew or had reason to know of its application towards a prohibited purpose. US v Van Hee, 531 F.2d 354 (6th Cir, 1976). Personal knowledge can constitute "technical data" for purposes of regulations restricting export of munitions. Exemption applying to "public domain" technical data applies only to technical data in published form. Exemption for public domain tech data must be claimed at time of export, not at trial, and package/letter must be marked as claimed exempt. US v. Donas-Botto, 363 F.Supp 191 (E.D.Mich, S.D. 1973). "Technical data" as used in the ITAR includes "technical knowledge", and its transmission is not protected by the First Amendment. The US also claims the right to prosecute for violations of the ITAR committed by non-citizens while located in foreign countries, even where the subject matter is owned by a foreign government. See US v Evans, 667 F.Supp 974 (S.D.N.Y, 1987) for a chilling story indeed. I'm going to be on vacation for a few weeks, and probably won't be checking the list; in order to avoid misunderstanding, the summaries above are my own and I believe they're accurate. I haven't found anything which overturns or disagrees with the above. My own ideas about "freedom of speech" don't match the above, but that's probably not surprising. My questions about the above: o When does export take place? When the technical data leaves the country, or when it is transferred to a foreign party? (my guess is the latter, at least with respect to "technical knowledge") o Do I need a license from the State Department if I know how to do RSA and I want to go bar-hopping in Tijuana? (just kidding, but there's a real question in there somewhere.) Discussion on the list about export regulation has focussed on violations of the ITAR; but prosecution is also possible under 22 USC 2778, with maximum penalties of $1M and 10 years in prison. Ouch. If there's going to be an LA-area C-punks meeting, will someone please send E-mail? -----BEGIN PGP SIGNATURE----- Version: 2.5 iQCVAgUBLdKAqX3YhjZY3fMNAQEXWQP9H+WGzXZYki4BXYJI1C4dYQItXHIxAj/9 rKpu5qvnLk3F/cG+vwBB7d6C9g/hRAJQwYSxw1OEI/GG4Es6rqDmpaD7oQeu+mX0 IV/B89gUQuP/YbARLlgH2nTbpxk8gXNQnRDXQlhjJzIzs+yiRGrL9ggTNfNTYh9R AOkTBh7aRTg= =/0G2 -----END PGP SIGNATURE-----

2 1

Yet Another T-Shirt Proposal
by perry＠imsi.com 17 Dec '03

17 Dec '03

Should I ever run in to Stewart A. Baker, I want to be wearing a T-Shirt that says "I oppose Clipper because I'm a bitter guy who couldn't go to Woodstock". (Actually, something pithier would be nice, but you get the idea. Maybe "Mad at the NSA for making me do my Trig Homework"?) Perry

2 1

Re: NSA Cheif Counsel in Wired (Rebuttal)
by Jeff Davis 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- A little further along in the AP story on the record level of Clinton Administration wire taps, Micheal J. Sniffen states: In a section on surveillances completed in 1993, the report said the longest and most expensive federal eavesdropping was accomplished by a microphone placed inside a New Jersey lawyer's office in a racketeering case. The microphone actually operated 435 days, overhearing a total of 65 people, at a cost of $517,673. ...The government said in court, "the purpose of utilizing the law offices ... was to evade electronic surveillance by fraudulently creating the appearance that these were legally proper meetings." This microphone recorded conversations in the office, not the telephone. As I stated to Dr. Dorthy Denning of Georgetown University, escrowed encryption is unnecessary for surveillance. In addition to "bugs", intellegence agencies also provide long range listening technology to the enforcement agencies like the FBI and DEA. Organized criminals don't use the phone to discuss business, it can be tapped. This sort of blows a hole in Stuart Baker's arguments for escrowed encryption being necessary in law enforcement. The next time he offends someone with his tired trig joke, I would hope that he is ask to rebut this. As well as to estimate how many conversations and participants are actually involved in the given figure of 333 1993 "wiretaps." Long range listening was found to fall under federal "wiretap" rules in the Smalldone case in Denver during the summer of 1982. Try to FOIA that info. -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLdKerV/ScHuGXWgVAQFziAQAuYTNTKjaqTWaOO3C42yKCWLM7+kU1gXp 4sGxHGQKfsDP333zLNA+ETGuVfs6si5YQVbsnlGVdS/v36oZp8bUj/8MgWYKLj66 1jRNf4mPl0Mb5LL7InrUwjKCqmOb/GLuHK7F0cHzZbsBE2FkmIqi27AcgJ/8nMxl lFiBbzWrBk4= =I+yV -----END PGP SIGNATURE----- -- PGP PUBLIC KEY via finger! JAFEFFM Speaking & Thinking For Myself! * eagle(a)deeptht.armory.com email info(a)eff.org * *** O U T L A W S On The E L E C T R O N I C F R O N T I E R **** ***** Committed to Free Public Internet Access for World Peace *****

1 0

MIT cypher talk
by Alan Wexelblat 17 Dec '03

17 Dec '03

[email joanne(a)theory.lcs.mit.edu for more info] > Thursday, May 19, 1994 > Refreshments at 4:00pm, Talk at 4:15pm in NE43-518 > > ``A Minimal Model for Secure Computation'' > by Uriel Feige > Weizmann Institute > > ABSTRACT > >We consider a minimal scenario for secure computation: Parties $A$ and >$B$ have private inputs $x$ and $y$ and a shared random string $r$. >$A$ and $B$ are each allowed to send a single message to a third party >$C$, from which $C$ is to learn the value of $f(x,y)$ for some >function $f$, but nothing else. We show that this model is >surprisingly powerful: every function $f$ can be securely computed in >this fashion. If the messages are required to be of polynomial size, >then we exhibit an efficient protocol for any function $f$ computable >in nondeterministic logspace. Using a computational notion of >security, we exhibit efficient protocols for any polynomial-time >computable function $f$, assuming the existence of one-way functions. >The above results generalize to the case where there are more than two >parties with private inputs. > >The minimalistic nature of our model makes it easy to transform >positive results achieved in our model to other more general models of >secure computation. It also gives hope for lower-bound proofs. We >give an alternative characterization of our model in terms of graph >embeddings, and use this to show that for most Boolean functions on >$\{0,1\}^n\times\{0,1\}^n$, the need to hide just one of the input >bits from $C$ requires a communication overhead of $n$ bits. \medskip > >Joint work with Joe Kilian and Moni Naor. > >Host: Michel Goemans

1 0

NSA's Baker to debate key escrow live on AOL, May 26
by Stanton McCandlish 17 Dec '03

17 Dec '03

________ begin fwd ________ NSA'S CHIEF COUNSEL TO APPEAR ONLINE Stewart A. Baker, Chief Counsel for the National Security Agency and featured writer in WIRED 2.06 will host a Q&A session on the Clipper Chip. He will appear on America Online in Center Stage (from AOL type keyword: "center stage") on Thursday May 26, 1994, from 7-9 p.m. EST. Baker is the NSA's top lawyer and supports the Clipper Initiative. He worked briefly as Deputy General Counsel of the Education Department under President Jimmy Carter. His article "Don't Worry Be Happy" refutes seven myths of key escrow encryption and is a WIRED Exclusive. ______ end fwd __________ [NOTE: chances are that's actually EDT not EST.] -- Stanton McCandlish * mech(a)eff.org * Electronic Frontier Found. OnlineActivist "In a Time/CNN poll of 1,000 Americans conducted last week by Yankelovich Partners, two-thirds said it was more important to protect the privacy of phone calls than to preserve the ability of police to conduct wiretaps. When informed about the Clipper Chip, 80% said they opposed it." - Philip Elmer-Dewitt, "Who Should Keep the Keys", TIME, Mar. 14 1994

1 0

Smart cards
by lile＠netcom.com 17 Dec '03

17 Dec '03

I thought you might want to see this. It does talk about the clipper so I hope people will note mind seeing it posted here... I hadn't thought things could get this bad... sigh... -lile ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Lile Elam | "Remember... No matter where you go, there you are." lile(a)netcom.com | Un*x Admin / Artist | Buckaroo Banzai ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

1 0