December 2003 - cypherpunks-legacy

The daily word of advice regarding Tim Mayo
by nobody＠cypherpunks.ca 17 Dec '03

17 Dec '03

What a joy to make a public mockery of Tim Mayo!

2 1

Eudora 3 EMS API stuff
by Dave Del Torto 17 Dec '03

17 Dec '03

Anyone interested in creating plug-ins for Eudora 3.x that provide encryption/remailer/etc support needs this information. dave ................................. cut here ................................. ABOUT THE API The EMS API is one of several complimentary facilities available for integrating 3rd party applications and services with Eudora email. "Plug-ins" interface with Eudora via the EMS API. For Windows the plug-ins are DLL's and on the Macintosh they are Components. Abstractly speaking, the EMS API is most suited for performing transformations, conversions and some other forms of processing on email messages as they are sent and received by Eudora. In practice it is very useful for encryption, digital signatures, and compression. These are all processes that involve encapsulating a message attachments and all. The EMS API is also very useful for text processing and conversions. These might include simple formatting and sorting utilities, or human language translation, and some forms of text-based encryption. EMS API plug-ins can be used to do data format conversions on things ranging from graphic images in attachments to international character sets. Another interesting category are plug-ins that process messages in some way, but don't necessarily change them. For example, a plug-in could be set up so that it scans certain messages for updates to a database like a catalog and automatically makes the change in the local database. Another could be used to scan incoming message attachments for viruses. In a sense, plug-ins can function as filters. DOCUMENTS AND SDK As of the end of July we have documentation and support available at <ftp://ftp.eudora.com/quest/emsapi> as follows: emsapi1.pdf - Adobe Acrobat document describing API version 1 (current on Mac) emsapi2.pdf - Acrobat document describing the API version 2 (current for Windows) chngs2.pdf - Acrobat document showing changes between version 1 & 2 EMS-API-SDK-1.0b3.hqx - Macintosh developer's kit ems1b2.zip - The Windows software developer's kit The developers kit includes a number of support utilities and source code for four sample plug-ins. CURRENT STATUS As of the end of August, Macintosh Eudora Pro 3.0 is shipping and supports the EMS API version 1. The current Macintosh Eudora Lite (the free version) does not support the EMS API, but it will and a beta version is expected in early September if not sooner. Version 1.0b3 of the SDK for the Macintosh is currently available from our FTP site. Windows Eudora 3.0 is in beta test right now and supports EMS API version 2. A copy can be obtained from our ftp site. Version 1.0b2 (a public beta release) of SDK for the API is also now available from our ftp site. MAILING LIST AND MORE INFORMATION If you have specific questions about the API you can write to <mailto:emsapi-info@qualcomm.com>. If you'd like to keep up to date on the API you can join a mailing list we have for announcements and discussions. To join, send a message to <mailto:msapi-dev-request@qualcomm.com>. This mailing list is manually administered and is not a listserv, majordomo or listproc, so no special formatting is needed in the request message.

2 1

Snooping ISP admin??
by Michael E. Carboy 17 Dec '03

17 Dec '03

Greetings All, Question for the group: I have encountered a situation that causes me to believe an ISP is snoopingthrough encrytped mail. It seems that PGP'd mail has aroused the curiosity of an ISP (not hooked.net) I have encountered "POP3 account in use by another user" several times in the past few days and I am the only user... wondering if that "in use" messsage is the result of a clumsy sysadmin being caught with his hand in the cookie jar. Any thoughts from the group??? If those more knowledgeable than I deem these NOISE... my sincere apologies. regards, Michael E. Carboy carboy(a)hooked.net carboy(a)carboy.com

7 6

Re: Public Schools
by Duncan Frissell 17 Dec '03

17 Dec '03

On Sun, 22 Sep 1996, snow wrote: >I would agree that parents can do as good or better at _most_ >subjects thru about the 3rd or 4th grade, and I do agree that >most of todays schools are shit, however there is one area-- >social skills--that homeschooling simply can't compete. >From John Holt's "Teach Your Own" "If there were no other reason to homeschool your children, protecting them from the 'valuable social atmosphere' of the schools would be sufficient." John was a commie liberal BTW. He felt that the schools had a very nasty and artificial social environment with rigid age segregation that bore no resemblance to real life where there are people of wider age ranges. Certainly, most people suffer worse mistreatment from their "peers" at school than they do later in life. As a libertarian, I would add that the social atmosphere of a Stalinist "brain factory" is not exactly the socialization I would choose for my children. I would choose a more market oriented model. DCF "Even if they manage to teach nothing else, the very existence of government schools teaches State supremacy."

7 10

Re: crypto anarchy vs AP
by jim bell 17 Dec '03

17 Dec '03

At 11:26 AM 9/23/96 -0700, Clint Barnett wrote: >>to create the appearance of as free a life style as they can for >>individuals > >ever read "1984"? the appearance of a free lifestyle is most definitely >not a free lifestyle. I am hardly a friend of the state, and far from >being an advocate of the church, but multinational corporations running >the world for their own fun and profit makes my sphincter clench. I think you're under a mis-impression here. At least in the libertarian circles where I do most of my political discussion, it is generally suspected that corporations grow large and powerful primarily BECAUSE OF assistance by and due to the policies of government. (heavy regulation favors large companies by keeping down small competitors, etc) Of course, that cause/effect relationship is intentionally hidden, and most of the public sees the show put on which appears to have the opposite effect: Anti-trust lawsuits, etc. Over time, the public is gulled into the false belief that if you get rid of the government, you get rid of the "only think that stands between us and the multi-national corporations." They believe this because the only government actions they see and recognize are anti-corporation. If they were aware of the truth, they'd realize that these large corporations are actually afraid of a free market, and that the companies consider big government to be their friends.. Just look at an excellent example: Intel versus IBM. Intel used to be this tiny upstart chip company from the Bay area and IBM was smokestack America. Now, Intel is the biggest (by dollar volume, anyway) chip company in the world and IBM is, well, considerably cut back from its heyday. At least in hindsight, IBM would have been "smart" to squash Intel, or buy it up, or have the government over-regulate it. Jim Bell jimbell(a)pacifier.com

2 1

Transforming variable-length to fixed keys
by pgut001＠cs.auckland.ac.nz 17 Dec '03

17 Dec '03

I posted this to sci.crypt recently but the response to it was rather underwhelming, so I thought I'd repost it here to see if anyone has any comments on it. What it is is a scheme for transforming arbitrary user keys (typically a long passphrase) into a fixed-length key for a particular algorithm. This has the following properties: 1. The user key 'userKey' is transformed into an algorithm-specific key 'key' using a number of 'iterations' of a hash algorithm 'hash()'. 2. The transformation is strongly serialized so that any form of attack involving parallelization or precomputation isn't possible. 3. The transformation is non-reversible, so that recovering the transformed key won't recover the original key. 4. The result of the transformation is algorithm-dependant, so that if an attacker recovers a transformed key for one algorithm they can't recover the transformed key (from the same user key) for another algorithm. 5. The transformation can be iterated as often as required to make password-guessing attacks difficult. 6. The transformation process is algorithm-independant and can use any type of hash algorithm and original and transformed key size. The transformation algorithm (which was designed with the help of John Kelsey) is as follows: key[] = { 0 }; state = hash( algorithm, mode, parameters, userKey ); for count = 1 to iterations for length = 1 to keyLength (in hash_output_size blocks) state = hash( state ); key[ length ] = hash( state, userKey ); The state acts as an RNG which ensures that the key hashing is serialized. The initial state depends on all encryption parameters, not just the user key. If we hashed the user key directly and then used it for a number of algorithms then someone who could recover the transformed key for one algorithm could compromise it if used for other algorithms (for example recovering a DES key would also recover half an IDEA key). Hashing all algorithm-related parameters means that a successful attack one an algorithm, mode, or configuration won't allow the key for any other algorithm, mode, or configuration to be recovered. The code which implements the iterated hashing is: /* Hash the variable-length input to a fixed-length output */ memset( key, 0, keyLength ); for( count = 0; count < iterations; count++ ) { for( keyIndex = 0; keyIndex < keyLength; keyIndex += hashOutputSize ) { /* state = hash( state ); key[ n ] = hash( state, userKey ) */ hash( state, state, hashOutputSize, HASH_ALL ); hash( NULL, state, hashOutputSize, HASH_START ); hash( temp, userKey, userKeyLength, HASH_END ); | | | output input input size /* Copy as much of the hashed data as required to the output */ length = ( keyLength - keyIndex ) % hashOutputSize; for( i = 0; i < length; i++ ) key[ i ] ^= temp[ i ]; } } Peter.

2 1

Re: AP [was: Re: Kiddie porn on the Internet] [NOISE]
by jim bell 17 Dec '03

17 Dec '03

At 12:38 PM 9/22/96 -0400, hallam(a)vesuvius.ai.mit.edu wrote: > >I agree with Bill, AP sounds like a very good club for the Exon types >to beat us with. But they're not doing it. They know about it, but they're NOT using it. This should tell you something. This really is no surprise to me; before I started publicizing AP, I tried to carefully consider whether letting the cat out of the bag was going to assist the government in stopping it. My conclusion was that they didn't stand a chance. If anything, they're afraid of publicizing the idea, because their opposition to it can look only self-serving. >Jim's latest comments sounded no different to me than the mealy mouthed >"understanding" messages that Sinn Fein publish about the IRA. Given >the opportunity to condem any attack on the President under AP rules >we get a non commital non-condemnation. Its the type of thing that Gerry >Adams says after his men have just killed two kids with a bomb in a litterbin >outside a MacDonalds. Don't try to tar me with that brush. I made it clear long ago that I'm opposed to truly random attacks on ordinary citizens. However, I hasten to add that I'd very much prefer that these attacks be re-targeted against the people who deserve them. However, I usually find that the people who make the biggest deal about disliking random citizen attacks are the ones who recommend replacing them with...nothing...and have no plan to solve the problems with any other method, violent or otherwise. [doctrinaire BS deleted] >As the Marxist idealogues got frustrated by their evident lack of progress >they turned to terrorism. The Bader Minehof gang believed that they could >spark the revolution by jolting society out of its complacency. Their >strategy was remarkably like AP. Somehow, I doubt it. Anybody who thinks an AP-like system would be effective for bringing in his pet communist (or fascist, or monarchist, or anything other than pure-anarchic) government is an idiot. AP is fundamentally anti-heirarchical in its leanings and effects. >If the heads of large corporations were >likely to be assasinated then noone would want to lead a large firm. Who said companies need "leaders"? Remember, I said AP is anti-heirarchical. Any heirarchies, particularly those forced on people, are strongly deterred. This includes social and religious heirarchies, BTW. Even corporate heirarchies would only exist if approved-of by the vast majority of the shareholders, most of whom will be the workers as well. > In >fact as any person with counter terrorism experience will tell you the >threat of death is remarkably ineffective as a means of intimidation. It >creates the opposite effect, strengthening the resolve of the target. I >discussed this point recently with someone close to Mossad who agreed. Under a set of circumstances similar to the status quo, that is not surprising. But that can change. That WILL change. People occupy positions of authority when they feel, personally, they are better off doing so than not. Give the average member of the public (not merely crazed "terrorists") the opportunity to remove those people, and you won't be able to keep them around. I mean, presumably they do their job for a salary and benefits, right? How many bombs or bullets will they tolerate, rather than resigning and taking a safer job? >Terrorism is becomming an increasing concern. The amount of damage an >individual can cause is much greater than that possible in the past. If anything, I think the advantage of AP is that it REDUCES the amount of damage that any given person has to cause, per citizen, to achieve his anti-government political ends. I've pointed out that it would only take a penny or two from each citizen, on average, to remove one of even the highest-level US officeholders. Far less than that if you include resignations. What I'm advocating is a system that makes it impossible for agents of the status quo to resist the opposition of the public, which is unlike the current system. Rather than have to build a bomb, or wait for somebody else to do so, the ordinary citizen only needs to make a modest contribution to the "revolution fund." And today, an officeholder can claim to resist "terrorism" based on the fact that he opposes the extreme action of a handful of people. What happens when they have to admit that a million people paid for that gun to be fired or that bomb to be planted? >There >are plenty of exhausted ideologies about which can be fashioned into >a justification of murder. Don't blame me for them. > Jim's post shows very clearly how Libertarianism >can be converted into a justification for terrorism. Its a very short gap >between being opposed to government and actively fighting against it. What, exactly, is the MEANING of "being opposed to government"? If you pay taxes to support that government (even if that support is not voluntary) those taxes mean that you are indeed acting to strengthen that government and assist its efforts. I suggest that you can't realistically claim "opposition to government" (other than rhetorical) under these circumstances. >I see libertarianism as the exhausted remnant of the mercantilism of the >1980s. As constructed it recognises only those rights which favour the >privileged in society and none of those which benefit the ecconomically >disadvantaged. Politically it reached its peak influence almost a decade >ago when Regan and Thatcher were at their zenith. The '94 congress will >probably be seen as the turning point in the political tide with the mainstream >of politics moving back to the left again. There will always be people arround >who conclude that the failure was not being close enough to the ideology. You've already made it clear you don't consider yourself a libertarian. Why go on with this stuff? >If the libertarians are not carefull they will be inexorably linked in the public mind >with the terrorists who act in their name. I think the stereotypical "terrorist" doesn't really claim to "act in their [the public's] name." The one recent counter-example that I've heard of, that of the EPR in Mexico, is a very welcome exception. If anything, I think the governments of the world are truly frightened that "terrorism" as commonly practiced will change from the "blow up random airliner" mode, passing through the "blow up nearby government building" mode, to "find and kill a particular official" mode. Far less collateral damage, far harder to whip up public opposition to in the press, etc. Jim Bell jimbell(a)pacifier.com

7 7

Re: provably hard PK cryptosystems
by tcmay＠got.net 17 Dec '03

17 Dec '03

I want to say something about "tiling problems," an area I find very exciting. I ordinarily would not have commented on "quantum computers," but will make just a couple of comments, since I want to comment later on tilings. At 9:58 PM 9/22/96, Adam Back wrote: ... >I'm not sure about quantum computers, some people who know much more >about particle physics than I do seemed initially sceptical, and >didn't think it was doable. However I have read some optimistic >sounding news clippings (on the list) which sounded as if things are >progressing well, with techniques being found using redundancy to get >around what were earlier problems of reliability. Is this accurate >reporting (thinking of garbled stories by over enthusiastic >journalists)? I'd be interested to hear opinions from anyone who does >know about particle physics about the likihood of practical quantum >computers being practical in the next 20 years or so. Caveat: I'm a skeptic on quantum computers. I've read a couple of the early papers, but am not current. And I certainly am not an expert. For what it's worth: * I think it's nearly certain that no significant results will be gotten in the next 20 years, probably not in the next 50 years. (I personally am skeptical that significant results will be gotten in 200 years, and probably never, but this is a harder point to make.) * Sure, there may be demonstrations of a "collapsed quantum state" which is isomorphic to factoring a number like "42," but not 50-digit numbers anytime soon. And, I feel, a 300-digit number is probably many, many orders of magnitude harder. * Experts may point to partial successes, and maybe they're right. But I'm skeptical. The recent post by someone outlining a bunch of "promising approaches" has got me thinking of spending some more time looking at recent results, though. I doubt anything will happen in the next 20 years...that just isn't much time in the high-tech business (which may sound crazy, but it's true). (A way to "creatively visualize" this point is this: for those of you who are now 25, when you are 45 the world will probably _not_ be radically different. Sure, computers will be faster and cheaper, but most things will look _mostly_ the way they look today. Not a compelling argument, perhaps, but one which makes sense to me. When I was 25 I suppose I expected dramatic advances to come...largely, they haven't. I can elaborate on this if there's sufficient interest.) >One other area that did sound promising was some kind of mapping >problem in n dimensional tiling that Tim was discussing at a physical >meet while I was over in the US. A news clipping posted to One of the most interesting books I've read is David Harel's "Algorithmics: The Spirit of Computing." (Warning: The book came out in a second edition, which as near as I can tell took out some parts and synopsized the book a bit. My reading was of the First Edition.) Harel described Huang's work on "tiling problems." These are easier to describe with pictures, and I don't have the time to try to generate ASCII representations of tiles. So, I refer people to Harel's book. Briefly, imagine a grid in a plane. Imagine a set of "dominoes" or "tiles" with different edge properties, e.g., some edges are blue, some red, some green, etc. (or the edges can be numbered, have symbols on them, etc.). Suppose one has an unlimited supply of, for example, N different type of tiles. Suppose a tile is placed at some place on the grid, and another tile (possibly a different tile, possibly the same type of tile) is placed some distance away on the grid. The problem is this: Can a "domino snake" be found which reaches from the first tile to the second tile, with the constraint that edges must match up on all tiles? (And all tiles must be in normal grid locations, of course) If the grid area is, say, a finite space, then clearly an exhaustive search of all legal domino snakes would answer the question. (However, even a relatively small grid of, say, 8 x 8, generates a truly enormous number of possible snakes, each of which must be tested. Of course, if a domino snake is "guessed" (a la the "nondeterministic" language one encounters in complexity theory), verification that it is answer is nearly instantaneous (one glances at the solution and verifies it). If the grid is unlimited, then even if the two initial tiles are located close together, there are of course an infinite number of possible snakes.... (Oddly, to me, the domino snake problem is "more undecidable" in the infinite half-plane that in the full infinite plane....) Huang proved that the domino tiling problem is formally equivalent to the standard set of NP problems (consult Harel, or Huang, for more details and more precise language...I'm just going from recollection here!). I like this approach because I can easily visualize the domino tiling problem and can make some points about expansion of knowledge into uncharted areas (research and tool-building is a kind of domino snake expansion, with local laws of physics, etc., the constraints on stepping stones (hand wave inserted here)). I find this easier to understand than more "algebraic" and "logical" versions, such as the representation problem and the word problem. (These other NP problems are also described in Harel (and in standard complexity books, such as Garey and Johnson, Papadimitrou, etc.). It has always been a great hope that some of the provably-hard problems be adapted for use in crypto. And yet factoring remains the core of popular modern crypto programs (discrete logs, too, for D-H). I don't think this is a major cause for worry, though. It's not as if factoring is suddenly going to become "easy." (I think the consensus is that factoring _is_ hard. Whether RSA is really equivalent to factoring also hasn't been proved, as folks have noted, but many think it is, loosely speaking.) --Tim May We got computers, we're tapping phone lines, I know that that ain't allowed. ---------:---------:---------:---------:---------:---------:---------:---- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Higher Power: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

6 5

possible solution to cyber S/N
by Vladimir Z. Nuri 17 Dec '03

17 Dec '03

I've written many times on the problem of signal to noise here in cyberspace and have tried to think through some elegant solutions to this very vexing problem. moments ago I just came up with an intense brainstorm fully worthy of sharing with the list <g>. it's clear that the web functions very much like a directed graph in many ways. sites and traffic between them are one graph. however, another graph involves thinking of each page or "article" on the web as a single node in the graph, and hyperlinks as the edges between the nodes. what we have today is a directed graph in which one can follow the forward direction of hyperlinks very readily. however, it is much more difficult to follow the reverse direction, i.e. find "all the sites that reference this paper". of course this is part of the beauty of the web, that anyone can link to anyone without actually having to register or something like that. the "referrer log" feature on a server is a mechanism that does allow a server to get some idea of who is linked up to that site. the basic idea I have is that references are an excellent way of discriminating Signal to Noise and are used routinely in the scientific arena. a paper that is pivotal and influential is referred to ad infinitum. obscure papers are forgotten and never referred to in subsequent literature. taking this idea to the cyberspace arena, the application is immediately obvious-- pages that are linked to by a lot of other pages are valuable, those that are not are not as valuable. (another closely related idea is how much hit traffic a web site gets-- wouldn't it be a *tremendous* improvement in the current search engines if they returned the pages ranked according to how many hits they get per time?) there are some problems with all the above, however. currently only the site that actually houses the pages can keep track of hits, and referrers are not very well kept track of at all. in a robust system, cheating, such as reporting more hits than one is getting, would not be possible. anyway, I tend to think that future S/N problems in cyberspace are increasingly going to be solved in some particular ways that are just now being tried out: 1. rating agencies. agencies that both reject and find "cool stuff" (like Point Communications etc.) 2. hit statistics. how many people are hitting various pages? if we could get an arbitron-like system that works the same way that newsgroup readerships are now reported (each site compiles statistics and sends them to centralized databases) we could have search engines that rank results according to hit statistics. I'm not saying it would be perfect, and there are all kinds of obvious nitpicky things that people here will harp on, but I still insist it would be better than no statistics at all. eventually a system like this that involved voluntary submissions of hit counts to the centralized servers (or just some way of search engines to get hit statistics from servers about the pages they own/serve) may evolve into a more robust system that makes cheating impossible (such as falsely reporting a high hit count to attract people to the site). 3. linking statistics. again, I think this is an extremely powerful way of separating signal from noise-- how many other sites are linked to the page in question? if it has few links, it isn't as interesting, if a lot of other people point to it, it's far more interesting. I encourage system designers to keep some of these ideas in mind when they are working on the current generation of software tools such as search engines etc. in particular there is a tremendous amount of innovation going on right now between search engine designers, and adopting some of the above ideas into a search engine might be a very powerful way of distinguishing it from the "competition" by returning more useable search results to the end user. again, I suspect we will be seeing increasingly ingenious and efficacious ways of dealing with what today is the horrible S/N problem. perhaps today will be thought of as the dark age of cyberspace because of all the muck we are routinely wading through <g>

4 3

Public Schools
by tcmay＠got.net 17 Dec '03

17 Dec '03

At 4:31 AM 9/23/96, Lucky Green wrote: >On Sun, 22 Sep 1996, snow wrote: >> I would agree that parents can do as good or better at _most_ subjects >> thru about the 3rd or 4th grade, and I do agree that most of todays schools >> are shit, however there is one area--social skills--that homeschooling >> simply can't compete. Children need to learn how to interact with one another >> in groups larger than a family unit. I don't think that homeschooling can >> accomplish this nearly as well as the public (or private) schools could. > >I understand that many parents that homeschool belong to organizations >that provide for meetings twice a week in which the children so educated >in a certain area get together. Homeschooling does not have to stand in >the way of a normal socialization process. Also--and I mean this point completely seriously!--many parents are not altogether convinced that the "public school socialization" is all that beneficial. Do kids really _need_ to learn to wear gang colors, smoke to be cool, get pregnant at age 14, and so on? An awful lot of accomplished persons grew up in isolated areas, on small farms, and were educated in very small classrooms (or at home). I think the "social skills" Snow talks about above are actually the _worst_ part of public schools in fin-de-Siecle America. If I had a kid, I wouldn't want him or her in the local public schools. (And my area is more remote than over in San Jose--San Francisco, which is even worse. And LA is even worse.) --Tim May We got computers, we're tapping phone lines, I know that that ain't allowed. ---------:---------:---------:---------:---------:---------:---------:---- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay(a)got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Higher Power: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

5 5