December 2003 - cypherpunks-legacy

New Money design announced
by nym＠netcom.com 17 Dec '03

17 Dec '03

I think this came from a Fidonet channel. an interesting topic: how will this affect the value of US cash? i.e. will it affect the value of the dollar worldwide? often when countries are facing economic trouble they change their currency. **** Conf. : 0008 - FINANCIAL WASHINGTON (AP) -- Move over Ben Franklin, tell Andrew Jackson the news: The portraits on the nation's paper money likely will be enlarged and moved off-center as part of the first major makeover in 65 years. ``Our plan ... is a pre-emptive step to protect U.S. currency from high-tech counterfeiting,'' Treasury Secretary Lloyd Bentsen said in announcing the redesign Wednesday. His department plans to have a final design ready sometime in 1995 and begin circulating new bills about a year later, starting with the most popular target for counterfeiters, $100 notes. Nothing has been decided for certain, but Treasury officials gave the House Banking Committee a rundown Wednesday of what was likely. The enlarged portraits -- Franklin on the $100, Ulysses Grant on the $50, Jackson on the $20, Alexander Hamilton on the $10, Abraham Lincoln on the $5 and George Washington on the $1 -- will allow for more detailed engraving of what is the most recognizable feature on a bill. The $2 bill, with Thomas Jefferson, is not being redesigned. And, moving the portraits will make room for a watermark in the form of a smaller version of the portrait, visible only when a bill is held to the light. Other likely changes include: -- Color-shifting ink that may, for instance, appear green when viewed straight on and gold from an angle. -- Computer-designed ``interactive'' patterns that turn wavy when illicitly copied. -- Iridescent planchettes in bills' paper. These are colored discs only a few millimeters wide that reflect light. -- Micro-printing and machine-detectable threads or fibers in the paper. The last major change in U.S. currency came in 1929, when bills were reduced in size and given a uniform look. Congress added the words ``In God We Trust'' in 1957. In 1990, several small changes made it more difficult to counterfeit with modern color photocopiers. They included adding a microscopic line of type and a polyester thread visible only when a bill is held to a light. Most lawmakers applauded the latest modifications, saying they were long overdue. ``We must protect the dollar's standing as the world's foremost currency. Our financial future depends upon it,'' said the committee chairman, Rep. Henry B. Gonzalez, D-Texas. But others warned against too radical a redesign, saying it would send a message of instability at a time when the value of the dollar is being battered on world currency markets. Peter H. Daly, director of the Bureau of Engraving and Printing, said the bills' traditional look will be maintained as much as possible. The paper will feel the same and the size of notes won't change. And the same engraving style will be used for portraits, borders, numbers and the historic scenes on bills' back. Bentsen stressed that old money will continue as legal tender. ``The redesigned currency will be introduced over a period of years and no U.S. currency will be demonetized, devalued or recalled,'' he said. And U.S. Treasurer Mary Ellen Withrow promised a worldwide educational campaign to encourage public acceptance of the new design. About $357 billion of U.S. paper currency is in circulation, more than 60 percent of it abroad. In fact, in some countries with troubled economies it serves as the principal medium of exchange. According to the Secret Service, counterfeiting abroad now far outweighs domestic counterfeiting. Last year, $120 million in phony bills were seized overseas, compared with $44 million domestically, said Secret Service Deputy Director Guy P. Caputo. Republican Reps. Jim Leach of Iowa and Bill McCollum of Florida suggested the administration may be minimizing the severity of the problem. Leach said a Middle Eastern country, which he did not name, reportedly has developed the capacity to counterfeit large quantities of $100 notes. He was apparently referring to trade press reports that Iran was distributing counterfeit bills through Lebanon. But Bentsen stressed that the changes were being made in anticipation of future problems and not in response to any current counterfeiting crisis.

2 1

Re: Hashed hash (and Kent's games)
by Ben.Goren＠asu.edu 17 Dec '03

17 Dec '03

At 12:19 AM 7/17/94, Kent Borg wrote: >Ben.Goren(a)asu.edu foolishly says: >>I'm planning on implementing the "cryptographic protection of databases" > >And wonders about the hash being too fast to compute, that a >brute-force traversal of the database would be too easy. The idea is >then to hash a bunch of times to burn CPU cycles, but what if the hash >is a group, extra hashing could be reversed quickly. (Did I get that >right?) On the nose. >Well, as the LOUD proponent of making secret keys s-l-o-w-e-r to >decrypt, I have thought about this a bit, and have a suggestion: > >Hash once, then do a zillion encryptions of the hash with a non-group >cypher like DES. I'll probably do just that. First thought, subject to revision: hash the name, feed it to DES with the output of a deterministic RNG (need not be secure, but the slower the better--BBS? (not that BBS is incesure)) as the key; repeat as needed. Hmmm...perhaps I'll adapt an earlier idea of mine: split the hash into two parts, a and b, and compute (a^(1/b))-1, and use some or all bits after the leading zeros. >Another idea (something I have thought less about): send every legit >user of the database a custom version with the parts encrypted with >that user's public key--and do the trick mailing list companies use, >scatter some dummy info in the list. When a dummy (not just me) gets >a junk mailing, go beat up on the user who's copy had to have supplied >the junk [. . . .] Nice idea, but there's neither the available resources to do that, nor, I think, the desire to beat up on careless users. Berzerk suggests a 0.1 S/N ratio (and in an earlier note a couple useable algorithms for the multiple encryption process); that would not be practical for any decent sized database, and I might have 100K or so people to deal with. But I almost certainly will mix in at least some random padding. I imagine that the database will always be the same length, even as people are added and/or removed with time. And the records, of course, will be premuted randomly. >-kb, the Kent who doesn't want to be thought of as only a card player Then here's a suggestion for you: develop some other primitives, like rolling dice, and you could implement just about any other game you like. Monopoly would need (aside from licensing issues) the dice, two decks of special cards, and some ecash. (Surely MM used as Monopoly Money isn't subversive? After all, it's teaching our young 'uns to be good capitalists.) Scrabble would need a deck of cards, each of which contains only a letter, with many duplicates. Trivial Pursiut is just a huge deck of cards; they'd probably be index positions to the database of questions, so special editions are just a file switch away. These are among the most popular games in the US, and probalby abroad. Build your primitives right, and these games are as simple as specifing paramaters (how many sides to the dice, what info the cards contain, etc.). And maybe you could license the stuff, each and every independent game, to the current owners of the games that aren't PD. So how about becomming "kb, the Kent who digitized the American family evening"? Go for it! And drop me a line when you want beta testers (sometime Thursday?). >Kent Borg +1 (617) 776-6899 b& -- Ben.Goren(a)asu.edu, Arizona State University School of Music net.proselytizing (write for info): Protect your privacy; oppose Clipper. Voice concern over proposed Internet pricing schemes. Stamp out spamming. Finger ben(a)tux.music.asu.edu for PGP 2.3a public key.

2 2

PGP bug *NOT* yet fixed
by Ben.Goren＠asu.edu 17 Dec '03

17 Dec '03

Y'all remember that bug that Colin Plumb told us about in the true random number generation part of PGP? It's still there, in the version from net-dist.mit.edu, as of late yesterday evening. There is *no* mention of the bug anywhere, in any readme files, in the documentation, anything. This strikes me as irresponsible. I would expect PGP 2.6a to have been released the day of the announcement, with the bug fixed. If there's some reason why that couldn't be done, then at least there should be some sort of prominent notice detailing the change, and probably a diff file--or even a simple shell script--to apply the (very simple!) fix. The signature on the following file checks as follows: >File has signature. Public key is required to check signature. . >Good signature from user "Colin Plumb <colin(a)nyx.cs.du.edu>". >Signature made 1994/06/01 14:04 GMT That's a month and a half ago. The *only* copies of PGP 2.6 out there that are free of the bug are those that have been fixed by hand. That's probably not all that many of them. I'm going to bite the bullet and paste in the original message here. Feel free to flame me if this is unnecessary re/cross-posting, but I'm not aware of any place to get this aside from digging through archives, and.... b& -----BEGIN PGP PUBLIC KEY BLOCK----- Version: 2.5 mQCNAi3L864AAAEEAKRe8j9QUqL4PDQSsliTKQ0yTkdLL8BFBm7c03RC9Ol5PP9K j/RtnsdxFMTtW7wkMwTpY1jF23HR+x54LrOpi8ig6HEmiXVVWuNByRjSMgz8jvrn MM0/tIOCPAgNMxiANUWqretPEWCZE9sLbylkJrrOd54ZKyXBTw/D7AL7u4qxAAUR tCFDb2xpbiBQbHVtYiA8Y29saW5Abnl4LmNzLmR1LmVkdT6JAJUCBRAtyxCUZXmE uMepZt0BAeiyA/4tNXz6loqEwyMv65TMGtqxTlT5ocGNzyE8mkZXvbmoS0m7sdsd aVBvHfK8lrkQz/anrzAHJMBOaZ0V6T7aCLAK6GnjHoeanP8ZyhaXpc2e7EVut4Zi hCpmq45uiA/1diwLXhC8OoHwKqZDT+uNnJLLdlAzrJiOaELAzXXeOvtMXokAYAIF EC3L/BnKPaH9hlqn8wEBXWgCWMgIh8Lsww5pFHRFbAe2HehjGIiOmQ+ZcnL3pOhw tLdoGm6lqWZ4njDSTULxDpKUtbe4pWNv6Go13t9p+1GmTh+RrnGoq6rs3Mlg+IkA lQIFEC3L+zgPw+wC+7uKsQEBDZkEAJYkHK5n02GXLwEEgFKpxQvWLqI2xz33rPDa 0eT6+RYMDcr/1vzTqX7CwNpCuTaFTVNRbRznvwNTDcQXVsnyPg5yGdRIIMPnWuGf gSEP7vjm8zzvfdh5te4ag6jobCN1PVyqIIxIV5S8iPv632gm4vQboJiQ+4+53qoS WJ6BNDq9 =Wjfi -----END PGP PUBLIC KEY BLOCK----- -----BEGIN PGP SIGNED MESSAGE----- I have the unpleasant task of reporting a significant bug in PGP's random number generation (for making primes), and that it's my fault. It *is* a significant problem, although it is *not* end-of-the-world severity. That is, the code is not doing performing as intended, and the results aren't as random as intended. On the other hand, this does not appear to make any generated keys easier to break. Because it has to do with random-number generation, there are no interoperability issues raised. Please read on for details. Thanks to the many people who have submitted other bug reports and porting patches. A new release from MIT is forthcoming with more cleanups. * The Bug In pgp 2.6 (and 2.5), there is a file named "randpool.c", which accumulates entropy from keyboard timings. These random numbers are used in generating session keys, although the primary random number generator for session keys, based on IDEA, is unaffected. The main use of these random numbers is the much more sensitive task of generating RSA secret keys. In that file, a tiny helper function is xorbytes: static void xorbytes(byte *dest, byte const *src, unsigned len) { while (len--) *dest++ = *src++; } A character is missing. '^', to be precise. That "=" should be "^=". I wrote it, and I knew when I was writing it that it was critical code. Since you can't test a random-number generator (except for the most trivial of flaws), you have to walk through the code very carefully. I did, or thought I did, yet still managed to miss this. Oops is too mild. That code is not supposed to have ANY bugs. In other words, I screwed up. There's a lesson in there somewhere. I'll try to learn it. * The Effect The randpool.c code works by maintaining a pool (buffer) of random bits and adding in new "noise" from the environment each time a key is pressed. This "adding" is done by exclusive-oring it with successive bytes from the existing pool. When the pool is "full", a cryptographic stirring operation is performed to mix all the information in the pool together and get ready for new noise. The bytes in the pool at the end are intended to be uncorrelated with the noise bytes that will be added, so the XOR adding does not cause any sort of "cancellation" of information. This stirring is done with a key, which is taken from the pool at the end of each pass. With the bug in place, the noise bytes *replace* the bytes in the pool rather than being added to them. So the information that was in the pool is obliterated. The only trace that remains is what's stored in the key. This is at most the size of the key, 512 bits, rather than the size of the whole pool, 3072 bits. PGP tries to ensure that generated RSA keys are completely unpredictable by accumulating enough Shannon information to make the whole key. Thus, infinite computational power would not let you predict a generated secret RSA key. This bug subverts that. * Security Analysis What effect does this have on someone's chances of breaking an RSA secret key generated with PGP 2.6? Not much, as far as I can tell. But it requires more careful thought and that eats into the comfort margin that should be there. Just for comparison, the RSAREF library's random number generation routines are also based on MD5, but use 16 bytes of seed. Successive random bytes are taken by computing the MD5 hash of the 16-byte seed, using those 16 bytes, incrementing the seed by 1 (taken as a 128-bit number), and repeating. Taking the MD5 of a 16-byte value involves one pass of the MD5Transform function, with 16 of the 64 key bytes unknown, 48 bytes are known (fixed, in fact), and the input hash is known (fixed, in fact). Compared to this, PGP 2.6, even with the bug, is excellent. All 64 bytes of key to MD5Transform are dependent on all of the seed, the input hash varies widely, and the output is XORed with some difficult-to-predict data. The reason that you can get away with less than perfect random numbers (less Shannon information than the size of the generated key) is that you only have to make sure that the weakness does not make any attack easier than the best known attack without the weakness. As long as guessing is only useful to a brute-force attack, it remains far easier to factor. Paul Leyland estimated that the work to try all possible 128-bit IDEA keys is equivalent to factoring a 3100-bit RSA key. Now, recent work by Arjen Lenstra on the number field sieve (Paul Leyland was assuming the MPQS used in RSA-129) has raised this RSA key length somewhat. Thus, an argument can be made in favour of RSAREF's use of a 128-bit random number seed, since that's all that is necessary. PGP prefers to be a little bit more paranoid. Still, once you have 512 bits of uncertainty, trying all possibilities is more work than trying to break a 1024-bit RSA key by trial division. So let's see just how much entropy is in there. Each keystroke, the following data is added to the random pool: - - The cahracter typed, an int (2 or 4 bytes) - - the time_t result of time() (4 bytes) - - the clock_t result of clock() (4 bytes) - - On MS-DOS, 2 bytes of hardware timer 0 - - On Unix, 8 bytes of gettimeofday() and 20 bytes of times() results - - On VMS, 8 bytes of high-resolution timer. The total is 12 bytes on MS-DOS, 32 bytes on Unix (this may vary, but that's very common), and 20 bytes on VMS. The information content of the bytes is taken at a maximum of 8 bits, although it's actually closer to 15 bits on MS-DOS, and less (maybe as low as 1 or 2) on a Unix system with a fast typist and a slow (60 Hz) clock. VMS is in between. This means that the entropy density in the added bytes varies from 1/12 (or better) in MS-DOS to 1/256 on Unix. Thus, the content of a pool's worth (3072 bits) is 256 bits (or more) under MS-DOS and may be as low as 12 bits on some flavours of Unix. The random number accumulation operation adds bytes to the pool until it is either full or the desired number of bits have been accumulated. Then it stors the pool. For a maximum-sized key (1024 bits), it will take many passes through the pool to accumulate the entropy, but owing to the bug, each time the pool is overwritten with the most recently collected data. The only entropy that remains from the previous pass is in the 512-bit key buffer. This applies to every stirring pass until the last, after the last noise data has been added and new data is about to be withdrawn from the pool. This last pass is very likely to be incomplete; some of the data at the tail of the pool is probably not overwritten. This can carry over extra entropy from the previous pass. No more than is there (the 12 to 256 bit range observed before), and then you have to add an unknown fraction of that for data that has been added in the current pass, but the total will vary from 12 bits (an average of 18) to 256 bits (an average of 384). Plus the entropy preserved in the key buffer. So there is from just over 512 to an average of 896 bits of entropy in the pool. 1016 random bits are used to make the starting values for the two primes in a 1024-bit key. This is clearly not the perfect Shannon entropy PGP aims for. As long as the stirring operation is still considered cryptographically strong, this reduction in the possible range of generated keys is not useful to a factoring algorithm, so it doesn't make a factoring attack any easier, yet a factoring attack is still far easier than a guessing attack, so the easiest attack is no easier. So I don't think anything is more attackable. Still, it's NOT what was intended, and that's always bad. My apologies to users of PGP. - -- -Colin -----BEGIN PGP SIGNATURE----- Version: 2.5 iQCVAgUBLeyVSw/D7AL7u4qxAQEjCQP/YlzY5DWT4FrSErQ8W0TP9ibRqpck4gKL YOkUgiMQnvCE2XHEvP1VTfUANgU9O/P7lClJ1oaOXIEbt5GW45DAVPgSZk5PoJ10 TZ5Ly4wqDzMa8YLDu4I2l2Use5wwIIYl5IbGEdZiRlYdox7eWaGRLfOiA8CPVb9p yZ7PgFZU10Y= =Bj83 -----END PGP SIGNATURE----- -- Ben.Goren(a)asu.edu, Arizona State University School of Music net.proselytizing (write for info): Protect your privacy; oppose Clipper. Voice concern over proposed Internet pricing schemes. Stamp out spamming. Finger ben(a)tux.music.asu.edu for PGP 2.3a public key.

1 0

Re: ID card from hell
by frissell＠panix.com 17 Dec '03

17 Dec '03

Three messages from Tim May concatenated: >Many of Duncan's refutations boil down to "What about tourists?" >The answer for the Feds is simple: temporary ID cards for tourists. >3. The card acts essentially like the one citizen-units would receive, >perhaps not fully authorizing certain things. > >4. Employers, even for casual work, would be required to check these >cards, > >Electronic "point of sale" terminals, similar to cash machines, could >make this "painless." Even if cash is not outlawed--a different, and >even more controversial topic--such "work permits" could be enforced >in various way. >I basically agree with Duncan's noble sentiments. I just don't think >it likely that the "What about the tourists?" argument will stop the >national ID juggernaut once it begins to move. >Perry makes my point well. That some people will be able to skirt the >system, or that the system will ultimately be unenforceable, does not >lessen my concerns. >Anyway, I see the imposition of internal passports--with a name chosen >for easiest acceptance, probably something like "Social Benefits >Card"--as very likely and not easily avoided. Just the tying of such >cards to driving would be devastatingly effective. As a reader of utopian & dystopian literature since the 1950's (remember, the Land of Oz was a utopian Socialist State), I was always hung up on control technology and the possibilities of the State getting carried away. Since I have been involved with computers and Cypherpunks my outlook has reversed. Let's assume for purposes of argument that the Feds have both the will and the money to impose tight financial and regulatory controls on American society. I will even assume that *we* don't exist and there is no high-tech opposition. Let's further assume that they rope in their NAFTA and G-7 (Gang of 7 Major Industrialized Countries - US, Canada, Japan, England, France, Germany, and Italy) partners in some sort of control regime to accomplish God knows what. I guess their psychology is something along the lines of Houseman: But no they will not, They must still Bend their neighbor to their Will And make me dance as *they* desire With jail and gallows and Hellfire. The "Coercive Metaphor" as I like to call it fits in well with Klinton's Bismarckian orientation. Note his lovefest with Helmut a few days ago. He likes the German model of labor markets (where *all* jobs require a certificate, where working during your holiday is a federal crime) and medical care. Where everything is either mandatory or prohibited. Opposed to this model is the Anglo Saxon model of individual rights. (X.25 vs TCP/IP to you networking types.) Here's the problem for Control Freaks: even if they get everything they want it does them no good. So they get their systems in place and they start spitting out data. What do they do with the data? You run checks on something the size of the G-7 economy, and you get tons of hits showing anomalies. You can't deal with all of them so you tighten your parameters until you cover the (small) part of the deviate population that you have the resources to do anything about. Oh, you can employ all sorts of AI programs to "catch" deviates and you can automate the punishment systems to deprive the deviants of "benefits" and permissions. We see this today with the automatic driver's license suspensions for a host of offenses." But cutting people off from "benefits" just gives them the incentive to *work* for a living and they come out of it stronger than ever with no ties to the government. Until you deploy killer robots to snuff deviates you can't make inroads into deviance purely by the application of force. The Commies tried. Slick Willie only has the balls to snuff a few Branch Davidians and Retail Pharmaceutical Salesmen. "Wet Work" is expensive. Our prisons are likewise. In New York City these days the big crusade is against unlicensed drivers running over kids. (Apparently, New Yorkers prefer their families to be decimated by *licensed* drivers.) The State automatically issues suspensions and people keep driving. The record is several hundred suspensions. Note if you look around that even though there are more control attempts by the State, things are less and less under control. Compliance is *not* increasing. Because direct application of force is difficult and expensive, it is rare. What Control Freaks need is *voluntary* obedience. But that is based on respect for powerful institutions (which is falling). We are primates and try to get along with our "tribe." In the past our "tribe" was village, then local area, then nation state. Today, technology has broken things up so that we can form many different sorts of "tribes." If we obey the mores of our "tribe" it is increasingly unlikely that that "tribe will correspond to the geographical nation state. I obey the mores of my *tribe* -- libertarian net dwellers. Others are members of other "tribes." We will collectively tend not to obey our nation states and their power will therefore decline. It is perfectly possible to have a rich full life even though you disobey your nation state. You need not even be an outcast any longer. There are *other* communities in which you can find full acceptance and support *without* having to move. I'm sure there will be much more to say later but I've run down... >And everyone go out and read or reread Brunner's "The Shockwave >Rider." Also Harry Harrison's "Stainless Steel Rat" series. Not for the technology but for the philosophy. DCF "Nineteen Eighty-Four Knocking at your door Will you let it come Will you let it run Your life." --- Awaiting proper definition of Mime sound file standards.

3 2

uh... excuse me
by nym＠netcom.com 17 Dec '03

17 Dec '03

Pardon me, but I think people are mistaking me for someone else. I read about this group from talk.politics.crypto (which I found poking around for info on Clipper). Are people objecting to the article I posted? I thought the interest in "digital cash" might warrant my posting that article about the U.S. currency changes. I guess I better just shut up and listen awhile longer on this list to find out what people are really interested in. I like to play an active role in posting contributions to every mailing list I subscribe to but I guess I was out of line here. Or maybe I should leave. You guys seem a little too paranoid for my tastes. (It's too bad, because I had some speculation I was hoping to share with you from some fascinating books I've been reading. You see, I'm a bit of a "new ager" and have been reading about prophecies that predict a new Hitler will arise partly via seizing the worldwide electronic infrastructure, and was curious what you would think.) Anyway, my apologies if my post was out of line. I'll be glad to leave if this list is not open to the general public or something. --nym(a)netcom.com

1 0

Re: Leaving the Country
by smb＠research.att.com 17 Dec '03

17 Dec '03

FYI - the states with no (general) tax on wages are: New Hampshire* Florida Texas Tennesee South Dakota Wyoming Nevada Washington Alaska* *No (general) sales tax either. How do they do it -- Magic. Magic? TANSTAAFL. Alaska has oil money (or has had it), and New Hampshire provides (relatively speaking) fewer services to its citizens, according to folks I know who have lived there. Knowing the politics of much of this list, that's probably considered a Good Thing by many; I disagree, but I won't clutter the list with (even more) politics. But if you're thinking of moving anywhere, find out what you *aren't* getting for your money, and see if it's worth it.

4 4

Remailer Detweiler filtering
by Hal 17 Dec '03

17 Dec '03

Nobody wrote: > But just from a philosophical view, I think any remailer operators > hunting for Detweiler-grams in all their incoming mail are employing > a procedure that is antithetical to their entire commitment. Isn't > it just a *teensy* bit hypocritical? Is this how you are going to build > cyberspatial-wide confidence in the use of your remailers for their > dependability and secrecy? Isn't it just a *teensy* bit hypocritical > to yell to the world that ANONYMITY IS THE RIGHT OF MAN and have a little > whisper in small print, "unless you are an official enemy of the > cypherpunks"? The issue is not a desire to deprive Detweiler of the benefits of anonymity; it is that he appears to do things which are designed to bring down the remailer network. By intentionally mass-posting to inappropriate news- groups, and injecting exponentially-growing messages into the remailer chain, he seems to be trying his best to deprive the benefits of the remailer net- work to others. This is, of course, in accordance with his well-known position against anonymity. If Detweiler succeeds, Nobody won't get to post anonymously (so to speak) anymore. I know that it is unfortunate that the remailer network is so fragile that a lone crackpot is a significant threat, but presently that is essentially what the situation is. > Cypherpunks, we believe in the philosophy that we are being oppressed > by numerous forces that seek to deprive us of our privacy-- big business, > the government, police, etc. But how can we claim to uphold the philosophy > of freedom of speech and privacy looking at our relationship with Detweiler? > Sometimes I think he was sent by God to test us. Pragmatically, I think that filtering Detweiler is more likely to provide privacy than not doing so. I understand the charges of inconsistency but IMO the particular facts of a case are a better guide to the proper action than abstract arguments. > What is it in the human psyche, rooted deep in our subconscious, that > pushes us to *vengeance* against those who offend us? That pushes us to > want to *expose* them (as the person said above, "EXPOSE THE BASTARD!!!"). > Do we have any consistent beliefs? How is that we, who are dedicated > to privacy, broadcast to everyone listening in a clear voice, that > "freedom of speech does not belong to people who offend us"? There is > a saying, "who will guard the guardians themselves"? Who will ensure > that those who advocate anonymity actually follow through, if they don't? If chained, encrypted remailing techniques are used, it is not a question of "exposing" anyone. Detweiler may be blocked from the network (if everyone agrees to do so) but it won't be possible to find out just what he is being blocked from doing. Had he been sufficiently careful in the first place there would not necessarily be any way of knowing who exactly was producing the offensive messages. In that case I believe most of the remailers would no longer exist. > I personally advocate that the Detweiler-Detritus be allowed through > the remailers unaltered as a blaring advertisement to the entire world > of cyberspace that yes, we believe that anonymity is *sacred*, even more > so than we believe that Detweiler is the AntiChrist of the Cypherpunks. I suggest, then, that you run a remailer (it only costs $20 a month on the system I use), and publicize the fact that Detweiler can use yours with impunity. Set up a mail-to-news gateway that other remailer operators can use so that they don't have to worry about the consequences of abuse. It's easy to talk about sacred ideas, but perhaps not so easy to keep an unpopular and misunderstood remailing infrastructure in place. "Xenon" also accused us of hypocrisy, started up a remailer, and stopped it in just a couple of days faced with these kinds of problems. > "I detest what you say, but I will defend to my death your right to > say it." --Voltaire > > "Freedom of speech does not end at the point that it offends; to the > contrary, that is where it begins" --Supreme Court justice (paraphrase) Detweiler is free to send any messages he likes; his service providers are free to continue or terminate his accounts as they see fit; and remailer operators are free to establish whatever policies they like for message handling. What better implementation of free speech could you want? Hal Finney hfinney(a)shell.portal.com

1 0

request for subscription
by SNMC62A＠prodigy.com 17 Dec '03

17 Dec '03

request for subscription

1 0

Clipper Costing / NSA ATM Crypto Venture
by NetSurfer 17 Dec '03

17 Dec '03

CP's, here's a couple of tidbits to get the juices flowing. They are from Infosecurity News July/August 1994 issue, page 10. The ATM part is particularly interesting... "Clipper Debate Rages Onnnn..." by Charlotte Adams Controversy continues to escalate over the government's Clipper escrowed-key proposal. Attacks include an analysis of just how much taxpayers would have to pay for it, and a Freedom-of-Information-Act request to hand over Clipper's escrowed keys. Meanwhile, the federal government continues to backpedal, saying that Clipper will not be mandatory, even for government users. What cost Clipper? To gauge Clipper's economic impact on taxpayers, Steve Walker, president of Trusted Information Systems Inc., offers the following analysis. Approximately 5,000 legally authorized wiretaps take place each year, based on a reported total of 800. He increased this reported figure to 1,000, for argument's sake, and multiplied by five to account for multiphone wiretaps.) There are approximately 500 million phones in the U.S., so the ratio of taps to the total number of phones is about O.001 percent. If the government taps 0.001 percent of AT&T's estimated market of 250,000 Clipper-equipped telephones, that works out to 2.5 key-escrow taps per year. Since the cost to run the country's two planned key-escrow centers is estimated at about $6 million per year, Clipper taps could cost taxpayers $2.4 million apiece (beyond the $250 million to buy the Clipper-equipped phones at $1,000 apiece in the first place). But if the number of Clip per phones sold is 100-fold greater than AT&T estimates--25 million devices--there would still be only 250 escrow taps per year and one call to key-escrow centers every 1.5 days, Walker figures. Each approved Clipper tap, under these cir- cumstances, would cost $24,000. Now, add to these escrowed-tap approval costs the estimated $50,000 to $60,000 that would be required to actually set up each wiretap. More than 1,000 Clipper crypto devices have been sold commercially since the products became available late last year, Department of Justice officials said. So far, the government has purchased another 9,000. Stalling tactics. A response to the Freedom-of-Information- Act (FOIA) request for Clipper's keys filed by Electronic Frontier Foundation cofounder, John Gilmore, is slow in coming. According to Gilmore's lawyer, Lee Tien, the FOIA applications--to the Department of the Treasury and Department of Commerce --have only generated requests for more time. The agencies, however, "seem to be making an effort to respond," Tien adds. Now, it's voluntary. Government witnesses testifying before back-to-back congressional hearings in May stressed the voluntary nature of key-escrow technology for both government and commercial use. But others demanded more than executive-branch reassurances. Raymond Kammer, deputy director of the National Institute of Standards and Technology, told a House Science, Space and Technology panel that he hopes government use will drive prices down far enough to make Clipper phones attractive to the public. Citizens may also wish to buy Clipper-equipped phones for communicating with government agencies, he added. Among Clipper critics, David Farber, professor of telecommunication systems at the University of Pennsylvania, told the House subcommittee that Congress needs to "weld into law" guarantees that Clipper will not be mandatory. And Trusted Information Systems' Walker told an earlier hearing before the SenateJudiciary Subcommittee on Technology and the Law that the administration should not "proceed on its own without separation of powers." He suggested putting key-es- crow centers under the judiciary branch, so that the executive branch "can't twist arms." --------------------------------------------------------------- NSA Launches ATM Encryption. Trying to stay in step with rapidly evolving telecommuncations technology, the National Security Agency has kicked off an Asynchronous Transfer Mode (ATM) encryption program to provide end-to-end encoding across synchronous optical network (SONET) systems. Called Fastlane, the ATM project specifies optical- channel (OC) rates, with OC12 desired. The work will parallel a SONET encryptor development program awarded to Motorola earlier this year. --------------------------------------------------------------- Disclaimer: these articles were scanned, not forwarded <hehehe> -NetSurfer #include standard.disclaimer >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> == = = |James D. Wilson |V.PGP 2.4: 512/E12FCD 1994/03/17 > " " " |P. O. Box 15432 | finger for full PGP key > " " /\ " |Honolulu, HI 96830 |====================================> \" "/ \" |Serendipitous Solutions| Also NetSurfer(a)sersol.com > >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

1 0

Nixon, Johnson, and the Dollar
by Jeff Gostin 17 Dec '03

17 Dec '03

Sandy Sandfort <sandfort(a)crl.com> writes: > Don't believe it? What did Johnson say just before silver was removed > from US coinage? For bonus points, what did Nixon say two weeks before > he closed the gold window to foreigners and devalued the dollar? I'll show my ignorance of history once again... I've not clue, but you've tickled my curiosity. Care to inform us? :-) --Jeff -- ====== ====== +----------------jgostin(a)eternal.pha.pa.us----------------+ == == | The new, improved, environmentally safe, bigger, better,| == == -= | faster, hypo-allergenic, AND politically correct .sig. | ==== ====== | Now with a new fresh lemon scent! | PGP Key Available +---------------------------------------------------------+

2 1