December 2003 - cypherpunks-legacy

RETURNED MAIL, ANYBODY?
by drzaphod＠brewmeister.xstablu.com 17 Dec '03

17 Dec '03

Has anybody else who has posted to the list in the past day gotten a return from css(a)netcom.com [Chris Schefter]? All my posts were mailed back to me by css with a note: "Unsolisted letter...returned" BTW: The spelling is quoted correctly. I've inquired css but havn't gotten a reply. TTFN. -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - DrZaphod #Don't Come Any Closer Or I'll Encrypt! - - [AC/DC] / [DnA][HP] #Xcitement thru Technology and Creativity - - [drzaphod(a)brewmeister.xstablu.com] [MindPolice Censored This Bit] - - 50 19 1C F3 5F 34 53 B7 B9 BB 7A 40 37 67 09 5B - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

5 4

Re: PGP posting validation
by trestrab＠GVSU.EDU 17 Dec '03

17 Dec '03

DrZaphod writes: > Ah ha! Another facet of a LIST run authentication system > [Cypherpunks Run Authentication System - CRASs?] > may be to allow users to mail THE LIST and ask their mail > to be filtered so they don't have to pay for msgs they don't > want. THIS IS DANGEROUS because a forger could mail in > such a request.. to patch this, THE LIST could mail out a > weekly msg -- a compiled list of which msgs got thru, and > which were filtered, and why. > > This opens up another possibility which may prove more > effective. If THE LIST can maintain a list of msgs/posters/ > PGP authentication, then those users who trust THE LIST > to authenticate their mail can select the msgs they want > to receive [from the same compiled list]. The list software that Ray Cromwell wrote for the Extropians list does all these things. Users can exclude [user|thread] and receive a regular ( I think daily) list of filtered msgs. Users can also set their own mode of receiving the list (reflected or digest), and can temporarily unsubscribe by excluding all. About the only intervention needed by list management is for subscribes and permanent unsubscribes, other than general maintenance. The list also has a security feature that when turned on only accepts posts from addresses previously authenticated as belonging to list members. This can still be spoofed, but at much greater difficulty than LD has had to expend heretofore. Jeff trestrab(a)gvsu.edu

1 0

Internet Banking
by dwomack＠runner.utsa.edu 17 Dec '03

17 Dec '03

As a recent subscriber, I'd like to ask a question or two regarding banking... 1. Such a bank would clearly need to be offshore, and in a location with no banking or taxation treaties with the US. Some of these are available, but, the record of quality accounting and regulation isn't always the best. What is the feeling about what depositors would want from such a bank before they "did business". 2. What is the minimum list of services a bank should provide? Deposit and transfer between inhouse accounts would be easy...wire transfer elsewhere would not be a problem...even handling some investments could be done. On the other hand, things like unsecured loans, credit cards, and immediate access to someone in "account services" would be a lot more difficult to implement. 3. What minimum account size would be appropriate? Domestic branches of offshore banks (i.e., Union Bank of Switzerland) generally require 100M, but the branches are subject to US record keeping requirements. Offshore, $1,000 or less is common. Would such levels, in your opinion, work? 4. Statements. Generating statements is time and accounting intensive. Generating statements on checking accounts is NOT cheap. Would accounts have to have scores of small transactions (i.e., $25 bucks for the xyz bill), or would it be possible to merely feed a domestic account from the offshore account. 5. Currency. US dollars? Swiss Franks? Other? Foreign currency is a nice option, but would up the costs for a startup bank. 6. Yield. Would depositors demand a yield? Current low rates at US banks seem to make this unimportant, but is it really unimportant? 7. Fees. What level of fees could the bank charge? A regular US checking account for a small business can cost $10 a month... 8. Any thoughts about marketing? Offshore banks, as I understand it, cannot lawfully advertise their services within the US. 9. Other thoughts? What do YOU think would be needed, unneeded, nice to have, in such an entity? What would frighten potential customers away? Thanks for any thoughts you might have.

2 1

Crypto and Taxes
by hfinney＠shell.portal.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I'll just briefly recap some of the points: Hal> I don't agree with the extreme position that cryptography will lead to Hal> the failure of the income tax and the destruction of the government. Perry> No one is arguing, by the way, that all the economy will go black. I'm Perry> merely noting that whereas right now its hard to lead a normal life Perry> entirely in the black economy (you suffer from a myriad of Perry> inconveniences), an anonymous offshore banking system that you have Perry> free access to changes all that. It's not clear to what extent Perry and I disagree here. I agree that some people will exploit the new opportunities. My doubt is whether the vast silent majority on which the government depends for its taxation revenues will do so. I read the other day that the government gets something like 70-80% of its tax revenues from people making less than $35,000 a year. These people are not financially sophisticated. Duncan> In traditional Black Markets, the transactions are illegal. In future Duncan> Black Markets on the nets, most of the transactions will be legal. Duncan> Legality certainly encourages transactions relative to illegality. I agree with the last point about legality helping, but I don't understand why most transactions will become legal in the future. I thought we were talking about ways to evade laws via cryptography. My assumption is that governments would actually crack down when faced with lost revenue, similar to what was described as happening in Italy: ??> Customers are fined for leaving businesses without a receipt. Your car may ??> be stopped and searched for undocumented merchandise at any time. Imputed ??> income taxes for self-employed people are at ridiculous levels (i.e. a ??> large degree of evasion is assumed). Taxes are levied on everything (car ??> radios, the width of your driveway, electric lighters for gas stoves). So I'd think even more transactions would be illegal in the future. The main issue, I thought, was whether people would widely risk violating these laws in order to save on their taxes, and whether cryptography would let them do so with impunity. Duncan> You left out a few information purchases: education, much of medicine, Duncan> all of financial services, design, marketing, supervision, and Duncan> management. Duncan is pointing out that more of the economy is in the form of information than I was counting. Even if we count these things as information, though, the question is still to what extent the providers of these services will be able to escape taxation. Take medicine as an example. I should be able to go to a doctor today and pay him cash, off the books, at a greatly reduced rate, for my medical services. Yet no doctor I've ever seen, and I've probably seen twenty or thirty in my lifetime, has ever suggested that. I don't see how cryptography will change this. Duncan> Once the interface is good enough, virtual offices with full workgroup Duncan> interaction built of pure information will spring up and the Duncan> "information" component of much of what we think of as physical work Duncan> will become apparent. Is this the key, people working for virtual businesses? No one knows the true name of anybody else, so no one is afraid of being caught? I am still skeptical. A whole nation of people tele-commutes to work for companies whose name they don't know, with co-workers protected by aliases, all so they can be paid in cash for their services. I find this notion implausible in the extreme. Joe and Jane Sixpack aren't going to want to work for a boss who wears a mask. Duncan> If you wander down the shopping street of a future MUD/MOO and you Duncan> buy or sell things, what nation has jurisdiction for tax purposes. Duncan> What if the MUD/MOO exists as a set of cooperative processes spread Duncan> around the globe. This may be uncertain now, but I don't see why it would always remain that way. There is nothing to stop governments from declaring, say, that residents in their boundaries are subject to their taxation, or that their citizens are subject to their taxes regardless of where they do business. More interesting from the crypto perspective would be the case where the business in the MUD refuses to disclose its true nationality or location. There again, though, I think running an anonymous corporation will pose many practical problems. Sandy> You don't need to Sandy> *own* a car, to have the *use* of a car. Imagine leasing a car Sandy> and using your cyberspace bank digital checks, digital money or Sandy> credit card to pay the monthly rent. No audit trail, and no Sandy> asset to be seized. Similar techniques can be used for virtually Sandy> all of your assets. How does this bear on the issue of government collapse due to failure of income tax? This example actually strikes me as an unobjectionable use of cryptography, one in which individual privacy is protected. The only tax consequence I see is possible avoidance of sales tax, which is col- lected only at the state level and not the national level. Sandy's example shows that car rental agencies might be able to operate on a cash basis, like the local fried chicken outlet. I don't see how this brings down the government. Sandy> I think Hal hasn't been reading Duncan or my posts very closely. Sandy> Here's a hint: A Cayman Islands corporation is a non-US citizen Sandy> even if it is owned by an American. Sandy is replying to my question about moving out of the country to avoid taxes. I gather that he is suggesting that people could set up corporations in the Cayman Islands and somehow divert some of their income to them, so that the income would be shielded from taxes. Can this be done today? Can I go to my boss and ask him to start sending my salary to this numbered bank account in the Cayman Islands, and to stop troubling the U.S. government with information about how much he is paying me? Sounds great. Why doesn't everybody do it, and why will everybody start doing it in the future? To sum up, I am willing to accept that people will be eager to avoid paying taxes, but I still doubt that cryptography will bring down the United States government. Particularly when we consider the lack of sophistication (both financial and technical) of the vast middle class who provide the bulk of tax receipts, I think that virtual corporations and offshore tax havens are not likely to become widespread enough to seriously endanger the government. (In response to John Kreznar's comments about my use of the term "cheating" to refer to violation of tax laws, I accept the thrust of his comments but I'll just observe that while preparing a false set of books may be justified and in some circumstances even honorable, it is not honest.) Hal -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLTnMY6gTA69YIUw3AQHqlgP/d6GTpNrK3c4B/jfxT5tQqAJ0uBkvc5Y7 fh1dtj16QrR/CmAHEJVV+JWXUjwaTnjqO1RdgPJfjjG1U7CaSiuy84OVlyQPSpAc JeIC7qa1HfqXRCK/bQmxcJMhbOULMKkk2plphcwDvL2Tlxe8DXvmgDLS21DUV6r+ bOT9RBf2U3c= =p55w -----END PGP SIGNATURE-----

2 1

Re: RETURNED MAIL, ANYBODY?
by drzaphod＠brewmeister.xstablu.com 17 Dec '03

17 Dec '03

> > Yep, happened to me too (subject was "Re: PGP posting validation", a > coincidence?). > > Felix I wondered about that myself.. |-] -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- - DrZaphod #Don't Come Any Closer Or I'll Encrypt! - - [AC/DC] / [DnA][HP] #Xcitement thru Technology and Creativity - - [drzaphod(a)brewmeister.xstablu.com] [MindPolice Censored This Bit] - - 50 19 1C F3 5F 34 53 B7 B9 BB 7A 40 37 67 09 5B - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

1 0

Re: RETURNED MAIL, ANYBODY
by Hal 17 Dec '03

17 Dec '03

I sent mail to Schefter about this, and got a reply last night. He said this was his form of protest for having his unsubscribe messages ignored. I suggested that he bit-bucket the unwanted messages rather than bouncing them, and give Eric a little more time to process his unsub request. Hal

3 2

MISC; reminder
by Brian D Williams 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Punkster's Just a reminder, tonight on Nova "Codebreakers" stories of the world war II effort, check local listings. Brian Williams Extropian Cypherpatriot Cryptocosmology: "sufficently advanced comunication is indistinguishable from noise." Chicago Current temp -19 wind chill -63 4pm CST -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLTxaZNCcBnAsu2t1AQFOdgP+NgsenZ1otjdCLK2aUKnhGW1Nwdr2Rk5+ V76MAqVWn+Iys190Hj3HTHBd6/aI7rhP9mRYWo8097pgYhMRmR35URa+hKRNLzP0 Iucwe9cWTLlbNnTPh+N04JJbwVhlDG3xXGykKeY0y9+IkUIeS0acu8nL01oGIurX hkqqkgmzxvE= =X6Ol -----END PGP SIGNATURE-----

1 0

Mr. Squish
by consensus＠netcom.com 17 Dec '03

17 Dec '03

I explained to a list manager about our problems regarding the source of SQUISH and the system administrator that does not respond to requests to deal with the problem. >Return-Path: <rburns(a)cap.gwu.edu> >Received: from cap.gwu.edu by mail.netcom.com (8.6.4/SMI-4.1/Netcom) > id MAA05785; Tue, 18 Jan 1994 12:47:43 -0800 >Received: by cap.gwu.edu (4.1/SMI-4.1) > id AA18334; Tue, 18 Jan 94 15:47:25 EST >Date: Tue, 18 Jan 1994 15:42:14 -0500 (EST) >From: Roger Burns <rburns(a)cap.gwu.edu> >Subject: Mr. Squish >To: Christopher Allen <consensus(a)netcom.com> >Message-Id: <Pine.3.07.9401181514.C17300-b100000(a)cap.gwu.edu> >Mime-Version: 1.0 >Content-Type: TEXT/PLAIN; charset=US-ASCII > > >> Do you mind if I forward the above message to the cyperpunks list? Or would >> you be willing to do so yourself? > >Sure. I don't have ready access to Usenet, so feel free to post my comments. >BTW, I found two typos in my original message, so below I copy a revised >version. > >-- Roger rburns(a)cap.gwu.edu > >---------- Forwarded message ---------- > >Thank you for response. If I may give some unsolicited advice, I would >take a "Margaret Thatcher" approach to the postmaster who is allowing this >problem to fester. > >I would (1) trace down his office phone number (through nic whois or >nslookup or whatever) and ring him up and explain that allowing abuse of >the Internet is an abrogation of his company's contract for obtaining an >Internet connection, and that if he doesn't consent to handling this problem >appropriately, you will seek to have his company disconnected from >Internet based on that abrogation. > >If he doesn't succumb to that threat, I would then (2) phone thru his >company's main switchboard and ask to be connected to the postmaster's >boss, and explain to the boss that you will be seeking to have his >company's Internet connection ended due to the postmaster's >irresponsibility in allowing an unstable person to wreak havoc on the >worldwide computer network community. > >Now while it might be a pain to actually follow through on this threat, if >you put enough of a poker face on what you say, the bluff might very well >be enough at either step 1 or step 2. > >Good luck! And I hope I don't have to deal with this problem! (But if I >do, I've outlined how I will proceed.) > >-- Roger Burns rburns(a)cap.gwu.edu ..Christopher Allen ..Consensus Develoment Corporation ..4104-24th Street #419 ..San Francisco, CA 94114-3615 ..(415) 647-6384 Fax ..(415) 647-6383 Voice ..email: consensus(a)netcom.com ..mosaic frontpage: .."ftp://netcom.com/pub/consensus/www/ConsensusFrontDoor.html"

1 0

BOF @Usenix?
by Sameer 17 Dec '03

17 Dec '03

Where/when is the Cypherpunks BOF @ Usenix? It's Wednesday right? (What's the hotel?) Thanks, Sameer

2 1

Re: RSA Questions
by norm＠netcom.com 17 Dec '03

17 Dec '03

At 8:54 1/18/94 +0000, SuperDupont wrote: >Hi Cypherpunks ! > >I've got a few questions about the RSA encoding (if they're answered somewhere >in litterature, just give directions, thanks) > > If the public encryption key is e (the exponent) and n=p*q (the modulus), > then the encryption scheme is: > > cypher= (plain^e) mod n. > > Number theory tells us that the reverse operation (taking the e-th root) > can be performed, as long as we know p and q: we know how to compute d > such that for any plain<n, (plain^e)^d=plain. > > Now my questions are: > > 1. Is there a way to determine ALL the possible values of d verifying: > (plain^e)^d=plain for any plain<n (or at least have an evaluation for > their number) ? > > In other words, is there a way to know the number of keys that unlock > what your public key locks ? > > 2. Is there a way to determine ALL the possible values of d verifying: > (plain^e)^d=plain for *a given plain* ? > > In other words, is there a way to know the number of keys that unlock > *a given message* ? > >Here's an example that's quite worrying (maybe because I chose p and q >to be random primes, and they have bad properties): > >e=17 # Exponent >p=967 # Prime p >q=1031 # Prime q >n=p*q=996977 # Public modulus > >phi=(p-1)*(q-1)=994980 >g=gcd(p-1,q-1)=2 >f=phi/g=497490 >d=(1/e) mod f=234113 # A possible value of d given by number theory > >Here's the result of the exhaustive search for the answer to question No. 2: > >plain=12345 >cipher=(plain^e) mod n >decipher=(cipher^d) mod n > >The possible values for d (138 of them) are: > >3393 10603 17813 25023 32233 39443 46653 53863 61073 68283 75493 82703 89913 >97123 104333 111543 118753 125963 133173 140383 147593 154803 162013 169223 >176433 183643 190853 198063 205273 212483 219693 226903 234113 241323 248533 >255743 262953 270163 277373 284583 291793 299003 306213 313423 320633 327843 >335053 342263 349473 356683 363893 371103 378313 385523 392733 399943 407153 >414363 421573 428783 435993 443203 450413 457623 464833 472043 479253 486463 >493673 500883 508093 515303 522513 529723 536933 544143 551353 558563 565773 >572983 580193 587403 594613 601823 609033 616243 623453 630663 637873 645083 >652293 659503 666713 673923 681133 688343 695553 702763 709973 717183 724393 >731603 738813 746023 753233 760443 767653 774863 782073 789283 796493 803703 >810913 818123 825333 832543 839753 846963 854173 861383 868593 875803 883013 >890223 897433 904643 911853 919063 926273 933483 940693 947903 955113 962323 >969533 976743 983953 991163 > >That makes a probability of 0.013% >Looks to me like it's a LOT. Maybe I'm wrong. > >-zap > >------------------------------------------------------------------------- >To find out more about the anon service, send mail to help(a)anon.penet.fi. >Due to the double-blind, any mail replies to this message will be anonymized, >and an anonymous id will be allocated automatically. You have been warned. >Please report any problems, inappropriate use etc. to admin(a)anon.penet.fi. Laudable Paranoia! In short the numbers: cipher, decipher, plain, d and e must all be relatively prime to p and q for all of this stuff to work. In practice, since p and q are very large, the probability of the cryptanalyst finding another value d that deciphers your message is about the same as him finding p or q. That is the same probability of him factoring pq by guessing. In your example 138 out of 996980 is about the probability of being divisible by either p or q. You might check to make sure that the message that you are enciphering is relatively prime to p and q. You could better spend your, however, verifying that your hardware had not made a mistake, which is more likely, unless, however you are sending one of your factors so that a friend can share your secret key. In that case, however, anyone with your public key can compute your secret key,

1 0