December 2003 - cypherpunks-legacy

RE: Distributed DES crack
by Andy Brown 17 Dec '03

17 Dec '03

On 22 July 1996 22:48, Ben Holiday[SMTP:ncognito@gate.net] wrote: > I've a few machines around that could be dedicated almost full time to the > task. What are the bandwidth requirements? Specifically, could the > keycracker be run over a 28.8 (with a 486 running linux)? If so, how many > 486's could I get over a single 28.8 (i.e. 28.8 -> multiple 486's daisy > chained with ppp over direct serial connection)? It's not a factor of the bandwidth, you search offline and send in your results to a central server. But first, a little reality check is in order. According to libdes, the 200Mhz Pentium Pro on my desk will do 1,827,997 ECB bytes/sec, or 228,499 ECB blocks. A DES crack would have to try, on average, 2^55 blocks. That would take my machine 43,798,875 hours, or 1,824,953 days. OK, so let's be reasonable and say that a week would be a good time to come up with a DES key. We would need 260,707 200Mhz Pentium Pro's to achieve this. Looking at that, 30 days seems not such an unreasonable target. We would need 60,831 200Mhz Pentium Pro's to achieve this. It seems obvious to me that DES is still *way* out of reach of anything other than special purpose hardware. Regards, - Andy (hoping he got his sums right) PS. For those more acquainted with Sun hardware, an Ultra-1 will do 1,683,647 ECB bytes/sec (gcc 2.7.2).

1 0

Returned mail.No such addressee
by Remo Pini 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- To: cypherpunks(a)toad.com Date: Tue Jul 23 13:53:56 1996 To: cypherpunks Date: Tue Jul 23 13:53:09 1996 Check out Pronto Secure (I think there's a mac version around!?) http://www.commtouch.com/ - --------< fate favors the prepared mind >-------- Remo Pini rp(a)rpini.com PGP: http://www.rpini.com/remopini/rpcrypto.html - ------< words are what reality is made of >------ -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: noconv iQEVAwUBMfS9VRFhy5sz+bTpAQHYFQgA1BzJQY9dM29KYpuzdhW9GRG/Ng0M9x5o d2obC6MWa3th3vCjr3qSb1yC4IKLXLACFvTa4/jHky8P3//3UOUVNZ0IffPSlY9/ a3dKSKFyMUaKtyzi7rzCV24NlBFT1eJVVNZjYsH8pbCGbxteH5+dRAvvbkmSPukX GGa1oY6u/XK7Ti8IaOifWFDvYi76W37UlLs9aSGAfpTWKlM88bnkUL3iPxHf8qs6 DE0PQZOE8M4JyQTc/H7E5oNkEhE9RxIOgJNpZGPSOazwh3MVjTBLIZZOpmsV3srv EH6aNobS5shKOs8t/t7aCXIzhvNRDEqB652bfPP79Q13ICOT7BBYyA== =4iqW -----END PGP SIGNATURE-----

1 0

DNA printing at birth
by William "Bud" Kennedy 17 Dec '03

17 Dec '03

[IMAGE] UK News Electronic Telegraph Monday July 22 1996 [IMAGE] Issue 432 See text menu at bottom of page [IMAGE] Labour plans DNA tests for everyone from birth By Rachel Sylvester, Political Staff _________________________________________________________________ External Links [IMAGE] Identity cards - A solution looking for a problem [IMAGE] Conservative Way Forward - Danger of compulsory ID cards [IMAGE] Centre for Computing and Social Responsibility - response to Green Paper on ID cards _________________________________________________________________ RADICAL plans to take the genetic fingerprints of everyone in Britain and put them on a compulsory ID card are being considered by Labour. A database of DNA identities would be used to solve crimes and cut benefit fraud. Frank Field, tipped as a possible Social Security Secretary in a Labour government, has drafted proposals to produce a genetic database of the nation. Blood samples would taken from babies and from people applying to live in Britain. The samples would be used to extract DNA. Only identical twins have the same DNA, making it a more accurate fingerprint. Chris Smith, shadow social security secretary, confirmed that the party was examining the plans. "We are not ruling them out. We are determined to cleanse the national insurance system of fraud." John Wadham, director of the pressure group Liberty, said it was incredible that a senior Labour politician could suggest such a "draconian" measure which breaches European law. "There is no evidence that such a massive invasion of our privacy would do anything very much towards stopping crime." Mr Field, chairman of the Commons Social Security Select Committee, also advocates taking compulsory fingerprints from every citizen as a fall back to confirming a person's identity. The combination of genetic and physical fingerprints would be used to "rebuild the national insurance system", Mr Field said yesterday. Each person would be given a number at birth, combining national insurance and health numbers, which would tally with their genetic code. The information would go on a computerised identity card. It would also contain an individual's address, medical history or criminal record. It would be impossible to claim benefit without the card. Mr Field believes that Mr Blair is "sympathetic" to the proposals and that any opponents would be "Old Labour". Mr Blair will express his determination to reduce fraud in a speech tomorrow. Mr Field, MP for Birkenhead and described as one of Blair's "gurus", believes that the scheme would eradicate the use of multiple identities by benefit fraudsters and wipe off a large part of the annual £2.5 billion cost of social security fraud. There are millions of bogus national insurance numbers in circulation. It would also allow police to solve more crimes because traces of semen, hair or skin found at a burglary or rape would lead to the culprit. Ann Widdecombe, the Home Office Minister, said: "There are huge practical and resource implications. Think what it would cost to test every person in the country." Mr Wadham said there was no guarantee that the benefits would follow. "Not all fraud is to do with bogus national insurance numbers. I do not understand why, even if you have an identity card, you would need a genetic database." Mr Field advocates setting up a commission to regulate the use of the information. It is unlikely that any policy would be announced before the general election. Mr Blair, who is known to be looking for radical ideas on social security, has told shadow ministers to "think the unthinkable" and remind the public that "with rights come responsibilities".

1 0

Re: Borders *are* transparent
by Remo Pini 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- To: cypherpunks(a)toad.com Date: Tue Jul 23 13:52:55 1996 > What about the possibility of using DSP's? Is there any brand of 28.8 K > modem which uses a "standard" DSP and EPROM firmware? Such a beast > might be > the easiest way to get a large amount of CPU horsepower operating > independently of the host computer. DSP's are optimized to execute a > large > number of instructions with little I/O needs. > > Jim Bell Zyxel modems (ISDN and V34) have a Motorola 56000 DSP and a Motorola 68000. - --------< fate favors the prepared mind >-------- Remo Pini rp(a)rpini.com PGP: http://www.rpini.com/remopini/rpcrypto.html - ------< words are what reality is made of >------ -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: noconv iQEVAwUBMfS9GBFhy5sz+bTpAQFQbggAp7H0bJyZYUzD87QQorzRQ8PTaTkkB1r3 tsEj6JWmq4PlppRd0lvjP2mN4LmfR700A8O7qdr6d9IfKrNmQzItDEPjq2zv+Lbf P9e2mi7Jz1xl1faZv3YiBAbdhv/jlnI0m4o3x6AwZNkAy1pe3xkj61H9n8tQ3cqf eAwDqZlOCCsjuN/hdJAiIHuiuqC2W0i59bZR59u6ek8iXE+8LnXXxeMxuUOZVIgI 2efgoJk6ev5/7IOoDaMlgffkHcWTTnjEClBI3JnGcIOnauacYG8t8UuPa5R8Td0t ZL/O8/gEDKGpos8j92DSyDgjb6XrRWq3CWZbfoXDhfRbnTKU/ZqXUg== =2afe -----END PGP SIGNATURE-----

1 0

Re: Digital Watermarks for copy protection in recent Billbo
by Deranged Mutant 17 Dec '03

17 Dec '03

On 22 Jul 96 at 11:19, Alex F wrote: [..] > > Other issues: what if an eavesdropper steals the music or video? It's > > If they steal it, well, who cares? If there is something worked out > so that they could trace STOLEN (not traded or sold) CDs then fine, > arrest them. Do you really think though that anyone would waste so > much time over $8? Sniffers aren't much effort, and if I sniff your tagged purchases and put them out over the net anonymously, they are traced to you. > > If it uses a credit-card number as (part of) an ID, that's pretty > > bad. Someone can sniff for CC numbers if they know how it's stored. > > Probably not done that way. My guess is that the disk ID is assigned [..] Probably, but the Billboard article discussed using CC numbers as an ID in the online watermarked transactions. Doesn't mean they were correct, of course. > > The system will have to rely on proprietary tech and security through > > obscurity. Even know how watermarks are stored without understanding > > the math, one must be able to somehow garble the sound without > > distorting it, but which renders the watermark useless. > > Actually, this would be quite easy. The "watermark" would be a > signal that plays inband, but out of our hearing range during the > entire CD. The human ear can only hear in the 20-20,000 (Hz, KHZ?, > whatever) range. It would be trivial to add a digital ID signal at, > say 30,000 or 15 or something like that. This could then be decoded, > if need be. This seems the easiest and most efficient way. This > could also be defeated with a lot of $$ (and/or a LOT of HD space). > If the frequecy is known (it can be found out) it can easily be run > through recording studio eqipment that can very effectively isolate > the frequency and cut it out. If you have a LOT of HDD space > (digital audio at 2 stereo tracks, not sure of the sampling rate or > bit resolution, takes about 20MB of HDD space per minute (2 tracks, > good sampling and bit rate) ) you could probably find the freq. > fairly easily by isolation and just edit it out, and write the new > stuff to a CD-R. If the signal is purely digital, I would imagine > that it might be even easier that if it were an analog signal (?). > Someone w/ good equipment (Digital Labs' stuff, or SAW (Software > Audio Workshop) would be able to do this w/o much problem. The > question is is the price/effort worth it? In quantity maybe. On an > individual basis, only if you already happen to have the erquipment. For someone in the misuc counterfit business, the equipment is probably not that expensive. It's innocuous enough (recording and editing equipment) that it wouldn't draw suspicion. AFAIK, most "bootlegging" is of unreleased concerts or out-takes. Digital watermarks would be of little use. > I have a suspiscion that this type of thing will not really come to > any kind of fruition due to not only the ability to defeat this, but > mainly due to things like buying at a garage sale, etc. If it did, I think it's intended for tagging online transactions. You connect to a company's site and download the latest album or single by some band, presumably with the rights to transfer that to a tape for personal use. If this becomes a predominant way of buying music or movies in an eventual future (when most people on the planet are wired) and anonymous purchases disappear (I doubt it) as well as radio broadcasts (another loophole) die out (quite doubtful as well). Of course you have to be foolish to pirate under such a system using your own name. Rob --- No-frills sig. Befriend my mail filter by sending a message with the subject "send help" Key-ID: 5D3F2E99 1996/04/22 wlkngowl(a)unix.asb.com (root@magneto) AB1F4831 1993/05/10 Deranged Mutant <wlkngowl(a)unix.asb.com> Send a message with the subject "send pgp-key" for a copy of my key.

1 0

Re: Borders *are* transparent
by Remo Pini 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- To: cypherpunks(a)toad.com Date: Tue Jul 23 13:52:47 1996 At 10:32 AM 7/22/96 -0006, Peter Trei wrote: >Any one up for a distributed brute force attack on single DES? My >back-of-the-envelope calculations and guesstimates put this on the >hairy edge of doability (the critical factor is how many machines can >be recruited - a non-trivial cash prize would help). I'm in with: 2x Pentium75 1x 486DX50 (all machines Win NT 4.0, it would have to be an Intel runnable algorithm) - --------< fate favors the prepared mind >-------- Remo Pini rp(a)rpini.com PGP: http://www.rpini.com/remopini/rpcrypto.html - ------< words are what reality is made of >------ -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: noconv iQEVAwUBMfS9FBFhy5sz+bTpAQFV6gf/ZkarvomYeMqyHSGz5yAtLPey6ucFx1AJ 1PfqZV/UJp92d9tX9DmfESHTqZcyCRDHq9+ziDh5vRr5PHovVVOkg9TClssYYk3l M75EZ20bNohI3ISTH28yUN9H/JdxvlPrDQp7Gwa0LU9QFhBsmpzaLbyL+aas1DA7 sUD6Yc8wBTg95OswYkOqc49DzyEdH6obfL0NhN2QuaSvJDIV/8vfdr08ZhW2ZGfF TmbGf8z3lWpuZpzhIDRypb74xrg0PJHuvL0OMsEe3HV0euUCpvCwK18YlAaJLoW9 R9Pep6Cq5u+13MDlYM20OZ+RVpUPvKrCY0t0//W8OArgiVCiaC7tVw== =tUpt -----END PGP SIGNATURE-----

1 0

RE: Responding to Pre-dawn Unannounced Ninja Raids
by jbugden＠smtplink.alis.ca 17 Dec '03

17 Dec '03

perry(a)piermont.com writes: >Jefferson could have been a mass murderer for all I care. His words >may be evaluated fully independently of his actions. They are not >interdependent. Many of our current politicians would be heartened by your sentiment: Do as I say, not as I do. Personally, I incline more towards the other cliche: Actions speak louder than words. It is possible to evaluate a persons words independent of their actions, but, given that environment has some effect on behaviour, it is not at all clear that you can treat them as independent. Nor, more importantly, would you want to. It could prove to be a good breeding ground for cynicism. Or is it sarcasm... ;-) James

2 1

Re: Another fascist
by Alex de Joode 17 Dec '03

17 Dec '03

In article <v03007807ae1a03b2bb71(a)[192.187.162.15]> you wrote: : One of the great friends of free speech on this list sent a forged cancel : message to the listbot to try to cancel my subscription. : The listbot, being reasonably well designed, ignored him and told me about : it, though I have no doubt less of a dunce could bring it off. : What a piece of slime! What do others think of this practice? : David Why, David, did you decide to subscribe to cypherpunks ? bEST Regards, -- -AJ-

2 1

Re: Brute Force DES
by Peter Trei 17 Dec '03

17 Dec '03

> Peter wrote: > >Any one up for a distributed brute force attack on single DES? My > >back-of-the-envelope calculations and guesstimates put this on the > >hairy edge of doability (the critical factor is how many machines can > >be recruited - a non-trivial cash prize would help). Duncan wrote: > I volunteer my 120 MHZ Pentium. A lot more Pentiums are out there now than > a year ago. That makes it more feasible. A lot more people with full net > connections. Like most Americans, I have a flat rate net connection and a > flat rate local phone connection so could run a cracking session permanently > (as long as no one tells my ISP). We need a full test of the Winsock > cracking client in any case. It wasn't working very well last time. > > DCF <back-of-envelope> In my terminology, 'hairy edge of doability" means we have a shot at success, but I wouldn't bet the farm on it. I thought that I might bet a couple hundred bucks, though. Sadly, after further calculation, I'm not so sure if it's doable just yet. What I'm looking at is a known plaintext attack on single ECB DES, using a brute-force test to cycle through the key space. People would get chunks of keyspace to test from a central server or servers, and would be motivated to take part by a cash prize for the lucky person who finds the key. Lets do the numbers: Single DES has the security of 56 bits of key - there are 64 bits in the keys, but 8 of them are parity bits which add nothing to security. 2^56 = 7.205e16 keys (which is a whopping big number) Let's guess that we can recruit the equivalent of full-time on 1000 machines. 7.205e13 keys/machine. Let's guess that we have about a month before people start to lose interest - so we want to be more than 1/2 done by then. Lets say we want to sweep the whole space in 40 days. 1.8e12 keys/machine/day ~21,000,000 keys/machine/second The fastest general purpose, freely available des implementation I'm aware of is libdes. by Eric Young. With this, I can do a set_key in 15.8 us, and an ecb_encrypt in 95 us/block. That adds up to about 9,000 keytests/sec (this is on a 90 MHz P5, running NT). I'm looking at ideas to speed up DES - if I'm willing to use honking great lookup tables, the permutation steps can be done more quickly than libdes. I'm also looking at implementing the algolrithm in hand-optimized P5 assembler. (It's been years since I've done a major assembler project - the P5 has some truely weird features to be considered, but also has (some) internal 64 bit registers to play with). Let's guess that I can speed up a key-test up by a factor of 10. (This is not a slur on Eric's code - it's extremely clever, but not optimized for any particular processor, or for key-testing. Note that the keytest described above takes about 10,000 cycles/test.) That gets my workstation up to about 90,000 keys a second, which is still almost a factor of 250 too slow. I'm going ahead with my work on a faster DES keytester, but unless optimizing gives an astounding win, I now think a distributed bruting effort is a bit pre-mature. What will make this brute doable, if not now, then in the near future? 1. Faster Processors - Moore's Law is still holding. A year ago, my 90 MHz Pentium was one of the faster machines taking part in the 40-bit RC4 crack. Now, it's passe. 2. More processors. The number of people on the internet continues to grow rapidly. 3. More interest - Crypto awareness has greatly increased in the last year, and a real cash prize (say, over $500) will generate both publicity and interest. These factors all multiply together. The number of cycles that could probably be recruited is increasing at a fast rate. A major part of the work will be a keyspace distribution mechanism which can handle the load (this was a major stumbling block last year). </back-of-envelope> Peter Trei trei(a)process.com Disclaimer: This has nothing to do with my employer.

8 9

RE: Distributed DES crack
by Joel McNamara 17 Dec '03

17 Dec '03

I'd like to see a very general hardware processing power equivalence table. For example, 1 MasPar equals how many Pentiums. This would be extremely useful in figuring out a range of the types of boxes required to do this. Joel At 09:51 AM 7/23/96 +0100, Andy Brown wrote: >But first, a little reality check is in order. According to libdes, >the 200Mhz Pentium Pro on my desk will do 1,827,997 ECB bytes/sec, or >228,499 ECB blocks. A DES crack would have to try, on average, 2^55 >blocks. That would take my machine 43,798,875 hours, or 1,824,953 days. > >OK, so let's be reasonable and say that a week would be a good time to >come up with a DES key. We would need 260,707 200Mhz Pentium Pro's to >achieve this.

1 0