December 2003 - cypherpunks-legacy

Where do Extropians live?
by D.C. Williams 17 Dec '03

17 Dec '03

I'd appreciate a pointer to the extropians-request address (email or post at your discretion). TIA. =D.C. Williams dcwill(a)ee.unr.edu

2 1

National I.D. Cards
by Brian D Williams 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- It would seem we are missing a point about having National I.D. cards. The rest of the world seems ready to rush in and adopt some form of "Clipper" chips to monitor their own citizens. (There doing it in the U.S.!) Do we really doubt they will adopt "The Card" as well? (don't leave home without it!) Step 2: International I.D. Cards...... Can you say "New World Order?" Sure you can....... Brian Williams Extropian Cypherpatriot "Cryptocosmology: Sufficently advanced communication is indistinguishable from noise." --Steve Witham "Have you ever had your phones tapped by the government? YOU WILL and the company that'll bring it to you.... AT&T" --James Speth -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLiWevNCcBnAsu2t1AQGhTQP/dOf1vfVoUs2wfXz70lPQ6WkwePzQCUPA prtlzzhJaJllobpM9cKUDL3x1Zx8BCV9+ebcgaZ62xDDHoGtK6K9l8RAPy/x5m5p QA+vV3Jgi0qI9UKRx4MOozaH6O2FhU6+4QgfWgE270yd4gv69cRHTEaiSB9Nfd7T KUvrt0g5b2c= =3fAd -----END PGP SIGNATURE-----

2 1

Re: National ID Card Info Request...
by frissell＠panix.com 17 Dec '03

17 Dec '03

At 09:47 AM 7/15/94 -0500, Jim choate wrote: >I would appreciate some kind sould e-mailing me the bill name which has >this in it. > >Take care. There is no bill as far as I know. At least not one very far along in the legislative process. Of course, the Health Security Act does include mandatory Worker's ID cards (your employer needs one to sign you up with your Health Alliance). DCF "The Health Security Act - the most expensive government program in the history of mankind: $1,000,000,000,000/year in total government spending $600,000,000,000/year in *new* government spending (& taxes)"

1 0

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

17 Dec '03

>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. > The grim necessity of working for a living precludes a proper answer at this time. For now though: "You can buy heroin in maximum security prisons." Controls don't always work. DCF "Eudora for Windows addict since approximately 9:05 this morning."

8 8

tran.shar
by Carl Ellison 17 Dec '03

17 Dec '03

BTW, it *was* at ftp.std.com -- but now I have it here, to e-mail to people who ask for it. (Sorry for the multiple messages.) - Carl

1 0

Re: National ID cards are just the driver's licenses on the Information
by jamiel＠sybase.com 17 Dec '03

17 Dec '03

At 10:38 PM 07/14/94 -0400, Dave Banisar wrote: >On Thu, 14 Jul 1994, Stephen D. Williams wrote: >> Do they realize how difficult it will be to get all the rural people, >> little old ladies, etc. to go along with this? >> > >Assuming that this becomes mandatory for all transactions that involve the >govt, most little old ladies would have to go along if they wanted their >medicare and social security checks. I dont see a huge liklihood of >most people wanting to give up those (or even being able to afford to). This conversation is beginning to remind me of parts of _The Illuminati_. Great paranoid fantasy of a book. Kept me enrapt for *hours*. >Dave -j -- "Blah Blah Blah" ___________________________________________________________________ Jamie Lawrence <jamiel(a)sybase.com>

1 0

Re: National ID cards are just the driver's licenses...
by frissell＠panix.com 17 Dec '03

17 Dec '03

At 08:19 AM 7/15/94 -0700, Sandy Sandfort wrote: >Over the period I've been on this list, I've seen the optimism of >various Cypherpunks wax and wane. Normally, I greatly respect Tim's >opinions. Today, though, I think Tim is reflecting an emotional response >more than a factual one. > > S a n d y > > Maybe Tim needs to leave the People's Republic of Kalifornia and go into "internal exile" in one of the United States which is less invasive. DCF "Haven't read much SF since I found the Net since it seems 'wrong' now. Left with military SF usually sited in 'other places/other times' where the lack of the Net doesn't interfere with enjoyment. Just finished "Guns of the South" by Harry Turtledove. Usual Lee vs Grant in the Battle of the Wilderness with AK-47s. Even if you don't like the Second American Revolution (Civil War) or military SF, a hell of a story. Every character in it (save those from 2014) is a historical character. Look for the cover art of Lee with an AK-47."

4 5

Key length security (calculations!)
by doug＠OpenMind.com 17 Dec '03

17 Dec '03

Tim Mays writes: >I refer readers to the sci.crypt FAQ, the RSA FAQ, or books such as >"Applied Cryptography." (Hint for those who don't want to: one time >pads (Vernam ciphers) and things like RSA with 1000-digit moduli.) > >("Enough effort" can be interpreted in a circular way to ensure the >answer is 'Yes," as a truism. This is meaningless, if "enough effort" >is impossible to achieve, as with OTPs, or is beyond the energy in the >universe. If "enough effort" is interpreted to mean theft or rubber >hose crytanalysis, all bets are off. But most people who ask the >question I cited don't mean these loopholes.) I have seen Tim posting statements to this effect many times, and because he is one of the more well respected and listened to voices on the list, I feel it important to examine this in some detail. While I agree that 1000 bit moduli in RSA is adequate protection *in all probability*, for even national security secrets, I think it is far from clear that this will definitely be true 10, or even 5 years from now. Instead of just waving vague generalities around, though, let's do some nitty gritty calculations: The people who cracked RSA-129 themselves have stated that they believe a 1024 bit modulus is at most 20,000 to 2,000,000 times more difficult to crack than RSA-129. For example, I recall Derek Atkins posting that he estimated a 1024 bit key to be 40,000 times harder than a 512 bit key, although I didn't save the posting. And Paul Leyland of Oxford posted: >RSA-129 is 425 bits; rather harder than 384-bit numbers. We estimate >that 512-bit keys are about 20 times harder than RSA-129, if a more >efficient but available algorithm is used. No-one knows how much >harder 1024-bit numbers are, but they will be no where near a trillion >times harder than 384-bit keys. Best estimates suggest that 1024-bit >numbers are about 10^4 to 10^5 times harder than 512-bit numbers. OK, so the people in the civilian world working on this today say it is possible that a 1024 bit key is only 20,000 times harder than RSA-129 *using known algorithms*. Now let's really get our hands dirty: cracking RSA-129 was estimated to take 5000 mips years. The NAL NWT 2/140 computer installed at the National Aerospace lab in Tokyo is estimated at 357 Cray YMP equivalents. I estimate this to be equivalent to 200 Gips for the purposes of this computation (this is possibly where I am most off). 5000 mips years = 1.58 X 10^17 instructions. This comes out to 9.13 days on the NAL NWT 2/140. If my estimates above are correct, scaling up to the 7400 Cray equivalent computer due to be installed 4Q95, from the 357 Cray equivalent above, we go down to 10.5 hours. This is all for the RSA-129, of course. Still sounds pretty safe so far... if it really takes at least 20,000 times as long to crack a 1024 bit modulus, then it would still take the 7400 C.E. (Cray Equivalent) computer 24 years to crack a 1024 bit number. BUT, the biggest worry is that no one knows how good the NSA's factoring algorithms are. I read recently that the NSA is the world's largest employer of mathematicians. The relative improvement in factoring algorithms since the introduction of the RSA-129 problem, to its factoring almost 20 years later, far exceed even the exponential increase in computer speed over that same period of time. (5 orders of magnitude? more?) We have no way of knowing how many orders of magnitude leeway we have, because as the moduli get larger, the factoring algorithm gets more and more important. Suppose the NSA has four orders of magnitude on us in the efficiency of their factoring algorithms. In that case, they might be able to crack a 1024 bit key as early as the end of 1995. (20,000 X 10.5)/10^4 hours = 21 hours required). Granted, this may not be likely, but I think we have to take the possibility seriously. At this point, 1024 bit keys cease to be secure for matters of critical national security (but still good for everything else). Now let's continue with our worst case scenario... suppose that computer speed doubles every 3.3 years over the next decade, and that further algorithmic breakthroughs continue to at least match this rate of doubling (not likely, perhaps, but *possible*). Then just one decade later, in 2005, the computer power of the NSA is 8 times greater, and the algorithms are 8 times faster, for a total speed increase of 64. At this point, they could crack a 1024 bit key in just 20 minutes (using all their resources), or 72 keys per day. At this point, I start to be uncomfortable trusting my security to a 1024 bit key length. So, it seems *possible*, even if by no means probable, that a 1024 bit key length is only good for the next decade or so. My intent is not to foster paranoia, but cypherpunks, of all people, should take as critical a view of key length security as possible. I suggest that people who state that the want 1200 bit or even 2000 bit key sizes in PGP be no longer ridiculed... the issue is subjective, as we have no way of knowing what the NSA's factoring algorithms are like. Doug ___________________________________________________________________ Doug Cutrell General Partner doug(a)OpenMind.com Open Mind, Santa Cruz ===================================================================

6 7

Re: National ID cards are just the driver's licenses...
by gtoal＠an-teallach.com 17 Dec '03

17 Dec '03

: I, and I'm sure Duncan, would be more than happy to discuss the subject : with anyone on the list who is interested. Let's do it offline, though, : unless there is a strong crypto tie-in. Sandy, I'd love to take part in a list for discussing practical ways of assuring privacy. I've been doing it myself for years. How about asking Robert Hayden to start up a specific mailing list for it? (Some name like 'obfusc', for obfuscating database entries etc...?) - description: 'discussions of practical real-life ways of improving personal privacy.' Also, you tend to mention things in dribs and drabs... in fact, didn't you do a sort of mini privacy-101 some years ago by posting little snippets of advice in your .sig file every post? Do you have any collected files with _lots_ of this stuff in it already that we could read? G PS Robert, did you say you could run anonymous mailing lists or was that someone else? This might be a good list to kick off as an anon list, just on principle...

3 2

Re: Triple encryption...
by DAVESPARKS＠delphi.com 17 Dec '03

17 Dec '03

Carl Ellison (cme(a)tis.com) wrote: > have you considered > > des | tran | des | tran | des ? That one's sort of your "trademark", isn't it? <g> (TRAN is really clever, BTW.) One scheme that seems to make even more sense, though, is: des | tran | IDEA | tran | des You get the benefits of 112 bits worth of DES keyspace along with 128 bits of IDEA keyspace, and thus don't stake your total security on the strength of EITHER algorithm. Other than making the code bulkier by requiring the inclusion of code for TWO crypto algorithms, and 64 bits of extra key material, what other drawbacks would there be to such a scheme (in a NON-commercial setting where licensing of the patented IDEA is not an issue)? If IDEA turns out to not be as secure as we've been led to believe, at least it, sandwiched between two layers of TRAN shuffling, should at least slow down a meet-in-the-middle attack on the remaining two layers of DES. As I recall, last time we discussed this over on sci.crypt you also advocated an additional step of "PRNGXOR". Is that still the case? Have you had the opportunity to read the Eurocrypt '94 paper by Eli Biham on triple DES modes, yet? /--------------+------------------------------------\ | | Internet: davesparks(a)delphi.com | | Dave Sparks | Fidonet: Dave Sparks @ 1:207/212 | | | BBS: (909) 353-9821 - 14.4K | \--------------+------------------------------------/

4 5