December 2003 - cypherpunks-legacy

FW: A third voice re: science and spirit
by Blanc Weber 17 Dec '03

17 Dec '03

From: Bhikkhu ... I don't know if you have been around many research-type psychologists, but they can be every bit as neurotic as you imply :-) --------------------------------- This is sort of amusing: Does this make it a case of the "blind leading the blind", or "it takes one to know one"? Blanc

1 0

FW: A third voice re: science and spirit
by Blanc Weber 17 Dec '03

17 Dec '03

OOps - sorry, I sent that message to the wrong list. Ingore, delete, forget it. Blanc

1 0

IS IT POSSIBLE?
by bmorris＠netcom.com 17 Dec '03

17 Dec '03

To: cypherpunks(a)toad.com PP> can put someone on a raft in the middle of a lake plinking away on the PP> keyboard -- and discern every keystroke from the shoreline. Then the typing of a passphrase could be intercepted - in the unlikely event that one is deemed important enough to be monitored by several people with some presumably expensive equipment - ah well, I'm not going to worry about it. * RM 1.4 B0037 *

1 0

Re: Dr. Dobbs Dev. Update 1/5 July 94 & Schneier
by tim werner 17 Dec '03

17 Dec '03

> >An interesting thought hit me when reading this. The "classic" >Cray series (Cray-1, X-MP, Y-MP) all have a rather curious instruction >generally known as population count. All it does is to take a register >and count the number of one bits in it, and return that count. > ... >Just a thought. It's the only plausable use that I have yet thought of >for this instruction. Has anyone else got any ideas? This instruction would be useful in all sorts of applications. I was just wishing I had such a thing only last week. I had to write a little loop to check the number of bits set in a word. Each bit represented an action, and in my particular case it was an error if more than 1 action was requested. The loop was really a waste when you consider that it could have been done in 1 instruction. tw

4 4

NSA and CSS Computer Resources
by tcmay＠netcom.com 17 Dec '03

17 Dec '03

Earlier I mentioned the list of supercomputer installations that Gunter Ahrendt maintains and I included a few (dated) excerpts. Well, tonight I found a new version posted. I'm only including the top handful of sites. Not reported are installations in non-U.S. intelligence agencies, such as Chobetsu, SDECE, Mossad, BND, etc. They probably don't have Crays, Connection Machines, or Paragons. I'll also edit the intro material a bit. The sites are ranked in order of what I'd call "Cray equivalents," where the Cray Y-MP/1 is rated at 1. For example, the top-ranked site has the estimated equivalent of 380 Cray Y-MP/1s. --Tim May From: gunter(a)yarrow.wt.uwa.edu.au (Gunter Ahrendt) Newsgroups: comp.sys.super Subject: JUL-1994 List of the world's most powerful computing sites Date: 3 Jul 94 01:37:07 GMT Organization: The University of Western Australia Message-ID: <gunter.773199427@yarrow> 03-JUL-1994 List of the world's most powerful computing sites ============================================================= (available by fingering me, joining my mailing list or in comp.sys.super) .... Ratings are ratios to a Cray Y-MP/1 based on NASA NPB BT Size A benchmark reports. Figures prefixed '~' denote approximations usually based on comparable programs, figures suffixed '?' denote relative guesses based on Intel iPSC/860 peak Gflops ratios. The current minimum performance for inclusion is a benchmarked rating of 4 BT, or 10.56 Gflops peak for unbenchmarked systems. ... 1) 380.82 - (27-JUN-1994) [NAL] National Aerospace Lab,Chofu-shi,Tokyo,Japan, nahirose(a)asuka.aerospace-lab.go.jp 1) NAL NWT 2/140 ~357 (5.1 @ 2 cpus) 2) Intel Paragon-336 18.38 (7 @ 128 cpus) 3) Fujitsu VP2600/10 ~ 5.44 2) 196.1 - (APR-1994) [CSS] Central Security Service,National Security Agency Headquarters,Fort George G Meade,Maryland,US,postmaster(a)dockmaster.ncsc.mil 1) TMC CM-5/512 44.8 (2.8 @ 32 cpus) 2) Cray C916-1024 35.52 = 2.22 * 16 cpus 3) Cray C916-1024 35.52 = 2.22 * 16 cpus 4) Cray C916-1024 35.52 = 2.22 * 16 cpus 5) Cray Y-MP/8E-256 8 = 1 * 8 cpus 6) Cray Y-MP/8E-256 8 = 1 * 8 cpus 7) Cray Y-MP/8E-256 8 = 1 * 8 cpus 8) Cray Y-MP/8E-256 8 = 1 * 8 cpus 9) Cray M98-4096 ~ 7.01 10) Cray 3/2-64 [-4Q96] 5.73? 11) NSA SMPP-2/2M [+4Q96] 7407.05? 3) 146.41 - (10-JUN-1994) [LANL] Los Alamos National Labs,Los Alamos,New Mexico,US,iaa(a)lanl.gov 1) TMC CM-5/1056-32 [-4Q94] 92.4 (2.8 @ 32 cpus) 2) Cray T3D SC128-8/3-64 15 (12 @ 128 cpus + 1 * 3 cpus) 3) Cray Y-MP/8-128 8 = 1 * 8 cpus 4) Cray Y-MP/8-128 8 = 1 * 8 cpus 5) Cray Y-MP/8-64 8 = 1 * 8 cpus 6) Cray Y-MP/8-32 8 = 1 * 8 cpus 7) Cray M98-2048 ~ 7.01 8) TMC CM-5/1056-128 [+4Q94] 92.4 (2.8 @ 32 cpus) 4) 142.08 - (10-JUN-1994) [NSA] National Security Agency,Dallas,Texas,US,postmaster(a)dockmaster.ncsc.mil 1) Cray C916-512 35.52 = 2.22 * 16 cpus 2) Cray C916-512 35.52 = 2.22 * 16 cpus 3) Cray C916-512 35.52 = 2.22 * 16 cpus 4) Cray C916-512 35.52 = 2.22 * 16 cpus 5) Cray C916-1024 [+2Q95] 35.52 = 2.22 * 16 cpus 6) Cray C916-1024 [+2Q95] 35.52 = 2.22 * 16 cpus 7) Cray C916-1024 [+2Q95] 35.52 = 2.22 * 16 cpus 8) Cray C916-512 [+2Q95] 35.52 = 2.22 * 16 cpus 5) 112.97 - (16-JUN-1994) [CRI] Cray Research Computer Network,Eagan,Minnesota,US,root(a)cray.com 1) Cray C916-256 35.52 = 2.22 * 16 cpus 2) Cray T3D 256 24 (12 @ 128 cpus) 3) Cray Y-MP/8E-128 8 = 1 * 8 cpus 4) Cray Y-MP/8D-128 8 = 1 * 8 cpus 5) Cray Y-MP/8D-64 8 = 1 * 8 cpus 6) Cray Y-MP/8D-32 8 = 1 * 8 cpus 7) Cray M98-4096 ~ 7.01 8) Cray Y-MP8I/6-64 6 = 1 * 6 cpus 9) Cray C92A-128 4.44 = 2.22 * 2 cpus 10) Cray Y-MP/4E-64 4 = 1 * 4 cpus 11) Cray T91 [+1Q95] 4.44? 12) Cray T3D 1024A [+4Q96] 204.8 ? 13) Cray T932 [+4Q96] 142.08? 14) Cray T3D 1024B [+4Q99] 640 ? 6) 106 - (20-MAY-1994) [SANDIA] Sandia National Labs,Albuquerque,New Mexico,US,arbreck(a)sandia.gov 1) Intel Paragon-1840 98 (7 @ 128 cpus) 2) Cray Y-MP/8-64 8 = 1 * 8 cpus 7) 104.38 - (29-JUN-1994) [MSCI] Minnesota Supercomputer Center,Minneapolis,Minnesota,US,consult(a)msc.edu 1) TMC CM-5/896VU 78.4 (2.8 @ 32 cpus) 2) Cray C916/9-512 19.98 = 2.22 * 9 cpus 3) Cray T3D 64 [-4Q94] 6 4) Cray T3D 128 [+4Q94] 12 8) 98.4 - (02-JUN-1994) [ATP] Angstrom Technology Partnership,Tsukuba,Japan,sato(a)jrcat.or.jp 1) Fujitsu VPP500/32 76.8 (4.8 @ 2 cpus) 2) TMC CM-5E/128 21.6 (5.4 @ 32 cpus) 9) 83.52 - (17-JUN-1994) [PSC] Pittsburgh Supercomputing Center,Pittsburgh,Pennsylvania,US, remarks(a)psc.edu 1) Cray T3D 512 48 (12 @ 128 cpus) 2) Cray C916-512 35.52 = 2.22 * 16 cpus 10) 82.4 - (10-JUN-1994) [CALTECH] Caltech,Pasadena,California,US,goss(a)delilah.ccsf.caltech.edu 1) Intel Paragon-555 30.35 (7 @ 128 cpus) 2) Intel Touchstone Delta-513 ~28.05 (7 @ 128 cpus) 3) Cray T3D 256 24 (12 @ 128 cpus) 11) 81.55 - (21-JUN-1994) [LLNL] Lawrence Livermore National Labs,Livermore,California,US, seager(a)llnl.gov 1) Cray C916-256 35.52 = 2.22 * 16 cpus 2) Meiko CS-2/256-512VU [-4Q96] ~ 24 (1.5 @ 16 cpus) 3) Cray Y-MP/8-128 8 = 1 * 8 cpus 4) Cray Y-MP/8-64 8 = 1 * 8 cpus 5) Cray 2S/8-128 6.03? 6) Cray T3D SC128-8/3-64 [+3Q94] 15 (12 @ 128+1 * 3 cpus) 7) Meiko CS-2/1280-2560VU [+4Q96] ~120 (1.5 @ 16 cpus) 12) 77.31 - (07-APR-1994) [TSUKUBA] Uni of Tsukuba,Ibaraki,Japan,root(a)ume.cc.tsukuba.ac.jp 1) Fujitsu VPP500/30 72 (4.8 @ 2 cpus) 2) Tsukuba QCDPAX-480 5.31? 3) Tsukuba CP-PACS/1024 [+3Q96] 113.68? 13) 75.44 - (25-JUN-1994) [NAS] NAS,NASA Ames Research Center,Mountain View,California,US, jet(a)nas.nasa.gov 1) Cray C916-1024 35.52 = 2.22 * 16 cpus 2) Cray C98-128 17.76 = 2.22 * 8 cpus 3) TMC CM-5/128VU-32/256 11.2 (2.8 @ 32 cpus) 4) Intel Paragon-204 11.16 (7 @ 128 cpus) 5) IBM 9076 SP2/160 [+3Q94] ~63.8 (3.19 @ 8 cpus)

1 1

Re: Password Difficulties
by kentborg＠world.std.com 17 Dec '03

17 Dec '03

lcottrell(a)popmail.ucsd.edu writes: >I make a point of using at least one non-dictionary word in every >passphase I make. Something pronouncable? Something that follows rules of some natural language, something short that could have been a word? Good, but not the whole cigar. Last I used VMS you could get it to suggest non-word word-a-likes to use as your password. Seems terrible brute forcable in 1994. Adding a non-word to a pass phrase is like increasing the size of the dictionary, and if you only do one non-word then only *that* word picks up more bits of entropy in the phrase. Yes, there are bits in where you put the word, but the whole phrase did not become made of deep bits. But my point is really that even these often-less-good-than-they-look measures are far better than what *real* people are going to do. -kb, the Kent who wonders whether real people will ever have decent security -- Kent Borg +1 (617) 776-6899 kentborg(a)world.std.com kentborg(a)aol.com Proud to claim 31:15 hours of TV viewing so far in 1994!

2 1

2D Political Spectrum (Nolan Chart)
by Vincent.Cate＠FURMINT.NECTAR.CS.CMU.EDU 17 Dec '03

17 Dec '03

As part of a Mosaic new page: ftp://furmint.nectar.cs.cmu.edu/security/cypheressay/what-is-cypherpunk.html I put together a file on the 2D political spectrum. This helps to explain why Libertarians and the "Left wing" folks are both on this list. -- Vince ----------------------------------------------------------------------- The two dimensional political spectrum was first developed by Nolan, so it is usually called the "Nolan Chart". A person gets a score on economic issues and on personal issues. This 2D plane is rotated 45 degrees so that "left" and "right" match with the old 1D spectrum. . ./ \. ./ . \. ./ . . \. ./ . . . \. ./ LIBERTARIAN \. ./ \. . . . ./ \. ./ . \.___.___.___./ . \. ./ . . ! . . . ! . . \. ./ . . ! . . ! . . \. ./ . . . ! . . . ! . . . \. 100 \. "LEFT" ! MODERATE ! "RIGHT" ./ 100 90 \. . . ! . . . ! . . ./ 90 80 \. . !___.___.___! . ./ 80 70 \. ./ . . . \. ./ 70 60 \./ . . . . \./ 60 Personal Freedom 50 \. AUTHORITARIAN ./ 50 Economic Freedom Percentile 40 \. . . / 40 Percentile 30 \. . ./ 30 20 \. ./ 20 10 \./ 10 0 0 Personal Freedom Issues (free or government controlled?): privacy, birth control, gambling, prostitution, homosexuality, drugs, pornography, censorship (even anti-American stuff), military draft, control of radio or TV content (profanity, violence, ...), etc Economic Freedom Issues (free or government controlled?): cable companies, taxi companies, licenses for businesses in general, mail delivery, stock market, minimum wages, farm prices, prices in general, banks in general, airlines, car gas-mileage, nuclear power, oil prices, money supply, current interest rate, maximum legal interest rate, exports, imports, employee hiring and firing, etc Common Agreement on Purpose of Government: All major groups on the political landscape agree that fraud and the initiation of physical force are both bad and should be punished. The con-man, fraud, swindler, or cheat, takes by trickery and deception. The "violent criminal" is someone who physically does something to someone else or their property without permission, or threatens to do so. Everyone also agrees that a government should defend the country against foreign invasion. Issues that do not fit nicely into Nolan Chart: Sort of economic: welfare, death penalty, foreign aid Sort of personal: abortion, Calibration of 2D Political Spectrum: Someone should calibrate a set of questions that break down the population by percentiles for the Nolan Chart. This could be done by someone like Gallup or maybe as a thesis project. It would be very interesting. Fun to see scores for differed countries and also how a countries score varies over time. If anyone knows of a large set of calibrated questions, please let me know. -- Vincent Cate vac(a)cs.cmu.edu This file is ftp://furmint.nectar.cs.cmu.edu/security/political-spectrum.txt

1 0

Re: Visual Passphrases
by kentborg＠world.std.com 17 Dec '03

17 Dec '03

tcmay(a)netcom.com writes: >P.S. I'm not hung up on passphrases as a major weakness. I think theft >of keys and keystroke capturing on compromised machines are much >more important practical weaknesses. Certainly passwords are not the only threat, and the rarity of Tempest equipment at the local computer store does mean that the TLAs keep an ability to spy on anyone they can aim an antenna at, but for some folks (read: most folks) there might be much easier ways. I actually don't worry so much about RF monitoring because the natural evolution of computers is slowly killing emmisions. Particularly for battery operated equipment and equipment near OSHA inspectors. -kb -- Kent Borg +1 (617) 776-6899 kentborg(a)world.std.com kentborg(a)aol.com Proud to claim 31:15 hours of TV viewing so far in 1994!

1 0

Re: Dr. Dobbs Dev. Update 1/5 July 94 & Schneier
by norm＠netcom.com 17 Dec '03

17 Dec '03

At 01:00 1994/07/03 -0400, tim werner wrote: >This instruction would be useful in all sorts of applications. I was just >wishing I had such a thing only last week. I had to write a little loop to >check the number of bits set in a word. Each bit represented an action, >and in my particular case it was an error if more than 1 action was >requested. The loop was really a waste when you consider that it could >have been done in 1 instruction. > >tw In C, {int j... if(j & (j-1)) not_exactly_one_bit; ...}

1 0

Re: Dr. Dobbs Dev. Update 1/5 July 94 & Schneier
by tim werner 17 Dec '03

17 Dec '03

>Date: Sun, 3 Jul 1994 01:24:39 -0700 >From: norm(a)netcom.com (Norman Hardy) >At 01:00 1994/07/03 -0400, tim werner wrote: >>... I had to write a little loop to >>check the number of bits set in a word. >>... The loop was really a waste when you consider that it could >>have been done in 1 instruction. >In C, {int j... if(j & (j-1)) not_exactly_one_bit; ...} I knew if I subscribed to this list I'd eventually pick up something useful. I already asked a couple of people at work if they knew of a trick to see if there was just one bit set, and no one did. Otherwise I would feel extra stupid. :-) tw

2 1