December 2003 - cypherpunks-legacy

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

RE: FW: A third voice re: science and spirit
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

Subj: FW: A third voice re: science and spirit > OOps - sorry, I sent that message to the wrong list. > > Ingore, ^^^^^^ > delete, > forget it. > > Blanc "Ingore"? Is that anything like an "AlGore"? Is that a verb to describe the state of the country, as in "we've just been INGOREd"?

1 0

Van Eck Detection
by Martin Spellman 17 Dec '03

17 Dec '03

> > To: cypherpunks(a)toad.com > > > > I've heard rumors through the years that CIA/NSA/whoever can aim a > > parabolic antenna at your window, read the electronic pulses > surrounding> your computer, and thusly determine what you are typing. > Is there any> truth to this? > How far away, and whether a van parked a block away can do it, is > unknown to me. There are people on this list who have done > this...perhaps they can comment anonymously. > > Note that the TEMPEST spec for shielding equipment is directly related > to this. In Britain we have 'TV detector Vans'. These are to detect licence evaders (you need to pay an annual licence for the BBC channels). They are provided by the Department of Trade and Industry. They use something like a small minibus and use Van Eck principles. They have two steerable detectors on the van roof so they can triangulate. But TV shops have to notify the Government of buyers - so that is the basic way in which licence evaders are detected. I read of a case on a bulletin board where someone did not have a TV but used a PC. He got a knock on the door. They said he appeared to have a TV but they could not make out what channel he was watching! Martin Spellman <mspellman(a)cix.compulink.co.uk>

1 0

Mysterious Deaths was: J.Bidzos life threatened
by Martin Spellman 17 Dec '03

17 Dec '03

> gtoal(a)an-teallach.com (Graham Toal) wrote: > Several years ago lots of British scientists died under > mysterious circumstances. They were generally described as > 'defence scientists' but what a significant number of them had > in common was work in the area of surveillance. Several of > them were connected to UK's system X. (One UK govt official was found dead in > his hotel room in an arab country while there at a trade fair > in an unofficial capacity helping to promotye system X). For those who want to follow this up a bit more there was a book published in 1990: 'Open Verdict - An Account of 25 mysterious deaths in the defence industry'by Tony Collins ISBN 07474 0146 2 Publisher Sphere Books Well worth reading if you can get hold of it. Collins picked up this story when he was a journo with 'Computer News' in England. He later became Executive Editor of Computer Weekly. It all happened in the mid 80s and most of those involved were computer programmers. 28 cases are dealt with in the book and there may have been more (or less) as much of the evidence is circumstantial and all may not have been connected. Collins reckons it has more to do with Electronic Warfare(EW) or C3i than digital exchange surveillance. Although System X was apparently a key component of IUKADGE (Improved UK Air Defence Ground Environment) an early warning system. The backbone for this was UNITER - a comms system to link over 100 USAF and RAF bases around the world. Many of the people worked for GEC Marconi and GEC Plessey (the main System X contractor). Jonathan Wash, fell from a hotel balcony in Abidjan, capital of the Ivory Coast, West Africa, on 19 November 1985, shortly after expressing a fear for his life. He was working for Telconsult, a subsidiary of BT (the telecoms transnational which is also a major defence contractor). He was helping the government there assess bids for a digital exchange contract. Some of the others that died worked for GCHQ, the junior partner of the NSA.Some of the deaths were very odd, to say the least! and some were not fully investigated as murder was not suspected. It is an inviting field for conspiracy theorists. Hilda Murrell, whose death may be linked to our nuclear power industry was also killed around this time (24 March 1984). System X was delayed for a long time and one rumour was that they had designed the basic exchange OK - the problems lay with doing all the 'other bits' that went into it.

1 0

RE: ANI numbers
by nobody＠shell.portal.com 17 Dec '03

17 Dec '03

> > The number is 1073214049889664 > 1-0732-1-404-988-9664 > a carrier code followed by an ani in the 404 area code...it > works, but it isn't free...should work just as well as a > 1-404-988-9666. ^^^ I thought that ANI only was supposed to work on 800 numbers, the theory being that since the callEE was paying the toll, he was entitled to know who was calling. So now we have to worry about ANI on non-800 numbers, too? I noticed you changed the last digit from a "4" to a "6" so that the number now ends in "666". Is this some sort of commentary on the "number from hell"? <g>

2 1

Privtool ("Privacy Tool") Beta release
by Mark Grant 17 Dec '03

17 Dec '03

Hi, recently a few people have been asking about PGP-aware mail programs for X-Windows, and I thought they might be interested in the one that I've been hacking together over the last nine months. At the moment, the Beta release is available from ftp.c2.org in /pub/privtool as privtool-0.80.tar.Z, and I've attached the README.1ST file so that you can check out the features and bugs before you download it. Currently the program requires the Xview toolkit to build, and has only been compiled on SunOS 4.1 and Solaris 2.1. I don't think that there should be any ITAR problems as there is no actual cryptography code in the program (you need a copy of PGP and ideally a copy of PGP Tools to run it), however as I'm not a lawyer and I'm in Europe anyway, if anyone in Europe is willing to make it available on an FTP site, contact me and I can upload it there. Also, if anyone wants to collaborate on a port to Xt, Motif, Windows etc, or knows how to fix some of the Xview funnies, mail me at mark(a)unicorn.com. I'm going to be off the net for most of the next two weeks, so don't expect a fast response in the immediate future. Mark Grant P.S. People wanting to integrate PGP with other programs might be interested in the pgplib.c and support files in the Privtool source, which give a high-level C interface to encrypt/decrypt/sign/verify messages, either calling PGP Tools or (on Unix) forking off a copy of PGP and examining the messages it prints out (though I haven't tested this with 2.6, only 2.3a). You're free (within the terms of the GPL) to use it in your own applications if it's of use to you (it only took me a couple of hours to add decryption/signature verification to Pine, for example). --- Privtool Beta Release @(#)README.1ST 1.9 6/31/94 ----------------------------------------------------- Privtool ("Privacy Tool") is intended to be a PGP-aware replacement for the standard Sun Workstation mailtool program, with a similar user interface and automagick support for PGP-signing and PGP-encryption. Just to make things clear, I have written this program from scratch, it is *not* a modified mailtool (and I'd hope that the Sun program code is much cleaner than mine 8-) !). When the program starts up, it displays a list of messages in your mailbox, along with flags to indicate whether messages are signed or encrypted, and if they have had their signatures verified or have been decrypted. When you double click on a message, it will be decrypted (requesting your passphrase if neccesary), and/or will have the signature checked, and the decrypted message will be displayed in the top part of the display window, with signature information in the bottom part. The mail header is not displayed, but can be read by pressing the 'Header' button to display the header window. In addition, the program has support for encrypted mailing list feeds, so that if the decrypted message includes another standard-format message it will replace the original message and be fed back into the display processing chain. When composing a message or replying to one, the compose window has several check-boxes, including one for signature, and one for encryption. If these are selected, then the message will be automatically encrypted and/or signed (requesting your passphrase when neccesary) before it is sent. Being an Beta release, there are a number of bugs and nonfeatures : Known Bugs : Message list scrollbar often set to stupid position when loading a mail file. When you save changes to the mail file, it throws away the signature verification and decrypted messages, so that the next time you view a message it has to be verified or decrypted again. 'New mail' indicator in icon does not go away if you open the window and close it again without reading any messages. Known Nonfeatures : Currently if you send encrypted mail to multiple recipients, all must have valid encrpytion keys otherwise you will have to send the message in plaintext. Also, the message will be sent encrypted to all users, not just the one who is receiving each copy. 'Add Key' button is enabled and disabled as appropriate, but does not do anything ! A number of other buttons and menu items do not work either. Passphrase is stored in ASCII rather than MD5 form, making it easier for hackers to find if you're on a multi-user machine (of course, you shouldn't be, but many of us are). Kill-by-subject does not work. Ignores Reply-To: lines, and could probably do with an improved mail-reading algorithm. Only one display window, and only one compose window. Message List window code needs rewrite. Code should be more modular to assist with ports to Xt, Motif, Mac, Windows, etc. Not very well documented ! Encrypted messages are saved to mail files in encrypted form. There is currently no option to save messages in decrypted form. No current support for remailers and pseudonyms (this will be added for the final release). Not very well tested on Solaris 2.x. Privtool can be compiled to either use PGPTools, or to fork off a copy of PGP whenever it is needed. There are also a number of different security level options for the passphrase, varying from 'read it from PGPPASS and keep it in memory' to 'request it every time and delete it as soon as possible', via 'request it when neccesary and delete it if it's not used for a while'. See the README file for information on compiling the code, and the user.doc file for user documentation (the little that currently exists). You should also ensure that you read the security concerns section in user.doc before using the program. Mark Grant (mark(a)unicorn.com)

1 0

ACAPULCO H.E.A.T.
by Sandy Sandfort 17 Dec '03

17 Dec '03

C'punks, Today's episode was, "Codename: Feminine Intuition." The Bad News: No crypto The Good News: No Fabio Gooder News: Lots of shots of Alison Armitage in swim suits. S a n d y

1 0

Password Difficulties
by sico＠hacktic.nl 17 Dec '03

17 Dec '03

Hi folks, Saturday July 02 1994 04:00, Kent Borg wrote: KB> Hey folks, passwords are hard to choose! It is indeed a problem. I hear that grady(a)netcom.com recently posted some useful info on this matter somewhere on UseNet, but I can't find it. I recall having found an interesting article sometime last year, but I lost it. In short, it said that pass phrases and such should be "shocking nonsense". The "shocking" element will help you remember it, and the "nonsense" element will make it difficult for others to guess it. KB> It boils down to this: I can't remember as many bits as the TLAs can KB> crack by brute force. My pass phrase is a couple of words long and contains deliberate misspellings and unusual capitalization. To be able to remember it, I simply use PGP every day, even if there's nothing to sign or decrypt. ;-) I'm a (long time) 2-finger typist but I have no problems with echo off. [rest deleted] CU, Sico.

2 1