December 2003 - cypherpunks-legacy

Alert--Admin. names escrow agents, no compromise on Clipper - 7 files (fwd)
by Mike Godwin 17 Dec '03

17 Dec '03

Forwarded message:

1 0

Crypto Regulation Reform
by rcain＠netcom.com 17 Dec '03

17 Dec '03

Mr. President, I am watching with great interest the activity with regard to cyrpto regulation and have an observation I would like to share. The following was excerpted from the Harris statement: > > The President has determined that vital U.S. national security and > law enforcement interests compel maintaining appropriate control > of encryption. Still, there is much that can be done to reform > existing controls to ensure that they are efficiently implemented > and to maintain U.S. leadership in the world market for encryption > technology. Accordingly, the President has asked the Secretary of > State to take immediate action to implement a number of procedural > reforms. The reforms are: > While I totally understand the concern here and am in sympathy with the reasoning, assuming benign adherence to the procedures, I think you are in effect jousting windmills with this attempt to control or regulate crypto. It is simply too easy to build and distribute inexpensive devices that are *truly secure*, without back doors to make it other than delusional to think that the people that we would not want to have this technology won't. A device can be made right now at lower cost than a computer modem, much lower, that could be inserted between any phone and the wall that would make it impossible, no matter what laws are in place, to tap either passively or acitively, communication that passes between two of these devices. I know how to do it, could do it and probably will just for the fun of it at least. If I can there are many others that can also. In fact I personally know several. These devices can be credit card size and even fit in a wallet. They can easily be smuggled in and will be. A black market will flourish and nothing will have been accomplished except the expenditure of a lot of futile money and creation of more crime in an inflated, lucrative market. We simply must accept that point-to-point secure communication is a part of our electronic environment and swallow the bitter pill that no matter what the valid arguments are for regulation, it is effectively not possible, so that national security and law enforcement are going to be denied, in the near future, a tool in their arsenel and will have to come up with new ways of gathering this intelligence. Please abandon this effort before we throw good money after bad and create a worse situation than we will have without it. I would like whoever processes this email to forward a copy to the following contact. > The contact point for further information on these reforms is Rose > Biancaniello, Office of Defense Trade Controls, Bureau of > Political-Military Affairs, Department of State, (703) 875-6644. Sincerely, Bob Cain -- Bob Cain rcain(a)netcom.com 408-354-8021

5 13

Clipper "Above the Fold"
by Duncan Frissell 17 Dec '03

17 Dec '03

Clipper and the Admin decision to adopt same is reported in a front page (above the fold) article in the Saturday New York Times. Usual errors about how the "backdoor" would work and about how warrants would be required to get the keys. All the usual suspects. Good placement though. DCF --- WinQwk 2.0b#1165

1 0

unsubscribe
by bgold＠tlcnet.aps.muohio.edu 17 Dec '03

17 Dec '03

unsubscribe

1 0

MAIL: tearlines, policies
by Karl Lui Barrus 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Fellow cypherpunks, Hm... I'm falling further behind list mail; the day after the security situation at Rice was fixed (~2 weeks off internet) the hard disk crashed. * About remailer policies: Try to gopher site (chaos.bsu.edu) in "Anonymous Mail"/"Remailer Policies" I can only really describe what goes on at elee7h5@rosebud, elee6ue@rosebud, and elee9sf(a)menudo.uh.edu. * About tearlines: There is no standard I'm aware of, although a quick and dirty trick is to place a single period in the first column. Most remailers pipe to /usr/lib/sendmail (and not "/usr/lib/sendmail -oi") so a single period will end a mail message. Try it before you rely on it to strip the rest of your message. I beleive Miron Cuperman (extropia remailer) invokes sendmail with -oi. * About old discontinued remailers: I remember another discontinued remailer ?(a)cs.buffalo.edu. I don't remember the name, but the student was forced to shut it down because the university said that running an anonymous remailer basically made computing resources available to non-students. * About the remailers I started/run: Remailer Fast? OpLog SysLog Subj Batch RD NL CPU Phys PGP BitB - --------- ------ ----- ------ ---- ----- -- -- --- ---- --- ---- ---------- menudo -- N SM - t1 ? Y Un H 23a ? rosebud ++/- N MQ - - - N Un M 23a ? elee9sf@menudo also accepts RIPEM encryption elee6ue@rosebud requires "digital cash" (basically random strings I made) Errors on elee9sf@menudo are forwarded klbarrus(a)owlnet.rice.edu where they are deleted. I still get mail at that address which is why I have it forwarded and not just dropped. Errors on rosebud are dropped Karl Barrus klbarrus(a)owlnet.rice.edu -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLVPSY4OA7OpLWtYzAQHDCgQAphyqkkgHtXblB1C5OlyCPZQD2/6IQ7YD FaYOHBG+NmnUMKl1bz8T9LcDKGvUKFSLW9SmI64MOqv78HF7QIXLILPG4mQ/Yn3j +zv5WyIEMofyMWUxkkWl8G/eIdCT2nB6vGNgQ8/hvhdG4DvGSpgNlwSB8itRTRwK j5DOz+wdQeM= =u1Y6 -----END PGP SIGNATURE-----

1 0

Re: how to solve this prob.
by aa377＠cleveland.Freenet.Edu 17 Dec '03

17 Dec '03

> > >There MUST be some way that the LISTSERV software can be modified >so that a user can send an unsubscribe message to the -request line >for another user. > >Take this Detweiler for example. If he forgets where to send his >unsubscribe message and sends it to the list instead, someone could send >an unsubscribe message to the proper address for him. ex: > > 'unsubscribe [ listname ] user(a)e-mail.addr' > >The system would note that the name of the person sending the >unsubscribe message ( user1 ) was different from the one who was being >unsubscribed ( user2 ) , and would, after unsubscribing user2 send a >message to user2 telling him that he had been unsubscribed from the list >by user1. ex: > > 'Dear user2, you have been unsubscribed from the Cypherpunks > list by user1. If you wish to resubscribe, send a message > containing...' > >That way, instead of the list readers bombarding the folks who send >the unsubscribe requests to the list, they could simply forward the >request to the proper place. > >Now, how do we get it implemented? > > > EEEEEEEEK! You've got to be kidding! Take this L. Detweiler guy. He sets up a script and every name that comes from toad.com gets deleted from the list. Good way to destroy the list. How many times do YOU want to resubscribe? Ken Kopin -JAFL (Just a F****** lurker) -- *** I Buy KOOL-AID Points *** |Internet: aa377(a)Cleveland.Freenet.Edu 1-499 1/3 cent each. | 500-1499 1/2 cent each. |Disclaimer: It'll never stand up 1500-? 1 cent each. | in court.

1 0

Some stuff about Diffie-Hellman (and more :-)
by rcain＠netcom.com 17 Dec '03

17 Dec '03

In the Diffie-Hellman exchange there is a well-known-prime, w, and a well-knwon-modulus, m. For those interested that don't know I think it then proceeds as follows (don't have notes in front of me so please someone correct me if I'm misremembering it) where ** is the power or exponentiation operator and % is the modulus operator: 1) Bob generates a one time random prime, b, then computes B = (w ** b) % m and sends B to Carol. 2) Carol generates a one time random prime, c, then computes C = (w ** c) % m and sends C to Bob. 3) Bob generates a session key: K = (B ** c) % m 4) Carol generates a session key: K = (C ** b) % m Carol and Bob have the same K because: K == (C ** b) % m == (B ** c) % m == (w ** (b * c)) % m >From just the knowledge of B and C a snoop cannot determine b from B, within computational reason (the root modulus being as difficult as factoring), nor c from C, and because K cannot be determined from B and C without knowing b or c, she is screwed. Now, the tutorial over :-), the question is; is there a "standard" well-known-prime, w, and a "standard" well-known-modulus, m, and if not, let's define one. I suppose that PGP uses a well known pair but they are big and not easy to hand around without going through media (I think.) When defined algorithmically they might be easier to actually incorporate in a program or a product than great big numbers. If this has not been done, I propose a simply stated algorithm for finding a "standard" w and m that will allow interoperation among all future implementations of D-H as follows: Let "standard" w be the first prime found probing from the starting point w' = n!, with a well-known n that should be small. I am not sure what n should be to generate a large enough w'. Let's just say the smallest n that generates a 1000 digit number. There is a well known primality testing algorithm by Lenstra that is pretty agreed upon by the number theory crowd (I have it coded by Lenstra and more on that later.) So, let w be the first number larger than w' that passes Lenstra's primality test. Any program or device employing D-H will have this algorithm in it somewhere for generating each session specific b and c so all we need to agree on is 1000 (or whatever is decided to be a large enough prime for all practical purposes.) I leave a "standard" for m up for discussion because I don't have the material in front of me that tells the criterion for selecting strong m's and there are some considerations. I would like it to be algoritmically defined though using standard long modulus, long integer arthmetic and some small, easy to remember number. Whatcha think? Oh, for those of you that actually code this stuff like me, I have Lenstra's long integer function package in C that I "ported" from K&R to ANSI and edited and reorganized the documentation in the process. I interacted with him in that process and it is a stable and reliable package. This was a year ago so he has most likely added to it by now but this snapshot I have is very complete and has way more than is needed to do nearly anything in crypto. And it is by Lenstra himself! A cool guy BTW. The problem: I did have to make some changes to macros and sundry things to ANSIfy it and may have introduced errors. It runs his demonstration programs that are part of the package and gives the correct results and these programs exercise a good part of it, especially the areas I had to mess with. BUT: I have not had the time to sit down and look hard at a true verification suite and he doesn't have one either. So, caveat emptor, I offer this package (and the original from which it was derived) to *one* person that can put it in a relevant ftp site. Is that you, Sameer? BTW, D-H is useless across a medium in which there can be an active snoop or spoof as I guess we call him. Whit, Marty and Ron agree as of a discussion a year ago. The spoof just has a pair of boxes and separately negotiates a session with Bob and one with Carol so that clear text passes between his pair. There is no way in theory to detect the presence of our friendly spoof. :-) I've found a solution to this that is more than sufficiently secure in practice and even theoretically secure in most practical situations. I'm not sure what to do with it. I would like to retire on it though (and get a couple "voluntary income tax" liens off my back :-) and perhaps even endow some kind of institute. Actually I worry more about being retired because of it if you get my paranoid drift. I guess that is why I'm lettin' y'all know about it here first. I am also curious about how you folks here feel about someone wanting to personally benefit financially from an algorithm/protocol invention/discovery like this but I don't want nor will get into any flame war. :-( Peace, Bob -- Bob Cain rcain(a)netcom.com 408-354-8021 "Morality is largely a rationalization of the point you happen to occupy in the power pattern at a given time. If you're a *Have-Not* you're out to *get*, and your morality is an appeal to a law higher than man-made laws--the noblest ideals of justice and equality. When you become a *Have* then you are out to *keep* and your morality is one of law, order and the rights of property over other rights." Saul D. Alinsky 1909-1972 --------------PGP 1.0 or 2.0 public key available on request.------------------

1 0

MAIL: Re: remailers revisted
by Karl Lui Barrus 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- - From a few weeks ago (recently for me :-) >Given that my understanding is basically correct, why couldn't >the remailer system be set up similarly to the way IRC is? Your system sounds great. However, don't you have to be root to run the server side of things (put it in /etc/inetd.conf)? Or the alternative is to leave a process continually running listening for connections, right? Leaving a process running isn't feasible for me, even if it forks all the time (especially now with the recent security problem on owlnet). Or is there another way that an ordinary user can pull this off? If so I'd like to hear about it and work on an idea I've had for a while. Karl Barrus <klbarrus(a)owlnet.rice.edu> -----BEGIN PGP SIGNATURE----- Version: 2.3a iQCVAgUBLVPkGoOA7OpLWtYzAQGa0wQAnh38YhoBl8fPemQRf79y44FgEVkRXHZX eNGNkNQ28Hy7aa21ni0FDViGLtauZO2akaYncL5GLEu6LYgr+pMjHThU0li16LQL ADOO8W1xUCyLu/hrNXKmlw+fQ0UoPPm8h10tTn+6D8XFzDDPGvKglRKpTkKVMHoa geMLZSbC8yI= =sBov -----END PGP SIGNATURE-----

1 0

unsub,add to announce pls
by Arlene Zeichner 17 Dec '03

17 Dec '03

Please unsubscribe. It's great but too technical for me.

1 0

Military & dependants
by nobody＠pmantis.berkeley.edu 17 Dec '03

17 Dec '03

Can American Military members or their family take copies of PGP or other encryption programs with them when being stationed at overseas bases? Aren't the overseas installations considered to be American soil while occupied, thus permitting such transfers? --- There can be only one!

1 0