December 2003 - cypherpunks-legacy

Dogwash (sorry if this isn't the proper procedure...
by m00012＠KANGA.STCLOUD.MSUS.EDU 17 Dec '03

17 Dec '03

for posting anouncements...but... ******************************** *Post-Preliminary Announcement:* ******************************** Inspiration: Dogwash was inspired by (some might say stolen from) Bruce Schneier's Blowfish. DogWash: A 128 bit block encryption algorithm/fiestal cipher, product encryption algorithm.... Rounds: The number of rounds can be defined at runtime. The range for the number of rounds is from 0 to 6140. The speed of the algorithm is correlated to the number of rounds used, with 16 rounds being reasonably fast. Keylength: The maximum keylength is determined by the number of rounds. For rounds=0, the maximum keylength is 16 bytes, or 128 bits. For rounds=6140, the maximum keylength is a ridiculous 49136 bytes, or 393088 bits long. For a standard 16 rounds, the maximum keylength is 144 bytes, or 1152 bits. Subkeydata: 64k bytes of high entropy (7.95 bits/char) subkey data are included in a header file. The subkey data is mutated with the key, and the result is used for encryption/decryption. Note that the actual amount of subkey data used is a function of the number of rounds requested. The subkey data is not fixed and may be changed so long as your correspondent is using the same subkey data. (Note: subkey data is later mutated with the session key. I only point out that the subkey data may be changed for those who might want to do that, for whatever reason.) Sourcecode: The program is written in c++. Currently, it requires a compilier that allows a 64 bit unsigned integer addition, %(1<<63). The use of destructors helps ensure that mutated subkey data is automatically destroyed after an encryption or decryption session. However, other risks exist, such as memory being swapped to disk, or the final programmer forgetting to protect his/her key. It should be very easy to convert it to standard C. Description: A PostScript file containing block diagrams and a well written (well, probably better than this) description may become available within the next few months. Security: I believe that DogWash is practically uncrackable, but I haven't the credentials to make such a pronouncement. (The only code I have cracked is the Sunday paper's Cryptogram.) But, unlike blowfish, this is not a fast encryption algorithm.

1 0

Crypto Chip in Eastern Europe
by adam.philipp＠ties.org 17 Dec '03

17 Dec '03

Does anyone know any more details on the crypto chip listed in the HTML http://jep.pld.ttu.ee/? It seems that this is a hardware PGP (uses RSA to encrypt IDEA key) project. They mention having prototype available RSN, as in December... or is this just some very interesting vapor-ware? Adam Philipp -- PGP Key available on the keyservers. Encrypted E-mail welcome. Sub rosa: Confidential, secret, not for publication. -Black's Law Dictionary

1 0

NYT on MS Network
by John Young 17 Dec '03

17 Dec '03

Laurie Flynn writes today twofer on MS Network and newbie 1/3 pres Robert Herbold. For combo send blank message with subject: NET_puf

1 0

DNA solution to Hamiltonian circuit?
by rishab＠dxm.ernet.in 17 Dec '03

17 Dec '03

srctran(a)world.std.com (Gregory Aharonian): [on Internet Patent News Service] Scientist uses DNA sequences to solve Hamiltonian path problem of combinatorial mathematics, a precursor of the PTO's headache of including biotechnology in it software prior art searches. Think of Hopfield's paper on using neural nets for the traveling salesman problem to predict where DNA computing will end up. Uhh! This was in one of Greg's 'random list of story titles' - he's yet to provide details. As Hopfield didn't really 'solve' the TS problem, but made it easier to solve a class of maps, this may not mean that there will be any significant effect upon Cypherpunk tech based on NP-hard graph problems (such as Zero Knowledge proofs) - but it would be interesting to know _what_ it's all about. ----------------------------------------------------------------------------- Rishab Aiyer Ghosh "Clean the air! clean the sky! wash the wind! rishab(a)dxm.ernet.in take stone from stone and wash them..." rishab(a)arbornet.org Voice/Fax/Data +91 11 6853410 Voicemail +91 11 3760335 H 34C Saket, New Delhi 110017, INDIA

4 3

Estonian RSA chip
by rishab＠dxm.ernet.in 17 Dec '03

17 Dec '03

adam.philipp(a)ties.org (Adam Philipp): > Does anyone know any more details on the crypto chip listed in the HTML > http://jep.pld.ttu.ee/? It seems that this is a hardware PGP (uses RSA to > encrypt IDEA key) project. They mention having prototype available RSN, as > in December... or is this just some very interesting vapor-ware? Jyri Poldre (jp(a)jep.pld.ttu.ee - probably on the list) was talking about this in May, while working on the design; we had a little discussion on the application of RSA patents in Estonia. The description of the hardware was pretty detailed back then for vapour-ware. ----------------------------------------------------------------------------- Rishab Aiyer Ghosh "Clean the air! clean the sky! wash the wind! rishab(a)dxm.ernet.in take stone from stone and wash them..." rishab(a)arbornet.org Voice/Fax/Data +91 11 6853410 Voicemail +91 11 3760335 H 34C Saket, New Delhi 110017, INDIA

1 0

Re: I Like ASCII, not MIME and Other Fancy Crap
by amanda＠intercon.com 17 Dec '03

17 Dec '03

> Now you're expecting me to hunt around for > viewers for .gifs and TIFFS and JPEGs and God knows what else you might want > to send me ? Actually, I'm not, any more than you're expecting me to hunt around for a copy of PGP just so I can verify your signature. Secondly, having to hunt around for viewers is something that has to do with your mailer, not MIME itself. For example, when I receive MIME email with images, they just appear inline with the text. Just because the easiest way for mailers like ELM to support MIME is to call out to metamail or the like doesn't mean that's necessarily how MIME was meant to be used... > .GIF is not part of the standard for the format of Internet email, is it ? Yes, it is. The MIME RFC specifies image/gif and image/jpeg as the standard formats for including images in Internet email messages. > I only switched to ELM a few months ago. I guess I'm actually getting pretty > comfortable with using it, which means it's time to ditch it. That's the problem with depending on UNIX :) :) Just as a reminder, I'll point out that I do in fact sympathize, which is why I refrain from using graphics, styled text, HTML links, and so on in my messages. I just have a strong disagreement with the sentiment expressed by Tim, where limiting email to what can be punched onto Hollerith cards is seen as a feature, not a bug. On the other hand, since designing UIs for Internet software is how I make my living, I'm no doubt biased in favor of it being a bug. Amanda Walker InterCon Systems Corporation

1 0

Re: Here's one for laughter
by Jonathan Cooper 17 Dec '03

17 Dec '03

> I know you don't need a lecture at your age---I was there recently. > But, what you have done for this administrator of your school sounds > quite illegal and most definitely immoral. Morality and legality mean one thing when your future is assured, and an entirely different one when your getting into college may depend upon the very people whom you would piss off by refusing. > Have you ever considered quitting your job to remove the legal and > moral questions surrounding the action, bringing the action to the > attention of the local media or helping the FBI nail the bastard. No. I'm a student, not a netadmin proper. And I *REALLY* don't need the headaches involved with getting the FBI involved, even if I thought it was a good idea, which I don't. I am extremely loathe to use the "authorities" on anyone, anytime. > Law related to communications) would do this administrator good. If And would do me extremely bad, unfortunately. > questionable sounds like grounds for an investigation. Threatening a > student to do the dirty work, makes me want to see this person of > power squirm like the dog he is. He is a wee bit of a twit, but on the other hand I would like to go to college and most of what I do is not reflected in my grades (2.3 GPA, 1370 SAT) for reasons which I won't go into here. Recommendations from the technical people at school are about all I've got, and I will do what I have to in order to get them. -jon ( --------[ Jonathan D. Cooper ]--------[ entropy(a)intnet.net ]-------- ) ( PGP 2.6.2 keyprint: 31 50 8F 82 B9 79 ED C4 5B 12 A0 35 E0 9B C0 01 )

1 0

Re: PGP .DLL
by Jonathan Cooper 17 Dec '03

17 Dec '03

> How do you plan to prevent passwords, plaintext, etc. being saved > in swap files, etc.? Munge the memory where they've been stored. There is really no way to protect absolutely, especially on something as unsecure as a PC where someone malevolent could just install a keyboard logger... It comes down to a "how-much-is-enough" question. -jon ( --------[ Jonathan D. Cooper ]--------[ entropy(a)intnet.net ]-------- ) ( PGP 2.6.2 keyprint: 31 50 8F 82 B9 79 ED C4 5B 12 A0 35 E0 9B C0 01 )

1 0

Re: School Admins
by Jonathan Cooper 17 Dec '03

17 Dec '03

> You definitely shouldn't be doing that! Unless you make it very clear to > the users that they should not have any expectation of privacy when using > the school's computers, then e-mail is covered by the Electronic Like I've told other people, it's easy to be objective when the rest of your life isn't on the line. But my college plans _DIRECTLY_ involve these people, and I can't afford to piss them off. And yes, the students now know that they're being logged, and I've taught them to use cheesy crypt(1) encryption to handle it. -jon ( --------[ Jonathan D. Cooper ]--------[ entropy(a)intnet.net ]-------- ) ( PGP 2.6.2 keyprint: 31 50 8F 82 B9 79 ED C4 5B 12 A0 35 E0 9B C0 01 )

1 0

Re: PGP .DLL
by Jonathan Cooper 17 Dec '03

17 Dec '03

> I'm way behind on messages currently, so maybe somebody mentioned > this... but have you looked at RSAREF? I know the license is > restrictive, but take a look at it architecturally. That is basically the way that I'd envisioned it functioning. Details have to be nailed down... anyone who is interested in helping, please mail me so we can take this chatter off the main cypherpunks list. -jon ( --------[ Jonathan D. Cooper ]--------[ entropy(a)intnet.net ]-------- ) ( PGP 2.6.2 keyprint: 31 50 8F 82 B9 79 ED C4 5B 12 A0 35 E0 9B C0 01 )

1 0