- cypherpunks-legacy - lists.cpunks.org

Colorado Cypherpunks meeting
by nates＠netcom.com 17 Dec '03

17 Dec '03

2 1

Re: stego
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

Jim Choate's comments on steganography having problems with images that are too complex or too simple were interesting. Obviously, cartoon-like GIFs aren't a good target, though scanned real stuff may be fine. Weather maps cna be good - back when I worked with the things, I found you could really see about5-6 bits worth of depth, and after that it didn't usually look much different - we stole one or two values from the color-map to draw lines on the satellite images to add state boundaries, various data values, etc., but could have stolen the LSB and maybe 7th bit without major loss on cloud-image pictures. (Radar pictures, on the other hand, were almost all black, with one or two other pixel values, compressed to 2% of original size, and would have been useless for hiding anything in.) Bill

2 1

Re: Dorothy Denning
by Black Unicorn 17 Dec '03

17 Dec '03

-> Also, I've gotten a message through to Ross Perot about our fight against Clipper, and how to reach me, so if I get a response, I'll let you guys know. Anthony <- Oh god, that is all the cause needs. I can see it now: "Transistor brain Ross Perot opposes it! It must be evil!"

2 1

Large Block DES Newsletter
by ritter＠cactus.org 17 Dec '03

17 Dec '03

Large Block DES Newsletter Vol. I, No. 1 Feb. 28, 1994 Terry Ritter, Ed. Current Standings for the Large-Block DES Proposals: I. NxM DES: A B v v k1 -> DES1 k2 -> DES2 v v C D Exchange Right 4 Bytes E F v v k3 -> DES3 k4 -> DES4 v v G H Falls to meet-in-the-middle like double-DES. Falls to a practical attack by Biham, now called "fix-in-the-middle." II. NxM DES Found Weak Announcement of above. III. Isolated Double-DES 2x construct found weak in original article. The 1x construct: A v k1 -> DES1 v B v km -> XOR v C v k2 -> DES2 v D was found weak by Chris Dodd <dodd(a)csl.sri.com> who pointed out that two different blocks of known-plaintext (A,D) and (A',D') will allow matching (B XOR B') and (C xor X'). (This is similar to Biham's "fix-in-the-middle.") Good going Chris! Also found by Stefan Lucks <lucks(a)namu01.gwdg.de>. IV. Ladder-DES A B | k1 | v v | XOR <- DES1-----| | | | k2 | | v v |---- DES2 -> XOR | | | k3 | v v | XOR <- DES3 ----| | | | k4 | | v v |---- DES4 -> XOR | | v v C D Joseph C. Konczal <jkonczal(a)nist.gov> points out that the construct is indeed vulnerable to meet-in-the-middle. I agree, but note that this seems to imply a 112-bit search. Since we don't need more than 112 or 120 bits of strength, I don't see it as a problem. (Indeed, if we could get more strength, we might want to trade it for speed anyway.) 112 bits (or so) is the design goal, which should be enough for a couple of decades. In a normal cipher design, I would expect each key bit to contribute toward strength, but these are hardly normal cipher designs. Especially when we try to expand block size, extra keys may simply provide another small block with the same strength as a previous small block. Keys will be delivered electronically, so the relatively rare delivery of 2x or 4x or even 8x the expected key material should not pose a serious problem. However, Biham reports: "ladder DES is not more secure than 2**88 steps and 2**64 chosen plaintexts." Now, 2^88 cipherings is 2^32 times as strong as the 2^56 currently in DES (and larger than Skipjack), but hardly the 2^112 intended. For the current design the current options are: 1) live with the 2^88 strength (so far!), 2) design the rest of the system to prevent chosen plaintexts, or 3) prevent more than, say, 2^32 block cipherings under a single key. Actually, we need to know exactly what the problem is, and the limits of it, before we can propose a fix, or decide whether the ladder-DES scheme is unfixable. Summary Three substantially different constructs proposed; of these, two fall, and one is wounded. To review, the intent is to find some relatively-simple construct which builds on the assumed strength of DES to deliver wide blocks and something like 112 bits of strength, with less processing than triple-DES. (I see no need for super-strength, unless it is free.) We still do not know whether or not this is possible.

1 0

Fax / Modem question
by fnerd＠smds.com 17 Dec '03

17 Dec '03

(Please respond to this question by private email.) Is it possible to set up most fax/data modems to be ready to receive calls from either fax machines or data modems? If so, in this mode, does it make the beep that sets a calling fax machine going, or does the caller have to push a button to initiate the connection from his or her side? Ditto for data. Doesn't the answer side usually put out the first beep? -fnerd please reply to: fnerd(a)smds.com - - - - - - - - - - - - - - - blue mass, Pharm. a preparation of metallic mercury with other ingredients, used for making blue pills. -----BEGIN PGP SIGNATURE----- Version: 2.3a aKxB8nktcBAeQHabQP/d7yhWgpGZBIoIqII8cY9nG55HYHgvt3niQCVAgUBLMs3K ui6XaCZmKH68fOWYYySKAzPkXyfYKnOlzsIjp2tPEot1Q5A3/n54PBKrUDN9tHVz 3Ch466q9EKUuDulTU6OLsilzmRvQJn0EJhzd4pht6hSnC1R3seYNhUYhoJViCcCG sRjLQs4iVVM= =9wqs -----END PGP SIGNATURE-----

1 0

Re: Dorothy Denning
by fischer＠charybdis.prl.ufl.edu 17 Dec '03

17 Dec '03

> I recall a profile of the Dr's Denning in one of the > popular cracker/phreax exposes, (maybe in Markov/Hafner?) > > Andy I believe you are referring to Bruce Sterling's "The Hacker Crackdown", interesting book because Sterling sympathetically presents many sides: paranoid telco's, testosterone-poisoned adolescent hackers, hand-wringing cops (Dr. Denning is presented in this milieu) and (my personal favorite) the "information wants to be free" libertarians. Randy Fischer

1 0

Applied Crypto
by fhalper＠pilot.njin.net 17 Dec '03

17 Dec '03

Does anyone know if Project Guttenberg(or anyone else) has electronic books such as applied cryptography? Thanks, Reuben Halper

3 2

PGP on the HP 100
by smb＠research.att.com 17 Dec '03

17 Dec '03

Has anyone tried putting ViaCrypt PGP on the HP-100? Is anyone but me crazy enough to contemplate it? How abysmal is the performance? --Steve Bellovin

2 1

Re: Insecurity of public key crypto #1 (reply to Mandl)
by Eric Johnson 17 Dec '03

17 Dec '03

: >Only "sensitive" information is worth encrypting and if you want to : >stay out of trouble you don't go giving sensitive information to : >someone you don't know. : : If part of your communications are encrypted and part are not you have : sent the message about what information is sensitive and what is not. : This difference in encoding is a fir-class message in it's own right. : : Therefore _all_ communications should be encrypted at all time. It is : no argument against the principle that this is difficult to do at the : current time. : : Eric : Would it not make sense, therefore, to publish a public cypherpunks mailing list key, which is returned with subscription requests? All incoming message cleartext to the mailing list server would then be encrypted in the server's key; not for security, but precisely for the reason you state above. That _would_ create quite a volume of encrypted communications to each receipient of the list. --Eric

3 3

Re: The answer.
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

> I think I have found the way to end all our problems: ESP. No good - the KGB has top psychics working for them, and the CIA has been working for years to keep up. I read it in the newspaper in the supermarket, so it must be true! Bill

2 1