- cypherpunks-legacy - lists.cpunks.org

Re: false positives
by catalyst＠netcom.com 22 May '93

22 May '93

>Has anyone tried to create a strong encryption algorithm that cannot >be broken by brute force? ^^^^^^ Brute force = exhaustive search. Therefore, if there is a solution, and the search terminates, the answer will be found. Your point was, can an encryption system be designed such that an exhaustive search yields multiple equally good, different (preferably contradictory) decryptions, for any given encrypted message. In "Communications Theory of Secrecy Systems", Bell Systems Technical Journal, Vol 28, pp. 656-715, Shannon measures the efficacy of an encryption system by the average number of plaintext messages that map to an arbitrary cyphertext (via different keys). Later, Hellman (in "An Extension of the Shannon Theory Approach to Cryptography", IEEE Transactions on Information Theory, vol 23, No. 3, pp. 289-284) emphasized Shannon's point about using compression with encryption so that decryption will yield more false positives. Ross Williams discusses this in his book: "Adaptive Data Compression". Note the limited definition of meaningful in these papers as 'makes words'. Given sufficient context, a correct decryption would not be able to hide in a forest of 'meaningul' false positives. (e.g. "Hmmmm, do you think it's 'cats often enjoy', or 'be ready by tuesday'). Of course, a one time pad affords a very large space of meaningful (much more meaningful than just 'makes words') decryptions for each encryption, hence its information-theoretically perfect security. A system which provides arbitrary mappings at the message level and no derivable component context enjoys this property as well. (e.g. a code book: 1-->be ready by tuesday; 2-->expect a guest. What does the message '1' mean? It can mean any message in the world, exactly as (when using a one time pad) the 17th character might mean any character in the world.) So in answer to your question: yes, a one time pad is just such a system. -- Scott + Scott Collins + "Few people realize what + + catalyst(a)netcom.com | tremendous power there is in one | + of these things." -- Willy Wonka +

5 4

PI Compression
by Kent Hastings 21 May '93

21 May '93

PI Compression It may have been discussed here months ago, but my favorite bogus compression scheme is "pi compression". Any number like pi or SQRT(2) can be represented as an infinite sequence of non-repeating bits (there are repetitive patterns, but eventually the sequence breaks out). A finite bit string can be represented simply as the starting location and length in pi. Since all possible finite bit strings are, by definition, contained in the unending cavalcade of bits in pi, all literary works, radio programs, TV, 3D holos, feelies, etc for all sentient species from the remotest past to the distant future, in every possible alternate universe is in little old pi. PI has been in the public domain from antiquity, therefore all parts of pi are public domain, just like every chapter, paragraph, sentence, word and bit are parts of a book. Who would dare argue against these reasonable assertions? Kent - kent_hastings(a)qmail2.aero.org.

3 2

LISTBIZ: Unsub during my abscence/USA-Legal PGP
by edgar＠spectrx.Saigon.COM 21 May '93

21 May '93

-----BEGIN PGP SIGNED MESSAGE----- Distribution: Cypherpunks Request <cypherpunks-request(a)toad.com> Cypherpunks <cypherpunks(a)toad.com> Please stop sending the Cypherpunks mailing list to edgar(a)spectrx.saigon.com (Edgar W. Swank) until further notice. I anticipate rejoining the list soon after my wife and I return from travel abroad around the end of June. (Since I'm retired I guess I can't say "vacation" or "holiday", can I?) For the general list: I would greatly appreciate it if someONE would save to E-mail me after my return any postings here or to alt.security.pgp about the progress of the USA-Legal PGP project or availability, or any response by Jim Bidzos or RSA to Phil Z's letter of early this month, asking permission to use non-standard RSAREF entry points. -----BEGIN PGP SIGNATURE----- Version: 2.2.1/EWS iQCVAgUBK/yZrt4nNf3ah8DHAQEQVwP8Cx6lG7kR4GnlRyka1B+l1BJisvcxfBGt i/YioZA0WF/4nlFOnj45vMMuPwWLfAFc/BtzjLmk3Iammiw2ZlUfs29rcF9peq+w YWjjVCePIbKE08pYiCUzaSlHT5l1II5Fl0Ex9NoMgjNfYCY2RxzGzrhlObeNZCra J+WXauK0l18= =Ipsd -----END PGP SIGNATURE----- -- edgar(a)spectrx.saigon.com (Edgar W. Swank) SPECTROX SYSTEMS +1.408.252.1005 Cupertino, Ca

1 0

Re: cypto + compression
by smb＠research.att.com 21 May '93

21 May '93

OK well if you encrypt a compressed file, there are bound to be lots more new redundencies created in the encryption process (unless it is something like ROT-13). Why not compress this AGAIN, squeezing more space out o Actually, you've got it backwards. A decent encryption algorithm, including DES, generates something with very little redundancy, and hence which cannot be compress further. $ compress </vmunix >vmunix.Z $ des -e -k foo <vmunix.Z >vmunix.Z.des des: WARNING: using software DES algorithm $ compress <vmunix.Z.des >vmunix.Z.des.Z -rwxr-xr-x 1 root 1875490 Jan 7 16:59 /vmunix $ des -e -k foo </vmunix | compress >vmunix.des.Z des: WARNING: using software DES algorithm $ ls -l /vmunix vmunix* -rwxr-xr-x 1 root 1875490 Jan 7 16:59 /vmunix -rw-rw-r-- 1 smb 794374 May 21 17:17 vmunix.Z -rw-rw-r-- 1 smb 794376 May 21 17:18 vmunix.Z.des -rw-rw-r-- 1 smb 1066555 May 21 17:18 vmunix.Z.des.Z -rw-rw-r-- 1 smb 2538235 May 21 17:21 vmunix.des.Z As you can see, compressing after encrypting *increases* the size of the file.

1 0

Dec. of Ind. < Govt. Imposted Crypto Restrictions
by Stanton McCandlish 21 May '93

21 May '93

I'd sure lend a .signature to ratify that! -- Testes saxi solidi! ********************** Podex opacus gravedinosus est! Stanton McCandlish, SysOp: Noise in the Void Data Center BBS IndraNet: 369:1/1 FidoNet: 1:301/2 Internet: anton(a)hydra.unm.edu Snail: 8020 Central SE #405, Albuquerque, New Mexico 87108 USA Data phone: +1-505-246-8515 (24hr, 1200-14400 v32bis, N-8-1) Vox phone: +1-505-247-3402 (bps rate varies, depends on if you woke me up...:)

1 0

Re: Cryptography show for ITR?
by John (FuzzFace/Fast-Eddie) McMahon 21 May '93

21 May '93

I just forwarded your suggestion to the guys who run ITR. Hopefully, it will go somewhere :-) Cheers! Fuzz

1 0

looking for distributed systems/crypto textbook.
by sommerfeld＠apollo.hp.com 21 May '93

21 May '93

At work, we're thinking of organizing an internal self-study seminar on areas related to distributed systems security. I know of Denning's textbook in the area, but would prefer not to use it for what should be obvious reasons. I also know of and have read the "Contemporary Cryptography" book by Simmons, but it seems to be a little too heavily oriented towards crypto/math as opposed to applications. Undoubtedly many of you would have opinions in this area; let me know what they are and I'll summarize to the list. - Bill

1 0

Cryptography show for ITR?
by technopagan priest 21 May '93

21 May '93

This week's NPR "Talk of the Nation/Science Friday" was on the internet. Both telephone callers, and internet "callers" were accepted. The show included the founder of Internet Talk Radio. There was a question along the lines of "how can internet mail privacy be assured?" Unfortunately, the experts only breifly mentioned public-key cryptosystems. I was wondering if they would go into a longer discussion about electronic privacy and crypto issues. I wonder if we could get a "cypherpunk" onto Internet Talk Radio to discuss cryptography and it's privacy concerns. Any ideas? -Thomas

1 0

SIGINT and TEMPEST follies
by fergp＠sytex.com 21 May '93

21 May '93

On Thu, 20 May 93 18:33:52 EDT, Bill_Stewart <anchor.ho.att.com!wcs> wrote - > This is distinctly not the case. You can take any countermeasures > you want. The precise standards are classified (some SECRET, some > CONFIDENTIAL COMSEC), so you can't find out how good the > government's abilities to eavesdrop are, or precisely what level of > protection the government thinks is necessary to protect classified > information, or how good the NSA thinks the Russians are, but as long > as you're not using classified information as your sources, you can > do anything you want. (If you're not protecting yourself *enough*, > the FCC will get on your case, but over-protection is fine.) I beg your pardon, but this is _not_ the case. TEMPEST requirements are _not_ classified and are available for public scrutiny. (You obviously do not know where to look.) I _know_ this to be true: I working in the COMSEC arena within the US military for the better part of five years. In fact, I helped design and construct the first tactically-based, X.25 packet switched, PC based, uHF interfaced network in Europe back in the early '80's using Zenith Z-248's and KG-84's. We did everything from designing the cabling pin-out's on up. > TEMPEST isn't particularly about transients or electromagnetic pulses, > it's about overall electromagnetic emissions. You're trying to separate issues that are one and the same. With the proper equipment, I can put you and an IBM Selectric (tm) typewriter on a wooden raft in the middle of Lake Superior, monitor and realize every keystroke that you make and you'd not be the wiser. This type of monitoring is easily defeated by low yield TEMPEST requirements. I've worked in this area, Bill, and have tested these _facts_. I don't care how you care to word it, it _is_ transient electromagnetic emissions. > Electromagnetic Pulses are the big fast spikes you get from nuclear > explosions (or similar slower spikes from lightning, etc.) and the > techniques you use for protection against EMP don't solve your > TEMPEST problems, and vice versa, though both kinds of protection > are some help for the other. > In my previous incarnation as a Tool of the Military-Industrial > Complex, I never saw TEMPEST expanded as an acronym in any of the > documents I read. As I stated above (and I place myself in jeopardy of being accosted by Big Brother by admitting it), I was a COMSEC flunky during my junket with the US Army. Originally, I was a 31S(ierra) and later a 29S(ierra) when two MOS's (Military Occupational Specialties, in Army jargon) were later merged. I was also a COMSEC Custodian (alternate, actually) and I can admit that I have a deep understanding of cryptographic key systems, cryptographic theory, cryptographic hardware design and integration and "BLACK" and "RED" separation principles. This type of technology does not require a rocket scientist (trust me, I've worked with them too at NASA)(oh yeah, I forgot -- I've also done contract work for Bell Labs and AT&T Secure Systems). > In the case of the Crippler Chip, however, you knew it had a built-in > wiretap when you bought it, which changes some of the reasonable > expectations about privacy a bit. This issue is one where I must disagree with you emphatically. The majority of the American public don't even know about Clipper. Hell, most of them think those AT&T commercials (Aretha singing "Freedom") are cute. Personally, I think its pretty sick, considering how they worked in collusion with Big Brother (No Such Agency) and announce that they would be immediately using this "technology" on the same day that the NIST made its announcement that Clipper would be the "weapon of choice" in their war on drugs, would-be criminals and the American public (April 16, 1993). I know the technical specifics of TEMPEST (it is an acronym, BTW) and the reason it was developed as a SIGINT (SIGnal INTelligence) standard. Cheers. Paul Ferguson | The future is now. Network Integrator | History will tell the tale; Centreville, Virginia USA | We must endure and struggle fergp(a)sytex.com | to shape it. Stop the Wiretap (Clipper/Capstone) Chip.

1 0

FOIA info
by fergp＠sytex.com 21 May '93

21 May '93

A quick addendum to my post on alt.privacy yesterday. After making a few phone calls, I found that these are the correct addresses to send requests for information under the Freedom of Information Act (FOIA): Central Intelligence Agency: Mr. John H. Wright Information and Privacy Coordinator Central Intelligence Agency Washington, DC 20505 Federal Bureau of Investigation: Federal Bureau of INVESTIGATION J. Edgar Hoover Building 9th and Pennsylvania Avenue, N.W., Washington, DC 20535 ATTN: FOIA/PA Section National Security Agency: Director, NSA/CSS 9800 Savage Road Fort George G. Meade, Maryland 20755-6000 ATTN: FOIA/N5 For those who live in The Commonwealth of Virginia, this is the address of the Richmond field office: Federal Bureau of Investigation 111 Greencourt Road Richmond, Virginia 23228 ATTN: FOIA/PA Section Paul Ferguson | The future is now. Network Integrator | History will tell the tale; Centreville, Virginia USA | We must endure and struggle fergp(a)sytex.com | to shape it. Stop the Wiretap (Clipper/Capstone) Chip.

1 0