- cypherpunks-legacy@lists.ogf.org

Re: .Comparing ViaCrypt and freeware.
by peter honeyman 28 Aug '93

28 Aug '93

> If you're worried about backdoors, reverse engineer it and verify > that it works as advertised. Given that the program has to largely > duplicate an existing set of source, this should be trivial. do you mean decompile it, or reverse engineer it based on the outputs? that former is tractable, but i suspect the latter is comparable to "busting" the idea algorithm. neither is "trivial" in my mind. peter

2 1

Re: Source Code NOT available for ViaCrypt PGP
by George A. Gleason 28 Aug '93

28 Aug '93

What about the idea of having someone who is known trustworthy examine the source code under nondisclosure? A passing grade from a couple of the respected members of the community here would be great for ViaCrypt's promotional efforts, if nothing else...! -gg

3 3

Cisco vulnerabilities?
by fergp＠sytex.com 28 Aug '93

28 Aug '93

On Thu, 26 Aug 93 21:17:16 -0600, L. Detweiler <uunet!longs.lance.colostate.edu!ld231782> wrote - > ===cut=here=== > > Date: Wed, 25 Aug 1993 12:56:54 -0700 (PDT) > From: Al Whaley <Al.Whaley(a)sunnyside.com> > Subject: Cisco routers > > Rumors abound that Cisco routers have a back door; that is when > a TCP port is disabled, it can still be accessed from Cisco's > IP number. > > I have personally verified this with the sendmail port. > > Al Whaley al(a)sunnyside.com +1-415 322-5411(Tel), -6481 (Fax) > Sunnyside Computing, Inc., PO Box 60, Palo Alto, CA 94302 > Sure, they have a backdoor -- it's called unsecured ports and lackidaisical security. Cisco routers don't really have "TCP" ports, per se. They have ethernet ports, or token ring ports, v.35 serial ports, and dial-up rs-232 for fail-safe configuration when some idiot drops your feed at the local rboc and you need to "look into" your net. If the "entrance" passwords are enabled properly, then I feel quite sure that this threat is minimal. However, I have learned recently that some facets of SNMP encapsulation can exploit _management_ but can not, however exploit the configuration of the router. It can add to the traffic overhead. Also, there is an additional "enable" password for configuration modification, such as changing IP addresses of the ethernet or serial interfaces (ports) and saving the configuration to NVRAM. I had a guy adamantly try to convince me the other day that the (Cisco) routers were in jeopardy because of the ability to TFTP a new (albeit, damaging) operating system directly into NVRAM (a sleight of hand), rendering the box useless. It can be done, in fact, Cisco would have to ship me a whole new box overnight if it happened, but if I mind my P's and Q's (read: adhere to proper security), he's pissin' in the wind. ;-) Cheers, Paul Ferguson | "Government, even in its best state, Network Integrator | is but a necessary evil; in its worst Centreville, Virginia USA | state, an intolerable one." fergp(a)sytex.com | - Thomas Paine, Common Sense Type bits/keyID Date User ID pub 1024/1CC04D 1993/03/15 Paul Ferguson <fergp(a)sytex.com> Key fingerprint = EE D2 93 7D 04 6D C6 05 AC 36 AD 9D 8E 4F 41 58

1 0

Re: What, Me Worry?
by Duncan Frissell 28 Aug '93

28 Aug '93

To: cypherpunks(a)toad.com Since I received 3 or 4 requests to explain my earlier signature line... C.>>"In his first three months in the White House, President Clinton C.>killed C.>>more people in the United States than during the 12 years of C.>Reagan-Bush." C.> C.>Could you expand on this? I'm curious. C.> Sorry, I was rushed. The "improved" version: During the first three months of the Clinton Administration, federal enforcement agencies killed more Americans than they killed during the entire 12 years of the Reagan and Bush Administrations. Waco! This is true because usually the Feds kill few Americans directly (fewer than 1 a month). DCF --- ~ WinQwk 2.0b#0 ~ Unregistered Evaluation Copy

1 0

Key Escrow Anecdote...
by wayner＠cs.cornell.edu 27 Aug '93

27 Aug '93

I was speaking with an international crypto consultant who told me this story about a business with an office in a South American country. They decided to start encrypting their link. Within a week, a team of soldiers burst through the door and smashed the encryption hardware to bits. On the way out, they asked, "What are you trying to hide from us?" -Peter

2 1

What, Me Worry?
by Duncan Frissell 27 Aug '93

27 Aug '93

To: cypherpunks(a)toad.com >From Thursday's WSJ news that suggests that the competition may be a tad busy to worry about suppressing strong crypto: Poor Man's Cruise Airliners Can Exploit U.S. Guidance System But So Can Enemies --------- Global Positioning Satellites Could Be Used to Direct Cheap, Accurate Missiles ---------- Defense Aide:It's a Quandry Washington--At the Pentagon, they call it "the poor man's cruise missile." It is a low-flying guided missile or robot aircraft that is relatively cheap to produce, but capable of hitting targets with a precision the U.S. once monopolized... (you can fill in the rest of the story) Reminded me of the late 60's early 70's with VonuLife and the gypsies and the troglodytes arguing that high-tech weapons would render large cities (aka target-rich environments) uninhabitable and people would have to spread out into communities too small to waste hardware on. Also, Vinge's "The Ungoverned" in which tornado-killer missiles are retargeted at the government baddies. This is an example of what I call the Sack-Full-Of-Cats-Thrown-In-The-River effect. As things get dicey for large "target-rich" institutions like governments, and their revenues drain away to the nets, internal and external struggles for fading influence begin. It ends up like a sack full of cats thrown into a river. Noisy, but fun if you don't like the cats. Duncan Frissell Who likes cats himself and could almost feel sorry for the rulers of others if they hadn't quite killed so many people. "In his first three months in the White House, President Clinton killed more people in the United States than during the 12 years of Reagan-Bush." --- WinQwk 2.0b#0

2 1

Re: Digital Coin Claim
by nobody＠pmantis.berkeley.edu 27 Aug '93

27 Aug '93

At 10:24 PM 8/25/93 -0400, Paul L. Moses wrote: >I think that creating a chain of title for digicoins would defeat the whole >purpose of the exercise. Creating a history of ownership is done in order to >increase accountability, not privacy. And such a history is necessary only to >clarify disputes over ownership, liens, defects....things which are simply not >problems with currency and bearer instruments. > >The question is, what is the end to which this chain of title is being put? >To verify that the holder is a "valid" holder? THIS IS IRRELEVANT. He who >holds, owns. Yes, but holds what? I think you missed the point. You're thinking that people use their identity-revealing keys to sign these public documents. I think Hnash intended that people use anonymous or even one-time-only keys. The bearer instrument is not the thing that gets publicized, it is the private key corresponding to the public key in the public document. When you generate that key, you put no identifying info in it. You use it only once, to transfer ownership. Hence, your true identity is not tied in to the history of ownership. This leads back to the question about settling disputes - if the keys associated with the owner of the coin do not identify him, he could attempt to use the same coin more than once. This could perhaps be circumvented by having a single registry. The merchant would wait until the registry showed her (her key) to be the owner before handing over the goods.

1 0

Re: Viacrypt PGP source code unavailable
by peter honeyman 27 Aug '93

27 Aug '93

> Yet still some people pay for software because > they feel better using a legal version. People who feel this way would > perhaps also prefer a legal version of PGP. i remind you (all) that the courts have long recognized an exemption from patent infringement liability for research and experimental use. this exemption makes my use of pgp legal, maybe yours, too peter

1 0

Matrix extensions of the rsa algorithm.
by Timothy Newsham 27 Aug '93

27 Aug '93

I have posted this before and never saw any reply. The following sounded appealing to me because it gives public key features with private key cryptosystem speeds. Matrix Extensions of the RSA Algorithm Chih-Chwen Chuang and James George Dunham CRYPTO '90 Abstract A new matrix extension of the RSA algorithm is proposed which is based on the Cayley-Hamilton theorem and a one-way function. The security of this algorithm rests on that of the RSA algorithm and the one-way function. The computational efficiency of the new algorithm depends on the dimension of the marix. The most efficient implementation is the 2x2 case in which both the encryption and decryption use a single modulo arithmetic multiplication and single evaluation of the one-way function. later in the paper.. 'Under a chosen plaintext attack on the key, the security of the new algorithm is equivalent to that of the RSA algorithm. Under a known plaintext attack on the message, the security of the system rests upon that of the one-way function.' '..theyse public key cryptosystems hvae the potential of a fast implementation.' Has anyone looked into this paper? Has anyone implemented the cryptosystem explained in it?

1 0

Re: Commercial PGP: Verifying Trustworthiness
by peter honeyman 27 Aug '93

27 Aug '93

> This means that I am trusting the "pseudo-random" stuff not to be > some secrets that PGP has read from my disk. trust? you could read the code, starting at about line 550 of crypto.c. of course, you have to trust your eyes, your editor (if you use one), and your operating system not to deceive you. (i think i've carried this too far.) > The only benefit > that I see to the pseudo-random stuff is to send the same message > to several people without revealing the fact that the messages are > the same except to those that can decode the messages. that is a big win, in my view, but the random prefix also also helps defeat chosen plaintext attacks, does it not? peter

1 0