- cypherpunks-legacy - lists.cpunks.org

Cypherpunk FTP site -- where is it?
by ferguson＠fiber.sprintlink.net 12 Aug '93

12 Aug '93

Could I entice someone to remind me of where the anon FTP site (and directory) is located? I seem to recall that it was on soda.berkeley.edu but beyond that, my memory fails me....

1 0

Re: >Clipper trapdoor?
by Peter Wayner 12 Aug '93

12 Aug '93

Dave Sobel has been wondering just how the Escrow Agency will check up on the requests for keys that it gets from the NSA. The procedures aren't decided yet, but from my understanding of the presentation given at the last CSSAB meeting in Washington, the Escrow agencies won't have a person/phone to escrow id number mapping that would allow them to check if a request for a key is valid. There are two reasons for this. 1) It would be pretty useless because people could sell their phones at garage sales or give them as Christmas gifts and screw up the list. 2) This is also a "feature". If two of the escrow agents (from different agencies) decided to go bad, then they wouldn't be able to look up their enemy Bob's escrow key by name. They would need some of the real-time hardware and access to the family key. Presumably, this would be handled by a third party. -Peter

1 0

ANONYMOUS CONTACT SERVICE
by Duncan Frissell 12 Aug '93

12 Aug '93

To: cypherpunks(a)toad.com H>A few months ago, someone subscribed to H>the list through the Penet service, H>and it ended up revealing the Penet H>aliases of everyone who posted. Each H>post was delivered to that subscriber H>marked as being from the Penet alias H>corresponding to the poster. All it took H>was a parallel non-Penet subscription H>to break the anonymity provided by Penet. H> H>Has this now happened again? Funny you should mention this. Maybe someone else just subscribed via penet since my morning post to cypherpunks was bounced back to *me* from penet (as well as being posted normally). The bounce was caused because penet was reading the originating address as *my* address (not cypherpunks) and since I have been using a penet password and my message did not contain the password, it was not forwarded. I have noticed that some postings show up here from cypherpunks and some from the address of the original poster. I suppose different mailing situations cause different results. Maybe all penet users on the list should activate passwords on penet.fi. Penet may search back along the chain of addresses for any familiar address to check against it's list of password-enabled addresses. Duncan Frissell H> H>At the time, there was some discussion H>about using "an..." versus "na..." H>forms of the Penet aliases, one of which H>would avoid this revelation. Has H>that been taken care of? H> H>Hal Finney H>hfinney(a)shell.portal.com H> H> --- ~ WinQwk 2.0b#0 ~ Unregistered Evaluation Copy

1 0

Re: >Clipper trapdoor?
by David Sobel 12 Aug '93

12 Aug '93

RE>>Clipper trapdoor? Marc - You wrote: >As for the example of snooping on Hussein and his officers, you know >exactly how much legal authorization the NSA needs to conduct that >wiretap. Exactly none. Hence, it needs no paperwork to get the key >to Hussein's phone. > >How the escrow agents make the NSA prove that the keyid in question >belongs to Hussen's phone is an exercise left to the legislature :-/ As you (and others) noted in response to my posting, there is either a trapdoor *or* a "national security exception" to the warrant/escrow arrangement. As your comment suggests, NSA has heretofore been able to act unilaterally in exercising its "legal authorization" to intercept communications overseas, but with the Clipper scheme must obtain assistance from a third (and fourth?) party -- the escrow agents. I think the question you raise is a critical one -- under what guidelines will the escrow agents determine the validity of an NSA request for the key without a FISA warrant? - David

1 0

chaos harnessed for encryption
by klaus＠mail.lds.loral.com 12 Aug '93

12 Aug '93

There is an article in some paper magazine (Technology) about some research at MIT for creating a signal-processing designs that use chaos to encrypt information. Voice telephone would be an excellent use of this technology. Anyone have more info on it? -- Christopher Klaus klaus(a)mail.lds.loral.com cklaus(a)hotsun.nersc.gov

1 0

No Subject
by nobody＠shell.portal.com 12 Aug '93

12 Aug '93

At 12:25 PM 8/9/93 -0700, Timothy C. May wrote: >despite the talk of >mandatory "trap doors" in encryption systems, encryption is >fundamentally easy to do and hard to detect. (For those who doubt >this, let me describe a simple system I posted to sci.crypt several >years ago. An ordinary digital audio tape (DAT) carries more than a >gigabyte of data. This means that thhe least significant bit (LSB) of >an audio DAT recordingng carries about 8megabytes of data! So Alice is >stopped by the Data Police. They ask if she's carrying illegal data. >She smiles inocently and say "No. I know you'll search me." They find >her Sony DATman and ask about her collection of tapes and live >recordings. Alice is carrying 80 MB of data---about 3 entire days >worth of Usenet feeds!---on each and every tape. The data are stored >in the LSBs, completely indistinguishable from microphone and >quantization noise...unless you know the key. Similar methods allow >data to be undetectably packed into LSBs of the PICT and GIF pictures >now flooding the Net, into sampled sounds, and even into messages like >this... Alice better not be carrying any software that could retrieve that data. The cynic in me suggests that this scenario is just an excuse for the data police to seize any equipment or data it feels like. Besides, Alice won't be stopped by the data police, Alice will have her door kicked in by the data police and they'll take everything electronic she has, including harmless music and video. And anything electronic they find in her residence, whether it belongs to her or not.

2 1

Real time OCR
by The Phantom 12 Aug '93

12 Aug '93

--------------- cypherpunks-list #4087 (54 lines) --------------- Date: Thu Aug 12 08:40:56 1993 From: wcs(a)anchor.ho.att.com (Bill_Stewart(HOY002)1305) <wcs/daemon> Subject: Re: [uk.transport] Speed Camera with OCR > > The ability to read a car's plate & decypher the number in real time > > without any human intervention must take a bit of work. > There's a lot of automated-highway research going on; most of it has > been oriented towards radio+smartcards, which are easy to make accurate, > and while they're not totally cheap, you can palm off much of the cost > on the car-owner and use cheap equipment in the tollbooth. > Bill Stewart If anyone can dig out the April issue of 'Advanced Imaging' (I seem to have given mine away), there was an article about imaging license plates for use at border crossings. The bugs looked like they were worked out and the system was ready for action. I don't remember where it was going to be put into effect, but.. mt Matt Thomlinson Say no to the Wiretap Chip! University of Washington, Seattle, Washington. Internet: phantom(a)u.washington.edu phone: (206) 548-9804 PGP 2.2 key available via email or finger phantom(a)hardy.u.washington.edu

1 0

WIRED #4
by Sandy 12 Aug '93

12 Aug '93

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ SANDY SANDFORT Reply to: ssandfort(a)attmail.com . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Punksters, Paul Ferguson asks if the new WIRED is out yet. The answer is "yes and no." Complimentary copies were given away at a couple of recent conferences. Also, subscription copies are going out now or in a few days. Newsstand copies should be available in three weeks or so. I would appreciate any Cypherpunk thoughts, opinions, questions or raves on my article, "Intelligent Island?" which appears in this issue. It and William Gibson's "Disneyland with the Death Penalty" are the lead articles in WIRED #4. Hope you like it. S a n d y >>>>>> Please send e-mail to: ssandfort(a)attmail.com <<<<<< ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

1 0

CryptoStacker
by Kent Hastings 12 Aug '93

12 Aug '93

CryptoStacker#000# Whatever happened to CryptoStacker(s) research? Please voluntarily comply with my request for a status report, or I'll padlock your business, throw you in jail, arrest all your friends, and firebomb your compound. It is your choice. Kent - <jkhastings(a)aol.com>#000#

3 2

Re: Secure voice software issues
by norm＠netcom.com 12 Aug '93

12 Aug '93

Eric Blossom <eb(a)srlr14.sr.hp.com> says: > I have seen estimates that a straight forward implementation requires > about 13.5 million Mulitply+Accumulates / second. Most of the time is > burned up using a brute force search for the best excitation vector to > use. There is a fixed 512 entry code book, and a dynamic code book > with 256 entries (it may be 128). Each code book entry is an > excitation vector that is 60 samples long. Therefore, to evalute each one, > you have to run a 60 element vector through a 10 pole filter to get > the predicted output, then compute some measure of error. This > requires an additional difference operation that is implemented as > some kind of "perception weighting filter" (I don't remember the > details). I have been reading the PowerPC 601 manual (MPC601, The Macs of early 1994). It is dangerous to believe performance figures. They give you the world in one chapter and then take it back here and there in bits and pieces. Here is what I see however. Simple single precision floating point operations can issue one per cycle. The book mentions several floating point ops that take more than one clock in a pipeline stage. They don't mention floating multiply-add. I think one can issue each clock. I-unit instructions can issue in the same clock as floating point ops. If you do the block trick used to multiply matrices then one load is required per multiply add. All this leads to the optimistic estimate that the 50MHz machine can sustain nearly 50 fmadd's per microsecond on a 50MHz chip. Inner products are much like matrix multiply which is a benchmark where the RS/6000 (The MPC601's father) achieved nearly one fmadd per clock, and that was double precision! 128 excitation vectors each of 60 single precision loats fit in the on chip cache, but it is tight. There may be enough margin here for it to work with no special DSP. I'll be in Yosemite for a few days so I won't be able to respond immediately to comments.

3 2