- cypherpunks-legacy - lists.cpunks.org

Re: Hiding conventionally encrypted messages in PGP messages to someelse.
by Paul Franklin 17 Dec '03

17 Dec '03

In article <199409030238.AA18130@xtropia> you wrote: > -----BEGIN PGP SIGNED MESSAGE----- > I have been thinking about steganography lately. Correct me if I am > wrong, but it seems to me that if one wants to hide encrypted data, then > all this public key encryption stuff becomes irrelevant. It seems that > the sender and the recipient must agree on a way to hide the data. The > time of this agreement is a perfect time to exchange conventional > key(s). > Speaking of conventional encryption, PGP uses conventional > encryption (IDEA). So if we wish to hide conventionally encrypted > data, why not use the purloined letter method, and hide it as the > conventionally encrypted data in a PGP encrypted file? > To create such a file, we would simply create as PGP usually does, > except that we specify or record the conventional IDEA key used. Then to > decrypt the file, we simply ignore the RSA headers and use the specified > or recorded conventional IDEA key. We could even insure that the IDEA > key in the RSA encrypted headers is wrong. So, obiwan can not reveal > the data even if Darth can seize him. > I have created a hack to PGP ui to do all of the above! Isn't this what pgp -c does?

2 1

Re: http://digicash.support.nl = NULL
by sidney＠apple.com 17 Dec '03

17 Dec '03

[Harry Bartholomew wrote about problems accessing digicash web page via lynx.] It looks to me like the web page requires Mosaic forms support. I could see a little bit more than you did using MacMosaic, which shows the pictures but doesn't support forms. It was a bit confusing, since I could see a number of entries for things with prices next to them, such as $5 for the e-cash software and $0.25 for "information". I couldn't tell how you are supposed to get either of those without either the software or at least the information. Maybe someone who has X or Windows Mosaic can tell us more about what's there. I sent e-mail to the info e-mail address that the page mentioned and got back the following. (I sent just the word "info" in the subject and body of the message.) What I find strange about it is that it looks like an automated reply, but it is asking me to reply as if the mail will be read by a human. Another interesting thing about it is that Chaum does not include consumers in his list, just people who are interested in selling products or clearing money. After reading all of the copies of the digicash press announcement that were sent to this list, I'm a bit hesitant to post this, but what the heck. Following is the reply I got back from info(a)digicash.nl: [begin quote] Date: Tue, 31 May 1994 14:51:25 CET From: "DigiCash Information" <info(a)DigiCash.nl> To: sidney(a)apple.com Subject: ecash-info Thank you for your interest in our electronic money technology. Because of all the reactions we have received, we have made available some background information on our Web server http://digicash.support.nl If you prefer, we would of course be pleased to provide similar information by electronic or paper mail. In any case, we plan to be keeping you informed of developments by email. It would be helpful for this purpose, if you could let us know more about the nature of your interest, and at least which of the following categories fits best: (a) potential acceptor of electronic cash for services offered over the network, (b) potentail provider of electronic cash service itself to other network service providers, (c) member of the press, or (d) interested in the technology for other purposes. We are looking forward to finding a way that we can work together. Very kind regards, David Chaum Managing Director [end quote]

1 0

I back your scratch...
by fnerd＠smds.com 17 Dec '03

17 Dec '03

So, the reason I asked about "orthogonal checksums" was this: If you can check the integrity of a copy of a file remotely, then people can trade storage space and have remote backups. Or have more complicated schemes where, say, a group of users pay for a common storage service by backing up some of the files (or other files) for the provider. The nice thing about this is that storage space is something that's immediately valuable and electronically tradable. And its value can increase by trading. In other words, we can trade one meg for one meg and both come out ahead in terms of safety. "I back your scratch and you back mine." I'm not sure storage space makes a good unit of value for anything else, since the cost of storage keeps going down. On the other hand, trading space for space is immune to that inflation. Some concrete applications: o Making bulletin boards' data safe from loss by vandalism o Archiving a mailing list -fnerd - - - - - - - - - - - - - - - nowhere man has gone before -----BEGIN PGP SIGNATURE----- Version: 2.3a aKxB8nktcBAeQHabQP/d7yhWgpGZBIoIqII8cY9nG55HYHgvt3niQCVAgUBLMs3K ui6XaCZmKH68fOWYYySKAzPkXyfYKnOlzsIjp2tPEot1Q5A3/n54PBKrUDN9tHVz 3Ch466q9EKUuDulTU6OLsilzmRvQJn0EJhzd4pht6hSnC1R3seYNhUYhoJViCcCG sRjLQs4iVVM= =9wqs -----END PGP SIGNATURE-----

1 0

Sternlight infested version of pgp26
by Vincent.Cate＠FURMINT.NECTAR.CS.CMU.EDU 17 Dec '03

17 Dec '03

The pgp26-RSAREF_3.0_beta.tar.gz file is not real. From some other post, it seems this came from the site in Italy that I mentioned awhile back. They have removed the file from there. It is a lot of bytes for a joke. It should probably be removed from soda too. -- Vince > pwd /alex/edu/berkeley/soda/pub/cypherpunks/pgp/pgp26 > ls -NOT_FOR_EXPORT pgp26ui-src.tar.gz pgp26uis.zip pgp26-RSAREF_3.0_beta.tar.gz pgp26ui-src.tar.gz.sig pgp26uix.sig pgp26.zip pgp26uis.sig pgp26uix.zip > gunzip -c < pgp26-RSAREF_3.0_beta.tar.gz | (cd ~/tmp; tar xvf -) gunzip: stdin: not in gzip format > strings pgp26-RSAREF_3.0_beta.tar.gz WANTED FOR NET.TREASON: |\/\/\/\| | | | (o)(o) C _) OFFICER David Sternlight | ,___| Net.Police | / a.k.a. PROUD _0--------/\/----\/\-------0_ DICK OF / /0 \----/ 0\ \ THE INTERNET / | \ \||/ / | \ / /|STERN || POLICE|\ \ _/_____/ ||LIGHT] || | [#] || \_____\_ |_______/ |\____/ || \__*_/| \_______| > ll pgp26-RSAREF_3.0_beta.tar.gz -rw-r--r-- 1 alexsrvr 567791 May 25 00:20 pgp26-RSAREF_3.0_beta.tar.gz

1 0

PGP 2.3 vs 2.6 outside the US
by rishab＠dxm.ernet.in 17 Dec '03

17 Dec '03

Jeff Barber <jeffb(a)sware.com>: > While creating a 2.6-like version from 2.3a seems a worthy goal, this > supporting argument is flawed. The original PGP was written in the USA > and, never having received the proper export approvals, must have been > "illegally exported." Isn't Phil Zimmerman being "investigated" by a > grand jury for this even now? So, it would seem to me that a bulletin > board carrying any version of PGP holds illegally exported software (wrt > US law). How does 2.3a differ from 2.6 in this respect? 2.0 to 2.3a were developed outside the US, released in New Zealand and the Netherlands, and _imported_ into the US. While a version of 2.3a in India, say, need not have come from the US, and hence does not indicate an ITAR violation somewhere, the version of 2.6 in Italian sites got there only because they were illegally exported from the US.

1 0

NSA breaks Russian PRNGs with neural networks??
by rishab＠dxm.ernet.in 17 Dec '03

17 Dec '03

tcmay(a)netcom.com: > An interesting article by Seymour Hersh is cited below. It says that > NSA had transcripts of the 1991 coup plotters (and presumably other > Russian leaders) and that Bush passed these on to Yeltsin to warn him. A recent article from the Daily Telegraph, another British paper, went on about the possible encryption techniques used by the Russians. It described how reused one-time pads led to the unmasking of Fuchs, the Rosenbergs, Philby et al. Then it suggested that the method the NSA broke was based on (presumably weak) PRNGs, a stream cipher. It suggested that the NSA might have developed techniques to find patterns in PRNG outputs through neural networks, or genetic algorithms!! While the latter sounds like crap to me, even though I've worked with and believe in the power of neural networks for amazing pattern recognition, unless the PRNGs were _really_ weak, I'm skeptical. I don't think the Russians are fools, and in these times one doesn't rely on secret weak algorithms for crypto, not when there are publicly well known strong ones. Humint? Maybe. The Atlantic Monthly will carry a story on the NSA-Yeltsin thing in June. It would be nice if someone posts it here. -------------------------------------------------------------------------- Rishab Aiyer Ghosh rishab(a)dxm.ernet.in Voicemail +91 11 3760335; Vox/Fax/Data 6853410 H-34C Saket New Delhi 110017 INDIA The National Fairness to Game Animals Association says: Support your right to arm bears! --------------------------------------------------------------------------

1 0

Cypherpunks' Electronic Book 4
by Gary Jeffers 17 Dec '03

17 Dec '03

Cypherpunks' Electronic Book (CEB) 4 Martin Janzen suggests using World Wide Web (WWW) to support & distribute the CEB. I am not at all familiar with WWW but hopefully its good. My concern with this distributed organization is that CEB could be controlled by the Cypherpunks' List & that all additions & changes to CEB would be refereed. No unauthorized changes would be made. As for updates, the point of CEB is to be made up of the most recent knowledge. People concerned with an area of electronic privacy & making contributions to the CEB would naturally update the CEB whenever they ran into new developments. People who wanted to read the CEB would be assured that by the nature of the CEB, what they found there would be the latest & the current best. Martin suggests that the CEB could be on several servers. I would suggest, then, that they be on different servers by chapter. Referees who are interested in particular chapters could maintain & administer that chapter. This could mean very little administration by Eric Hughes (cringe). For those who are kindly offering their archives of Cypherpunks' list, they are not needed now, but if & when the CEB gets established, it would be good to have your names on the CEB as archivists. Yours Truly, Gary Jeffers

1 0

Re: IMP (was Re: ecash-info (fwd))
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

> > 1.) Chaum's e-cash coupled with WWW/Mosaic is a de facto internet > > mercantile protocol. It ain't de facto until a lot of people are using it. I suspect PGP with credit card numbers in free-form ASCII is almost as common, at least for now; there are 4 or 5 groups I've seen that will accept that. And phone calls or faxes to the number on the bottom of the ad on Usenet probably outmumber those... In particular, digicash isn't a standard until there's at least one bank a lot of people can access via digicash. However, it's certainly a nice approach. Credit cards do take care of one of the objections Hal mentions, which is how you can trust your vendor to ship you the goods instead of absconding with your cash. Some of the digicash protocols can help with that process, at the cost of privacy - the spender can reveal the number of the coin that was ripped off, but the receiver may have cashed it via anonymous remailers, making it difficult to prove who's telling the truth in court, especially when the goods were software delivered by anonymous remailer chains as well. Bill

1 0

Re: PGS bug! (fwd)
by ER CRAMER 17 Dec '03

17 Dec '03

I tried to reply to this messages but it was bounced back to me... So I mail it to this list... > > I like your PGS shell a great deal. Thank you for writing it. > > Thank you for liking it ;-) > > > However, I have found what appears to be a small bug. I do not have my > > keyrings in my pgp directory. They are in a separate directory, specified > > via the PubRing, SecRing, and RandSeed directives in 'config.txt'. When I > > load PGS, it checks my pgp directory for these files, does not find them, > > and asks if I want to create a secret key ring. When I answer no, the > > program generates a General Protection fault #13, which is gracefully > > caught and handled by QEMM386, my memory manager. > > We don't support keyrings that aren't in the PGPPATH (at least not yet). > To tell you the truth I didn't know of hte PubRing, SecRing and RandSeed > directives... I guess I can take a look at it, maybe it's easy to change... > > If you answer NO on the question to create a keypair PGS should quit. I have > know problem with QEMM myself. What version of QEMM do you use??? > BTW: I tried out the PubRing and SecRing directives in my config.txt but they where nog supported... Does anyone knows something about it??? ... If you outlaw Privacy, only the Outlaws will have Privacy! Eelco Cramer <crame001(a)hio.tem.nhl.nl> ------ --------------------------------------------------

1 0

Clipper in patent trouble?
by Mike Ingle 17 Dec '03

17 Dec '03

(From alt.security.pgp) Rich Lethin <lethin(a)toast.ai.mit.edu> wrote: > >I was chatting today with someone moderately well-informed about the >clipper controversy (unlike me). He pointed out the following work to me >by Prof. Silvio Micali at LCS on a technical scheme which can serve as a >compromise between the needs of society for legitimate wiretaps and the >need of individuals for strong privacy. Basically, it seems to be a >protocol for extending a public key algorithm into a k-escrow system. This >apparently differs from the Clipper chip in that algorithmic details are >well publicized. > >The work was in Crypto '92 apparently, and an MIT lab for CS tech report >numbered TR-579b. Even worse, Micali is claiming that his patent on fair cryptosystems (#5,276,737) covers Clipper as well. In the Wall Street Journal (May 31, 1994, p. B6): Mr Micali, whose patent was issued in January, says his patent covers the concept of breaking an encryption key into multiple parts that are guaranteed to work, and are held by escrow agents. It seems to me that Clipper does not guarantee that the multiple parts will work in anywhere near the same way as his scheme does (see my book for details); Clipper is simply a secret splitting scheme. On the other hand, Micali filed his patent application in Apr 92, a full year before Clipper became public. Bruce ************************************************************************** * Bruce Schneier * Counterpane Systems For a good prime, call 391581 * 2^216193 - 1 * schneier(a)chinet.com **************************************************************************

1 0