December 2003 - cypherpunks-legacy

Mosaic / CEB / CypherEssay Archive and List
by Vincent.Cate＠FURMINT.NECTAR.CS.CMU.EDU 17 Dec '03

17 Dec '03

I think that Mosaic pages are the way to go these days. These are not hard to write and they are very nice to use. For some interesting examples of what can be done, check out some of the existing cypherpunk mosaic pages: ftp://soda.berkeley.edu/pub/cypherpunks/Home.html ftp://ftp.u.washington.edu/public/phantom/cpunk/README.html http://www.quadralay.com/www/Crypt/Crypt.html http://digicash.support.nl ftp://furmint.nectar.cs.cmu.edu/security/README.html http://pmip.maricopa.edu/crypt/cypherpunks I think it would be good to have a "cypherpunks-essay" mailing list that was not moderated but people only sent essays that they thought were good enough to be archived. For starters, people can just CC: mail to "vac+cypheressay" and it will get dropped into a separate file in: ftp://furmint.nectar.cs.cmu.edu/security/cypheressay My "vac+cypheressay" was done with one line in my .maildelivery and a short shell script I call "save-numbered". Both are below (note the A.I. methods used to moderate mail). Having a "cypherpunks-essay" mailing list on the same machine as "cypherpunks" could be nice for folks who want a lower volume group. Could have mail to it also go to cypherpunks, and Reply-To: also go to cypherpunks. -- Vince Addr vac+cypheressay pipe R /usr/vac/bin/save-numbered /usr/anon/security/cypheressay #!/bin/csh -f # # This is not totally safe as it is not atomic. If two messages # came at the same time there is a chance one would get lost. # For a low volume list this will not be a problem. cd $1 umask 022 set current = `cat num` @ next = $current + 1 echo $next > num cat >> $next # Moderator says no reply messages grep "Re:" $next if ($status == 0) then /bin/rm $next endif

2 1

Re: Mailer questions
by usura＠vox.hacktic.nl 17 Dec '03

17 Dec '03

ghio(a)cmu.edu (Matthew Ghio) writes: >> What is the maximum length of a "Request-Remailing-To:" header? > >Generally, as much as you can fit on the line. If you're running out >of space in your particular mail editor, try using "Anon-To:" instead. The remailers at VoX are running under dos so the maximum length is 128 symbols. VoX accepts X-Anon-To: and the lenghty Request-Remailing-To: it does _not_ support Anon-To: [note to some hip edu site !!] >> I understand that several of the headers can be changed by using >> a "##" token instead of "::". Which ones can be changed and >> which ones can't? VoX doesn't support ##. >> Has anyone had any experience chaining from a Cypherpunk-style >> remailer to anon.penet.fi? Can a "::" separator be used >> successfully, or which format will work? > >I think it can be done... I know it can be done... -- Exit! Stage Left. Alex de Joode <usura(a)vox.hacktic.nl>

1 0

CLIP: flaw found?
by ddt＠lsd.com 17 Dec '03

17 Dec '03

"Nothing can be made to be foolproof: fools are far too ingenious." -dave >Newsgroups: xpress.news.usa >From: xpress (Associated Press) >Date: 2 Jun 94 03:57:00 GMT >X-Category: NA*N**** >Subject: Times: Flaw Found In Wiretap > >NEW YORK (AP) > >A computer scientist has discovered a basic flaw in coding technology that the >Clinton administration has been promoting as a standard for electronic >communications, The New York Times reported Thursday. > >Matthew Blaze, a researcher at AT&T Bell Laboratories, told the Times his >research had shown that someone with sufficient computer skills can beat the >government's technology by encoding messages so that no one, not even the >government, can crack them. > >The administration has been urging private industry to adopt the so-called >"Clipper chip" as a standard encoding system. The government says telephone and >computer messages sent with the chip cannot be read by an outsider but can be >decoded by government law-enforcement agencies. > >Officials fear that without such a system, wiretaps would be useless against >criminals and terrorists because their communications could be hidden in >unbreakable codes. But communications executives and privacy-rights experts >fear the potential for snooping and worry that foreign customers wouldn't buy >the equipment if Washington could snoop on it. > >Blaze said the flaw he discovered in the Clipper design would not permit a >third party to break a coded computer conversation. But it would enable two >people to have a secret conversation that law enforcement officials could not >unscramble. > >Blaze said a draft report of his findings has been circulating among computer >experts and federal agencies. > >The National Security Agency, which played a leading role in developing the >technology, does not dispute the flaw's existence, but believes the Clipper >remains useful anyway, the Times said. > >Michael A. Smith, the agency's director of planning, told the Times in a >written response to questions that the flaw found by Blaze was difficult enough >to exploit that most people wishing to circumvent the system would find other >ways to do it. > >Martin Hellman, a Stanford University expert on data encryption who has read >Blaze's paper, said: "The government is fighting an uphill battle. ... People >who want to work around Clipper will be able to do it."

1 0

MacPGP 2.6ui yet?
by ddt＠lsd.com 17 Dec '03

17 Dec '03

Everyone, Has David Sternlight's worst nightmare made it to a Mac version yet? At 5:42 pm 5/27/94 +0100, mathew(a)mantis.co.uk (mathew) wrote: >-----BEGIN PGP SIGNED MESSAGE----- > >This is to announce an unofficial release of PGP, based on 2.3a, >modified for interoperability with MIT's PGP 2.6. dave

1 0

Anon Remailers + Clueless people = funny
by cdodhner＠indirect.com 17 Dec '03

17 Dec '03

-------- Forwarded message -------- Newsgroups: alt.sex.pedophile.mike-labbe,alt.sex.intergen,alt.sex.bondage,can.infobahn,can.general,can.legal,soc.culture.canada,ont.general,alt.usenet.kooks Date: Thu, 2 Jun 1994 12:08:30 GMT From: nightfly(a)netcom7.netcom.com (Nightfly) Subject: Re: Huge Porno Ring uncovered in Canada - audio report available In article <2sj9a0$akq(a)agate.berkeley.edu> Tommy the Tourist <nobody(a)soda.berkeley.edu> writes: Notice to all computer geeks in Canada! Got any "erotic" Gifs ? Delete them quick! Various Canadian Law enforcement agencies are about to redouble their efforts to stamp out pornography. Canadian Police are calling for a large scale crackdown on child pornography. It all started with two video tapes found by the river. The investigation has blossomed and resulted in more than 30 arrests. [yabbada yabbada yabbada] Take this with a big grain of salt, people. Tommy the Tourist has to be the biggest dust case on the net. Verify this through another source before you start chewing up and swallowing your floppies. Note addition of alt.usenet.kooks in the Newsgroups: line. *blink* Nightfly *blink*

1 0

Re: News Flash: Clipper Bug?
by pcw＠access.digex.net 17 Dec '03

17 Dec '03

>> One thing this shows, even if the application of the technique turns out >> to be too difficult to be practical, is that Dorothy Denning's evaluation >> of the design was worthless. That team should have found this themselves. >> No wonder she was trying to play it down in the NYT. > >It was my understanding that Denning was just looking at the Skipjack >algorithm, and not the clipper unit as a whole. I wouldn't be too quick >to condemn her on this point. Of course, Dorothy Denning could simply say, "Gosh, the Blaze result shows how to make Clipper _more_ secure for the average user. Now they don't have to worry about the government. I stand by my assessment that it is secure."

1 0

Re: NYT article
by joshua geller 17 Dec '03

17 Dec '03

> joshua geller says: > > an interesting commercial product might be software to > > do blaze's trick.... > You can't build it -- the EES is not available as an algorithmic > specification to the public -- only as hardware. quite true. I was mistaken. josh

1 0

Denning and Clipper...
by pcw＠access.digex.net 17 Dec '03

17 Dec '03

>> One thing this shows, even if the application of the technique turns out >> to be too difficult to be practical, is that Dorothy Denning's evaluation >> of the design was worthless. That team should have found this themselves. >> No wonder she was trying to play it down in the NYT. > >It was my understanding that Denning was just looking at the Skipjack >algorithm, and not the clipper unit as a whole. I wouldn't be too quick >to condemn her on this point. Of course, Dorothy Denning could simply say, "Gosh, the Blaze result shows how to make Clipper _more_ secure for the average user. Now they don't have to worry about the government. I stand by my assessment that it is secure."

1 0

Matt Blaze & Tessera Linked in NYT Again
by Duncan Frissell 17 Dec '03

17 Dec '03

Check the first page of the Bussiness Section of this morning's NYT for an article by John Markoff on Matt and Tessera. " 'The point is that the back door has a broken hinge,' said William Ferguson, vice president of Semaphore Corporation" DCF

1 1

Re: HACK - U HAVE NO PRIVACY
by gtoal＠an-teallach.com 17 Dec '03

17 Dec '03

Please tell me you posted this as a joke and didn't even contemplate it for a microsecond! Haven't you ever heard of Craig Shergold? Say, you're not the same lefty who's listed in the alt.syntax.tactical FAQ are you? G

1 0

Re: no subject (file transmission)
by gtoal＠an-teallach.com 17 Dec '03

17 Dec '03

See Tim May's "blacknet" spoof from last year. Anonymous mail pools have been the subject of long discussions. We were talking about it just last week actually - I pointed out that with suitable newsclip software you didn't even need a designated pool group. G PS I got thrown off the machine I was using to run the newsclip stuff on, so no more experiments with that for a long time...

1 0

The Delhi Detweiler
by rishab＠dxm.ernet.in 17 Dec '03

17 Dec '03

gtoal(a)an-teallach.com (Graham Toal): > PPS While in paranoid mode, I thought I'd mention that I've noticed > some of Detweiler's mannerisms emanating from email in New Delhi of > all places. Weird. I think I must be cracking up. I've been worried myself, BBSes here have had a number of posts against those who oppose our Government's recent moves to be nasty to BBSes. The style of these posts is very Detweiler; maybe you can now get a degree in Detweilerese? Or did you mean me ;-) -------------------------------------------------------------------------- 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

IMP (was Re: ecash-info (
by Duncan Frissell 17 Dec '03

17 Dec '03

H > 2.) It seems to me that that e-cash, contrary to the status quo's H >thinking, H > is *critical* to internet commerce. H > H >No, it's not critical. Some form of transaction mechanism is H >critical. Privacy is not critical to the bulk of the economy, H >though. Face up to it. If it were, it would be so obvious that we H >wouldn't be discussing it on a mailing list. In fact, _we_ wouldn't be H >discussing it, but rather a whole bunch of bank vice presidents. The nicest thing about electronic money is not that anonymous digital cash is possible. but that the money is electronic. This means cheap transaction costs. This means that even if "banks" deploy ID-based rather than anonymous transaction systems, civilians can quickly step in and offer to broker anonymous transactions for those who prefer their privacy. Just as you can cash checks for other people today, you will be able to carry out transactions for others on the nets. Since it is much easier to operate these kinds of services in an electronic payment system with automation and on-line verification, suppliers should appear if the demand warrants. Since people will be in different physical locations and legal jurisdictions proof of wrong doing will be hard to come by. DCF "It is perfectly possible to overthrow the old order without picking up a gun or violating any (significant) law. If the conditions of human life change as radically and rapidly as they seem destined to, the (always bizarre) notion that groups of small oligarchies are competent to run the lives of everyone on earth will be swept aside almost accidentally as people move into their new lives." -- DCF --- WinQwk 2.0b#1165

1 0

more info from talk at MIT yesterday.
by sommerfeld＠localhost.medford.ma.us 17 Dec '03

17 Dec '03

The folks from the NSA said the following about key generation: - each escrow agency provides a "seed key", seed1 and seed2 - the box which programs the chip generates two random keys, random1 and random2 - for each chip programmed during that batch (which is "12 to 14 hours of production"), the box computes a classified deterministic function (U1, U2) = F(serial, random1, random2, seed1, seed2) to generate the unit keys They did *not* explicitly say that the random seeds were destroyed at the end of the production run. Also, someone asked "How do we know that the unit key isn't a hash function of the chip serial number?" The answer was: "You don't". They also confirmed Tom Knight's suspicions about what they're going to do when someone reverse engineers the chip and publishes the Skipjack algorithm & the family key: they've got a patent application filed, under a secrecy order; if the algorithm is published, they'll lift the secrecy order and have the patent issued, and use that to go after anyone making a compatible version. They also had a comment that they considered Blaze's findings to be mostly irrelevant, as the only people who would use it would be persons who *didn't* trust the escrow system, but *did* trust the algorithm... - Bill

6 7

Re: Faster way to deescrow Clipper
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

> > ...not be able to decrypt the communications, but they still get your ID. > >"your ID"? You mean your phone's ID. Goodness gracious, if you were > >a criminal, you wouldn't go out and steal someone else's Clipper > >phone, would you? Let's not get too high tech here, just because we > >have the ability. > > Or you could just steal someone else's LEAF, by keeping a copy of it, and use > that for spoofing. Then you could have a valid IV too... The IV is session-dependent, and both ends generate it. We don't know where in the LEAF the chipid is, but if they use a fixed format and don't do a key-dependent permutation of the LEAF bits, it shouldn't be hard to figure out (unless the checksum comes first and they use a block-chaining encryption, in which case you know you lose.) That would let you create rogue LEAFs with known users' chipids, which would be interesting - does anyone want to make 65536 calls to clipperphone(a)whitehouse.gov :-) ? (Yeah, it's not quite that simple.) (If you do need a lot of data, cellphones are a good source, since the cellphone operators' chipids are likely to be wellknown, though rapidly tapped.) Paranoid-speculation-mode: Of course, if you can forge LEAFs with their chipid, they can forge LEAFs with yours, which could be used to manufacture interesting evidence.... Bill

2 1

Re: more info from talk at MIT yesterday.
by werner＠mc.ab.com 17 Dec '03

17 Dec '03

>Date: Fri, 03 Jun 1994 10:36:37 -0400 >From: "Perry E. Metzger" <perry(a)imsi.com> > >Since when can the government patent its work? I thought that works >produced by government agencies could not be copyrighted or patented. I'm not sure of the details, but I know the law in this area was changed during the Reagan administration to unfetter U.S. companies who had done research that was partially govt. funded. Under the old rules, foreign companies could take advantage of work that could not be patented or otherwise protected by trade secret. Thus, technologies that were developed through govt.-funded work were acquired for free and used against the U.S. in global economic competition. Not sure about work that is completely government-produced, but what did the government ever come up with all on its own, anyway? I do know that my company is currently engaged in several joint projects with the government as a result of the new regulations, and part of the deal is that we retain the rights to the inventions. tw

1 0

Re: more info from talk at MIT yesterday.
by smb＠research.att.com 17 Dec '03

17 Dec '03

Bill Sommerfeld says: > They also confirmed Tom Knight's suspicions about what they're going > to do when someone reverse engineers the chip and publishes the > Skipjack algorithm & the family key: they've got a patent applicatio n > filed, under a secrecy order; if the algorithm is published, they'll > lift the secrecy order and have the patent issued, and use that to g o > after anyone making a compatible version. Since when can the government patent its work? I thought that works produced by government agencies could not be copyrighted or patented. The government can patent things, but not copyright them. In any case, they cannot refuse to license a patent, so this isn't real protection anyway. (The hope behind people patenting things they may release in the future is to make it commercially less attractive, not to utterly prevent use.) Why can't they refuse to license a patent?

1 0

Re: more info from talk at MIT yesterday.
by Perry E. Metzger 17 Dec '03

17 Dec '03

smb(a)research.att.com says: > Why can't they refuse to license a patent? I was under the impression that the law obligated you to license patents -- albeit not necessarily at an attractive price. I am quite likely to be mistaken on this -- my memory on the topic is very sketchy, as demonstrated by the fact that I didn't realize the government can patent things (although I was right on copyrights.) Perry

1 0

Re: no subject (file transmission)
by f_griffith＠ccsvax.sfasu.edu 17 Dec '03

17 Dec '03

Date: Thu, 2 Jun 1994 17:58:38 -0700 > From: nobody(a)shell.portal.com > > Attn Pr0duct Cypher > -----BEGIN PGP MESSAGE----- > -----END PGP MESSAGE----- > >Now that's an interesting idea for contacting people anonymously. However, it would have been kinder to the rest of the list members to put Attn Pr0duct Cypher in the subject.

1 0

Re: more info from talk at MIT yesterday.
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

> Bill Sommerfeld says: > > They also had a comment that they considered Blaze's findings to be > > mostly irrelevant, as the only people who would use it would be > > persons who *didn't* trust the escrow system, but *did* trust the > > algorithm... Defense Messaging System is supposedly going to use Skipjack, so I assume it's reasonably secure - and if there *are* NSA-only backdoors in the algorithm, at least they won't be admitting it to your neighbor hood cops and FBI wiretappers, so you'd have to be an *interesting* suspect to get cracked. Bill

2 1

Re: Black Eye for NSA, NIST, and Denning
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

Perry writes: > > However, it can be done in advance, and you can conceivably reuse > > forged LEAFs. > > I will point out something that I didn't quite understand myself but > have since discussed with Matt Blaze in some detail -- LEAF checksums > are tied to session keys. You CAN do this in advance but only if your > key exchange will permit you to generate your session keys in advance, too > Obviously, reusing forged LEAFs requires reusing session keys. More precisely, as Steve's summary pointed out, it's tied to the IV, which is tied to the session key. (It makes sense - assuming the descriptions of the LEAF contents are true, the only session key component in the LEAF itself is encrypted with the chip-unique backdoor key, and tying it to the IV accomplishes key-dependence, though they could also use the session key externally from the LEAF.) Unfortunately, most Clipperphones will probably use Diffie-Hellman key exchange, since it reduces or eliminates the need for prearranged public-key management (depending on whether they're using radio or a medium that can be actively wiretapped), so precomputation will generally not be usable. I suppose some crude Diffie-Hellman implementations might always use the same half-key for every conversation, rather than generating a random one each time, and you could precompute session keys for talking to them. For email applications, however, most standards will probably use sender-generated session keys, so it would be simple enough to make secure Tessera mailers if you don't worry about subliminal channels in the hash. Bill

1 0

Re: more info from talk at MIT yesterday.
by smb＠research.att.com 17 Dec '03

17 Dec '03

> Bill Sommerfeld says: > > They also had a comment that they considered Blaze's findings to b e > > mostly irrelevant, as the only people who would use it would be > > persons who *didn't* trust the escrow system, but *did* trust the > > algorithm... Defense Messaging System is supposedly going to use Skipjack, so I assume it's reasonably secure - and if there *are* NSA-only backdoors in the algorithm, at least they won't be admitting it to your neighbor hood cops and FBI wiretappers, so you'd have to be an *interesting* suspect to get cracked. Not only that, there have been too many spies found in the various intelligence agencies for them to risk such a thing.

1 0

Re: Faster way to deescrow Clipper won't work
by sidney＠taurus.apple.com 17 Dec '03

17 Dec '03

Perry answered most of my questions with the new information that the checksum is a function of the IV and session key. That still leaves one big one unanswered: Doesn't the Clipper chip at the other end of the message have to compute and send a LEAF too? Otherwise there would be no way for the wiretappers to identify it. And that LEAF will contain the proper session key unless both ends are using the same hack. Also, what does Clipperphone have that the Tessera board doesn't which keeps this hack from working? Is enough known about the workings of the two so that someone could build a computer-based telephone using Tessera that would be compatible with Clipperphone yet use this hack? -- sidney <sidney(a)taurus.apple.com>

1 0

Re: Faster way to deescrow Clipper
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

Sidney Markowitz writes: > If that's the case, then 1) How does the second chip find out what the > session key is? That's a separate protocol issue; Clipper doesn't do any key exchange itself, though Capstone does. Unless manufacturers are bullied/bribed into using a standard implementation, everyone will probably roll their own. > 2) Doesn't the second chip also have to generate and send a > LEAF, if for no other reason than to identify itself to the wiretappers, > and if so won't that give away the session key if that chip's device is not > also hacked? If you use the same session key for both directions of the conversation, which most Clipperphones probably will, then yes, it's true. That means you can only have private conversations with other people who also care about privacy, which is somewhat appropriate. On the other hand, a big use of Clipper is traffic analysis, and Matt's method *will* prevent them from getting your Clipper serial number from your conversations, though they'll get the number for the other end if they're not also hacking LEAFs. That can be a big win, especially if the other end is a well-known person, like your local cellphone provider or president(a)whitehouse.gov. However, one danger of doing this for cellphone calls is that they might notice that calls from your cellphone keep having different LEAFs, and suspect that you're a Potential Troublemaker. 3) If all that is needed for this hack is a LEAF with a proper > checksum, why go through the brute force method of generating random LEAFs? > Why not just buy (or steal or whatever) another Clippered device that you > never use for real communication so the wiretappers have no record of who > has that serial number, and get LEAFs from it? For that matter, why can't > you obtain one LEAF from listening to anybody's Clippered transmission and > use it over and over again? The LEAF depends on the IV for the session, which depends on the session key. Therefore, it's probably different for each call; otherwise you *could* just reuse someone else's LEAF. (This should be obvious, but I wasn't thinking about it when I first read Matt's paper, though the "but the IV will be wrong so that won't work" had been a sufficient distraction for many of us when CLipper first came out.) Remember that they don't record Clipper chip keys when you buy your Clipperphone - otherwise stealing one would be effective. They record the chip-unique backdoor keys when they make the chip, so they can tap *any* conversation they hear without needing to keep track of who owns what phone. On the other hand, for cellphones, it's *real* easy to find out who uses a given chip, since the phone call setup protocols tell them what phone it's coming from, and they _can_ look that up with the phone company, so they can easily do that correlation. (If the Clipper chips are socketed, you could always swap them for occasional more-paranoid-but-still-tappable calls, but that would probably just annoy them.) Bill

1 0

Re: News Flash: Clipper Bug?
by wcs＠anchor.ho.att.com 17 Dec '03

17 Dec '03

> > [ discussion of chip operations ] > > Where did you get the information about the internals? > I've seen no references at all. Back when Clipper was first announced, you could call up Mykotronx and ask them for data sheets on the chip. Don't know if they're still giving them out or not.

1 0