December 2003 - cypherpunks-legacy

Transforming variable- to fixed-length keys
by pgut001＠cs.auckland.ac.nz 17 Dec '03

17 Dec '03

In preparing the next version of cryptlib (which is going to have some cool features when it's ready, which should be before the end of the millenium), I've run into a problem in writing a general-purpose n-byte input to m-byte output transformation function. What this does is take an arbitrary-length user key and transform it to a fixed-length encryption key (for example an entered passphrase into a 112-bit triple-DES key). The constraints on memory usage are: - The input (user) key can't be altered (you can't change data passed in by the caller) - The user key can't be copied to an internal buffer (it can be of arbitrary length, and is sensitive material so shouldn't be copied elsewhere) In other words there's no temporary storage available apart from what's provided in the output key. This is almost always a different length from the input key. Some other constraints are: - The transformation must be algorithm-independant (it shouldn't, for example, rely on SHA1 to transform an input string into a fixed output of 160 bits and assume you'll never need a key longer than 160 bits). This means you can't just use a single pass of a hash function to generate the output key, since the output can be smaller or larger than the hash function output. - The transformation must be able to be iterated to make a password-guessing attack harder to perform. This one is tricky, since the lack of temporary buffer space means you can't just feed the output back to the input and iterate. Here's my initial approach, if anyone has any comments to make on this or knows of a better way to do it, please let me know. Peter. -- Snip -- Initially, the user key is passed in as a byte string: +-------------------------------------------------------+ | User Key | +-------------------------------------------------------+ The first stage in the key hashing prepends the length of the string as a big-endian 16-bit count to the user key: +------+-------------------------------------------------------+ |Length| User Key | +------+-------------------------------------------------------+ The aim of the hashing is to reduce this variable-length input string to a fixed-length key appropriate to the encryption algorithm being used. This is done by treating the user encryption keys as circular buffers and repeatedly hashing chunks of the user key and xoring the result into the output buffer. Thus the first chunk of the encryption key would be obtained with: +------+-------------------------------------------------------+ |Length| User Key | +------+-------------------------------------------------------+ | | | _ / | Hash _ / | _ / | / | | +-----------------------+ | Encryption Key | +-----------------------+ The second chunk of the enryption key would be obtained with: +------+-------------------------------------------------------+ |Length| User Key | +------+-------------------------------------------------------+ | | | _ / | Hash _ / | _ / | / | | +-----------------------+ | Encryption Key | +-----------------------+ Since the input to the hash function is much larger than its output, a significant amount of the user key affects each chunk of the encryption key. The size of each "chunk" is determined by the hash function being used. For example with the MD4 hash function, 64 bytes of user key affect each 16 bytes of encryption key. Once the end of the user key or encryption key buffer is reached, the hash function wraps around to the start of the buffer and takes its data from there. A pass over the user key is considered complete when the hash function input has wrapped around completely and is back at the start of the buffer. The amount of wraparound depends on the length of the user and encryption keys. For example with 8-byte (strictly speaking 56-bit) DES keys even a single application of MD4 will wrap around the encryption key buffer twice, shrinking up to 64 bytes down to 8 bytes in a single operation. On the other hand a 4-byte user key will wrap the user key buffer around twice, expanding it to fill 8 bytes of the encryption key buffer (without, however, actually giving 8 bytes of effective key space). In order to avoid repeatedly hashing the same data (which results in the output key cancelling out every second round), the input data is varied by adding the iteration count mod 256 to each byte before it is hashed. Therefore for five rounds of key hashing the user key "This is a key" would give the following effective input to the hash function: \x00\x0DThis is a user key \x01\x0EUijt!jt!b!vtfs!lfz \x02\x0FVjku"ku"c"wugt"mg{ \x03\x10Wklv#lv#d#xvhu#nh| \x04\x11Xlmw$mw$e$ywiv$oi} [Is this nice? Problems are that you might be able to perform some sort of related-key attack, and that if you know the input value to round n you can get the input value to round n+m without having to go through all m rounds. However I can't see how this would aid an attacker].

1 0

ratings
by bryce＠digicash.com 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- Of course a real rating would be signed by my ratings private key and not by my e-mail private key. I might start a service where I will pay you for your ratings and then distribute them for a fee. (Note that you get paid for generating ratings, as long as your ratings get good meta-ratings. You have to pay to use other people's ratings. This is how it should be.) Regards, Bryce -----BEGIN PGP SIGNATURE----- Version: 2.6.2i Comment: Auto-signed under Unix with 'BAP' Easy-PGP v1.1b2 iQB1AwUBMd7iRkjbHy8sKZitAQGTEQL+IchRcLv9r7/WpDQb8tmJ0QJD7tN8KUNG AEX/UtzApwffH2kS90ThHVnsVt/8WKgI+WfsZ0Z0PtYoE5uLwBUJDzXydbZ8zHEx B5Ti4pfF0wuXwWD6kA/ISYyhZHRewcol =Dwsd -----END PGP SIGNATURE-----

1 0

Re: Net and Terrorism.
by Arun Mehta 17 Dec '03

17 Dec '03

At 00:37 05/07/96 -0700, Rich Graves wrote: >Since the two muslims who used to give a shit about this list seem to have >left in disgust, I suppose I should register my "That ain't representative >of Islam... Not so long ago, when Moslems were fighting the Soviet Union in Afghanistan, they were heroes to the western world, supplied arms and money enough to destabilise the whole region. Once the Soviet menace faded, the same fighters were branded terrorists. Mixed signals like this are responsible for much of the animosity that one finds in the Islamic world (possibly even in other parts of the world) against the US. Arun Arun Mehta Phone +91-11-6841172, 6849103 amehta(a)cpsr.org http://mahavir.doe.ernet.in/~pinaward/arun.htm

1 0

Re: Net and Terrorism
by vinnie moscaritolo 17 Dec '03

17 Dec '03

>On Fri, 5 Jul 1996, vinnie moscaritolo wrote: > As for a solution to governement problems, I have always and still belive > in the ballot box first. And the only way to win votes is to appeal to the > morons out there who do vote.. Maybe there is hope in the next generation > of internet literate kids.. or maybe I am just a dreamer. > >snow(a)crash.suba.com wrote > You are a dreamer. Netscape, TV for the internet. BS. I am not being that idealistic. look all I am saying is that maybe the net just offers kids the ability to see a variety of views instead of the mainstream liberal (or whatever it will be next week) controlled media. For now anyways any entity that has something to say and can write his way out of a paper bag has pretty much the same ability to influence on the net (at least newsgroups) as say Ted Copulate does. Maybe freedom we have here wont last for long, maybe the Pointcast of the future will just create another MTV generation...but at least for now we have a voice. Vinnie Moscaritolo ------------------ "friends come and friends go..but enemies accumulate." http://www.vmeng.com/vinnie/ Fingerprint: 4FA3298150E404F2782501876EA2146A

1 0

Re: [RANT] Giving Mind Control Drugs to Children
by James A. Donald 17 Dec '03

17 Dec '03

> > Er... Tim... Ritalin is an amphetamine. At 07:25 AM 7/7/96 -0700, Sandy Sandfort wrote: > Yes, normally, but doesn't it have a paradoxical reaction for > hyperactive children (i.e., it acts as a depressant for them)? Not really: Ordinary college students who use it to facilitate cramming report that it has the same effect on them as on hyperactive children. A well known symptom of amphetamine abuse is that the abusers will cheerfully persist in pointless and boring activities for hours on end, such as folding paper bags or stirring long overcooked spaghetti. Furthermore, people generally have to be forced by the threat of violence to take depressant drugs, especially neuroleptics, whereas everyone cheerfully takes their Ritalin. --------------------------------------------------------------------- | We have the right to defend ourselves | http://www.jim.com/jamesd/ and our property, because of the kind | of animals that we are. True law | James A. Donald derives from this right, not from the | arbitrary power of the state. | jamesd(a)echeque.com

2 1

NYT/CyberTimes on CWD article
by declan＠well.com 17 Dec '03

17 Dec '03

"We are writers, not crytographers." -Declan --- http://www.nytimes.com/library/cyber/week/0706patrol-reporters.html July 6, 1996 Reporters Claim to Have Lists of Blocked Sites By PAMELA MENDELS Reporters Brock N. Meeks and Declan B. McCullagh say they've got a little list. Several actually. The lists are of Internet sites that, in the eyes of several companies making parental control software, could be considered inappropriate to children. The lists are supposed to be secret. But Meeks and McCullagh say they have obtained lists compiled by Microsystems Software, Inc., the Framingham, Mass.-based manufacturer of Cyber Patrol; Los Altos, Calif.-based SurfWatch Software, a subsidiary of Spyglass, Inc., and Santa Barbara, Calif.-based Solid Oak Software, Inc., maker of CYBERsitter -- three of the leading producers of parental- control filtering software. McCullagh said that he and Meeks were able to view the complete Cyber Patrol and CYBERsitter lists and part of the SurfWatch list. In an article published this week in CyberWire Dispatch, a report on Internet-related issues distributed through e-mail, Meeks and McCullagh wrote that they had taken a peek at some of the sites contained on the lists and had then contacted groups that might be concerned about the listings. Representatives of organizations ranging in advocacy from feminism to gun lobbying to animal rights said they been disturbed to learn that some sites they endorse had made the lists. Kim A. Gandy, executive vice president of the National Organization for Women, said Friday that she was upset to learn that CYBERSitter blocks access to NOW's Web site. Further, she said she did not like the company's rationale: that the NOW site contains links to, among other things, sites about homosexuality. "It's ridiculous," Gandy said. "It's insulting. And I think most parents would not approve of that kind of censorship. Lots of parents don't want children surfing pornography, but would not think of denying them access to legitimate information." Marc E. Kanter, director of marketing for Solid Oak, confirmed Friday that NOW's site had been included on the CYBERsitter not-for-children list because of its links leading to "sexual preferentation" sites. "This is what our users want," he said. "If they don't want to restrict access to this material, they don't have to buy it or they can simply turn it off. We are not trying to play any political role. We are simply providing a tool for parents." Officials of the Gay & Lesbian Alliance Against Defamation were also upset that the Cyber Patrol list blocked several Internet discussion groups devoted to news of interest to the gay community. "We feel that this is the kind of thing important to gay and lesbian youth, to read about our community," said Lauren R. Javier, director of information systems for the Gay & Lesbian Alliance, adding that the newsgroups contained little if any sexually explicit material. Javier added that Cyber Patrol officials had been responsive in the past to complaints, so he wanted to give them "the benefit of the doubt" and intended to contact them about the matter. For his part, Nigel R. Spicer, president of Microsystems, said he had not examined the reasons that all the gay newsgroup sites named by the article were included on the Cyber Patrol list. The one site he did check after reading Meeks' report, however, was on the list because it contained links to personals ads, he said. McCullagh is keeping mum about how he and Meeks got the lists in the first place, although he denies that either of them personally decoded the software. "Brock and I are not cyptographic analysts," he said. "We don't spend our days de-encrypting files. We are writers, not crytographers." Spicer was less than happy about the prospect that Cyber Patrol's list may have fallen into outsiders' hands. He said that, so far, he had been unable to confirm whether the reporters had the true list for Cyber Patrol and, if so, how they had managed to obtain it. "It's always a concern if you believe people are getting access to material you've gone to the trouble to not make available," he said. "If we believe the encryption scheme has been compromised, we will make another one." Kanter, of CYBERsitter, said the list mentioned in the Cyberwire Dispatch article was, indeed, his company's. "I hope that list doesn't get out beyond where it was," he said. Jay S. Friedland, vice president of marketing for SurfWatch products, said Friday that he had not yet read the article. He said the blocking companies keep their lists secret for two reasons: to prevent their misuse and to keep their competitive edge. "Clearly, each company has a proprietary advantage," Friedland said. "One of our competitors could take and use the same information." ###

2 1

Re: CCC Crypto Lock
by Deranged Mutant 17 Dec '03

17 Dec '03

On 6 Jul 96 at 4:10, anonymous-remailer(a)shell.port wrote: > MicroPatent, 4 July 96 [..] > Abstract: Disclosed systems and methods for protecting a > software program from unauthorized use and copying > through the removal at least one of a plurality of > instructions comprising a software program, and > encrypting the removed instruction utilizing an > encryption algorithm to produce an encrypted instruction, > the encryption algorithm responsive to a randomly > generated key. Would certain computer viruses be considered prior art here? (Be it that they encrypt for the purposes of hiding rather than copy protection though.) Rob. --- No-frills sig. Befriend my mail filter by sending a message with the subject "send help" Key-ID: 5D3F2E99 1996/04/22 wlkngowl(a)unix.asb.com (root@magneto) AB1F4831 1993/05/10 Deranged Mutant <wlkngowl(a)unix.asb.com> Send a message with the subject "send pgp-key" for a copy of my key.

2 1

Re: What remains to be done.
by Deranged Mutant 17 Dec '03

17 Dec '03

On 6 Jul 96 at 13:32, Anthony Daniel wrote: > Should try SECURE DESK-TOP as well (PC only), it can encrypt (DES and IDEA) [..] > And it can add PEM capabilities to most e-mail clients and it's WIN95 and PGP capabilities would be nice. :) > user friendly. Try it at: > > http://www.systems.it/secure > > There are NO export restrictions on it as well because it's Italian made. Well, it can't be downloaded from a US site to a non-US site anyway. It's nice to see some non-US strong crypto in that it will be all the more impetus to relax ITAR. Rob --- No-frills sig. Befriend my mail filter by sending a message with the subject "send help" Key-ID: 5D3F2E99 1996/04/22 wlkngowl(a)unix.asb.com (root@magneto) AB1F4831 1993/05/10 Deranged Mutant <wlkngowl(a)unix.asb.com> Send a message with the subject "send pgp-key" for a copy of my key.

1 0

Re: CWD -- Jacking in from the "Keys to the Kingdom" Port
by Alan Olsen 17 Dec '03

17 Dec '03

At 06:27 AM 7/6/96 +0000, you wrote: >On Fri, 5 Jul 1996, CyberEyes wrote: > >> NOT talking about Lynx, what DOS-based Web browser is there? > > Net-Tamer. > > Requires a PPP connection, and precious little else. The problem is getting PPP to work under DOS. If your kid can do that, then he will have no problem in disabling any sort of filtering, as well as wiping the hard drive and installing Linux with X11R6. DOS stacks are a pain to get functioning, usually have little to no useful instructions, and tend to be harder than hell to find. (Or as Homer Simpson once said: "Mmmmmm! Packet drivers!") I expect to see a case where some kid gets in trouble for filtering out what his parents can see, read, or hear. Or sets the school filter to only allow going to porno sites. I find it humorous how many people think that they can use technology to babysit their kids when the kids understand the technology much better than they do in most cases... --- Alan Olsen -- alano(a)teleport.com -- Contract Web Design & Instruction `finger -l alano(a)teleport.com` for PGP 2.6.2 key http://www.teleport.com/~alano/ "We had to destroy the Internet in order to save it." - Sen. Exon "Microsoft -- Nothing but NT promises."

1 0

Re: [RANT] Giving Mind Control Drugs to Children
by AwakenToMe＠aol.com 17 Dec '03

17 Dec '03

Did I miss something?? I fisrt came here and asked something about protected mode and was yelled at for asking it in this newsgroup. Now we're onto mind control drugs? uhhhhhh ok.

1 0