- cypherpunks-legacy@lists.ogf.org

Re: Apple+Clipper
by lefty＠apple.com 31 Aug '93

31 Aug '93

Tim May says: >Yep, I was just making this point, that Apple is _already_ releasing >hardware with OS incompatibilities. (By the way, many/all? of the >newer Macs need a "system enabler" patch to run System 7.1, that is, >the standard System (OS) no longer runs on all machines. Actually, this has always been true. It's just that it was done in the form of patch resources internal to the System file prior to the release of System 7.1. It used to be that, in order to save disk space, one could go into the System file with ResEdit and remove the various 'PTCH' resources that didn't apply to the particular machine you happened to be using. The System Enabler scheme is simply a somewhat cleaner method of accomplishing the same thing: you now have a "generic" System file, with the various hacks segregated in the Enabler. -- Lefty (lefty(a)apple.com) C:.M:.C:., D:.O:.D:.

1 0

PGP: question
by Karl Lui Barrus 31 Aug '93

31 Aug '93

Cypherpunks, Are there any heavy implications, security and otherwise, to the truncation of the environment variable PGPPATH that occurs in buildfilename()? Background: I've been trying all sorts of stuff getting pgp to read config.txt when I'm anywhere else in my directory structure. My home directory is /home/klbarrus which as it turns out is a symbolic link another directory (it may even change from time to time). Since I never could get PGPPATH to work on my old NeXT account, and my home directory there was also a link to something else, I thought: AHA! PGP uses stat() and/or some other functions which don't follow symbolic links!! Nope, it does, so that wasn't it. So, I poked around the code more and found out that buildfilename() returns null if the length of getenv(PGPPATH) is greater than 50. This probably explains why it didn't work on the NeXT; PGPPATH was set to some huge path like /private/Net/tree/Users/barrus/Cryptography/pgp or something close, so buildfilename() returned null. But the pathname on my new account is definitely shorter than 50 characters, but I decided to try increasing 50 just to see what it would do. Right before remaking pgp I realized the mistake I had made, one that is (export) almost too embarrasing to admit :-) Anyway, any "deep" reason to return null if the length of the environment variable is > 50? Or is it just to keep the path relatively short, maybe to keep from breaking a system call on some machine out there? -- Karl L. Barrus: klbarrus(a)owlnet.rice.edu keyID: 5AD633 hash: D1 59 9D 48 72 E9 19 D5 3D F3 93 7E 81 B5 CC 32 "One man's mnemonic is another man's cryptography" - my compilers prof discussing file naming in public directories

1 0

Attacks on remailers
by hfinney＠shell.portal.com 31 Aug '93

31 Aug '93

>>>>> On Wed, 25 Aug 93 11:39:11 EDT, J_G_Thomas%CAASD1(a)mwmgate1.mitre.org said: Joe> A new protocol is probably the cleanest way to solve the Joe> problem of traffic analysis of messages addressed with Joe> encrypted address blocks. The best way to get security Joe> in a remailer chain is to nest your encryption, so only Joe> one layer gets peeled off in each remailer hop. That Joe> isn't possible with encrypted address blocks, since the Joe> sender will only know the address (and public key) of the Joe> first remailer in the chain. All hops after the first Joe> one must send the same message out as they got in, with Joe> just a layer off the encrypted address block. As I indicated in my long posting, it is not necessary to send out the same message that was received. Chaum proposed encrypting the message (the non-address-block portion) with a secret key at each stage, a key which would be revealed to the remailer (along with the address of the next address in the chain) when it peeled off its own layer of encryption. But if Joe> remailers talked to each other by first doing RSA-signed Joe> Diffie- Hellman key exchange, then encrypting the Joe> traffic, a packet snooper wouldn't be able to correlate Joe> incoming and outgoing messages. If no encryption is done on the message body, there is another attack for this case that I didn't mention. It is: Run a remailer. For every anonymous address floating around on the net, try sending a message to it. Look at the messages which pass through your own remailer and look for matches to the message you sent. Any anonymous address which includes your remailer as one of the elements will pass through you. You have then defeated all of the stages of the chain before yourself. In particular, if you happen to be the last remailer of the chain, you have broken the anonymity of the anonymous address. This attack, while not the most powerful on the list, defeats many of the principles of remailer chains, such as that the chain is as strong as its strongest link. It requires you to strongly trust at least one remailer in the chain (the last one), whereas without this attack you would not have to especially trust any single remailer. So it is sig- nificant. Diffie-Hellman encrypting messages between remailers would not help against this attack. Also, rather than DH it would be just as effective to use the public key of the next remailer in the chain, and more convenient: some remailers are not able to participate in TCP exchanges, being connected to the net by occasional uucp connections. This lack-of-TCP problem also impacts the proposal to use a public telnet port for message communication. Another problem with that proposal is that it would need the remailers to run as background processes. With the current software they can run as mail filters, which makes them much less conspicuous to system managers. The suggestion for remailers to send messages by telnet connection to port 25 of some other machine (rather than by piping to sendmail as they currently do) is perhaps reasonable (for those systems with TCP access), although it makes the remailer somewhat harder to set up since you have to find some other machine which will let you connect to their port. Also, I think some machines may log incoming or outgoing telnet connections to this port, since it is a common technique for mail forgeries. I have heard that most systems will actually not allow public telnet connections to this port. I don't know that much about how widely available telnet and other TCP/IP services are on the net, so if these techniques are more usable than I am suggesting I'd like to hear about it. Hal Finney hfinney(a)shell.portal.com

5 7

encryption program posted to comp.sources.misc
by khijol!erc＠colossus.apple.com 31 Aug '93

31 Aug '93

Anyone else seen "syf" posted to comp.sources.misc? Someone oughta talk to Paul Vixie about posting XOR encryption programs... <sigh> -- Ed Carp, N7EKG erc(a)apple.com 510/659-9560 anon-0001(a)khijol.uucp If you want magic, let go of your armor. Magic is so much stronger than steel! -- Richard Bach, "The Bridge Across Forever"

1 0

Re: Free Electronic Cash
by thug＠phantom.com 31 Aug '93

31 Aug '93

> Exactly which corrupt EFF did you have in mind? Howabout some serious > facts rather than random accusations? Or did you just want to libel us? > > John Gilmore > Board of Directors, Electronic Frontier Foundation True or False, the EFF accepted money from AT&T in return for the EFF's support on various issues/bills before congress wherein AT&T came out on top in it's effort to commercialize parts of the Internet for it's own monetary gain? I've that's not bribery and corruption of what otherwise is supposed to be a non-profit organization, then I don't know what is. 'nuf said. Thug

3 2

Re: Help/Advice requested (fwd)
by Superuser 31 Aug '93

31 Aug '93

Through e-mail, gnu(a)toad.com writes: | | > Anyone have some ideas about how to implement such a beast? | | First you have to tell us what it's supposed to do. [ ... ] Sorry, I had assumed that people would know about registration schemes here. By the use of a (perhaps complex) algorithm, generate serial numbers that would be entered by the registering user which would effectively 'register' the program. There would be a routine in the executable that could detect a valid serial number and act accordingly. The algorithm to do this should be well-hidden into the code, as to discourage simple hacking and alteration... There are a wealth of schemes out there. For example, SCO UNIX uses a (albeit weird) scheme that involves a serial number and activation key which actually decrypts the executable (not what I want to do, really). INFORMIX uses a scheme that brands the executable with a serial number (very interesting). I thought of having an external program that would brand the executable with a serial number and perhaps some other hidden information in reserved spaces. Something like the program mentioned below would be effective in hiding the code. What else do you need to know? | > have run something like PKLITE on your executable) encapsulates the | > EXE file, encrypted, with CRC checks and debugger traps. | | Fine, it uses encryption and CRC and debug traps. Now, to what purpose? | What's the objective? What's the goal? Why? [ ... ] Here's a README from the distribution... Protect! EXE/COM 3.1 encrypt and still run .EXE's Registration $ 25.00 SDN_UTIL AUG93 PROTECT EXE COM CRC SECURE ENCRYPT SCRAMBLE FILES: PrExCm31.SDN Author: Jeremy Lilley Protect! EXE/COM is a powerful EXE/COM protection utility which thoroughly encrypts any EXE or COM file but still allows it to be run. And, files protected by it make a full CRC INTEGRITY CHECK every time they are run. This NEW version fixes many bugs found in previous versions of this program such as an infamous screen bug in v.3.0. Additions are made to make this program to prevent CRACKING, REVERSE ENGINEERING, and MODIFICATION even more. This program is one of the most powerful and cost-effecive solutions to problems with your programs' being modified by malicious hackers. With every other form of file security being nullified, Protect! has been renewed and can now secure your programs more than ever. What would you do if somebody were to add a TROJAN or 'MODIFY' a few screens to a program of yours ? Protect! can make this circumstance extremely unlikely for only $25. If a problem is detected, a message of your choice can be outputted to the user to tell the user exactly what to do. Instead of burying your head in the sand, get Protect! to protect your programming and software investment. Requirements: DOS 2.0 or higher, IBM or compatible PC, 256k RAM, & any executable file compression utility such as LZEXE, PKLite, Compack, Diet, or another program for compressing EXE files before they are scrambled and protected. [ .SDA Format is (c)Copyright 1993 The SDN Project ] [ SDN Authors-Only Info Line is 203-634-0370 USA ] -------- If you need more information, I can give you the author's email address...

1 0

What's wrong with PEM?
by collins＠newton.apple.com 31 Aug '93

31 Aug '93

After reading the RFCs for PEM (1421-1424), I am curious what other people think about PEM. For cypherpunks agenda, in what ways is PEM lacking? My take is: 1. PEM is a protocol, only applicable to mail (perhaps only to internet mail) while PGP is program that provides similar services for mail, but is also applicable to non-mail related encryption tasks. 2. PEM and PGP don't aggree on the symmetric algorithms (DES, IDEA). 3. PEM certificates are bulky, and transmission is encouraged. 4. PEM certificates are issued by Certificate Authorities, which would seem to preclude PGP's 'web of trust' model. These all seem to have answers: 1. PEM is protocal, PGP is a program that implements much of what PEM is... why not make PGP PEM compliant. 2. Propose IDEA as a symmetric algorithm for PEM. 3. Ha! PGP already has key servers. 4. Propose a revion to the certification scheme where USER certificates would be created by the owner and signed by non-certificate-authority acquaintances ala PGP. Yes, this would take time and effort. No, this should not be taken as an affront to our current and previous efforts. I think that we should persue _every_ avenue. If the only real problem with PEM is the trust model, and we can change that, then this would be a strongly legitimizing action. Scott Collins | "Few people realize what tremendous power there | is in one of these things." -- Willy Wonka ......................|................................................ BUSINESS. voice:408.862.0540 fax:974.6094 collins(a)newton.apple.com Apple Computer, Inc. 1 Infinite Loop, MS 301-2C Cupertino, CA 95014 ....................................................................... PERSONAL. voice/fax:408.257.1746 1024:669687 catalyst(a)netcom.com

1 0

Re: Commercial PGP: Verifying Trustworthiness
by doug＠netcom5.netcom.com 31 Aug '93

31 Aug '93

--- Forwarded mail from Marc Horowitz <marc(a)Athena.MIT.EDU> >From owner-cypherpunks(a)toad.com Mon Aug 30 23:40:01 1993 Received: from relay2.UU.NET by mail.netcom.com (5.65/SMI-4.1/Netcom) id AA14421; Mon, 30 Aug 93 23:39:57 -0700 Received: from toad.com by relay2.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA10745; Tue, 31 Aug 93 02:40:50 -0400 Received: by toad.com id AA14781; Mon, 30 Aug 93 23:33:56 PDT Received: by toad.com id AA14701; Mon, 30 Aug 93 23:31:26 PDT Return-Path: <marc(a)Athena.MIT.EDU> Received: from Athena.MIT.EDU ([18.72.1.1]) by toad.com id AA14688; Mon, 30 Aug 93 23:31:23 PDT Received: from OLIVER.MIT.EDU by Athena.MIT.EDU with SMTP id AA00837; Tue, 31 Aug 93 02:28:59 EDT Received: by oliver.MIT.EDU (AIX 3.2/UCB 5.64/4.7) id AA14903; Tue, 31 Aug 1993 02:28:52 -0400 Message-Id: <9308310628.AA14903(a)oliver.MIT.EDU> To: bbyer(a)BIX.com Cc: honey(a)citi.umich.edu, cypherpunks(a)toad.com Subject: Re: Commercial PGP: Verifying Trustworthiness In-Reply-To: Your message of Tue, 31 Aug 93 00:14:18 -0400. <9308310014.memo.72462(a)BIX.com> Date: Tue, 31 Aug 93 02:28:52 EDT From: Marc Horowitz <marc(a)Athena.MIT.EDU> Marc Horowitz <marc(a)Athena.MIT.EDU> said: >> I dunno. The early versions of UNIX had a back door in the login [...] >I've let a lot of stupid comments go by, but I have to respond to this one. > >It is true that Dennis Ritchie (I believe, if not him, one of the >other original UNIX authors) proposed such a login/compiler virus. >But it wasn't in any early version of UNIX. Stupid? Watch the flame bait...he merely overstated a touch. The back doors weren't part of any of the full distributions, it's true, but they were quite a bit more than proposals. Ken Thompson actually distributed those back doors via a compiler update, warning of a security problem and urging all sites to recompile. Most did, which inserted the back doors into the programs. That's close enough to the original claim. See the Ken Thompson & Dennis Ritchie Turing Award Lecture, which goes into detail about this. The level of sneakiness involved was amazing. Compilers are the ultimate security breach. Doug

1 0

Re: Apple, privacy, and AOCE
by collins＠newton.apple.com 31 Aug '93

31 Aug '93

>I found the idea that RSADSI will be generating folks' key pairs >particularly chilling. What I gathered from actually using this software is that you personally generate a key pair, on your own machine, and then transparently send your public key to RSADSI. Some time later, you receive a certificate (with an expiration date) that allows your 'signer' to function. RSADSI does not make, or even see, your private key. >The article accompanying these sidebars suggests that folks' private keys >will be stored on the server; My understanding is that address books on the [optional] servers may have copies of certificates, for people who have certificates and want them published. >the article [...] must be the result of miscommunication yes >The NSA recently signed an agreement with the Software Publishers >Association that will provide expedited approval of RC-4 encryption based >on 40-bit keys. Not surprising, since a pre-computation attack allowing a direct key lookup against RC-4 with 40 bit keys is economically feasible for anyone who can afford a CD-ROM jukebox (128 mips-years of computation + 8 terabytes of storage). >NSA [...] will allow slightly more-powerful scrambling capabilities >[in AOCE] AOCE uses 64-bit keys, and larger keys mean better security. This could mean anything. They might actually be using 64 bit keys (which would be good, although 80 bits is recommended), or they might be using 40 bit keys with 24 bits of salt (or worse: 32 and 32). Salted keys (key+salt of sufficient size), stop the precomputed attack, but if the actual key size, without salt, is still only 40 bits, then exhaustive search of the keyspace, after the salt has been seen, will only take 64 mips-years. Scott Collins | "Few people realize what tremendous power there | is in one of these things." -- Willy Wonka ......................|................................................ BUSINESS. voice:408.862.0540 fax:974.6094 collins(a)newton.apple.com Apple Computer, Inc. 1 Infinite Loop, MS 301-2C Cupertino, CA 95014 ....................................................................... PERSONAL. voice/fax:408.257.1746 1024:669687 catalyst(a)netcom.com

1 0

Internet commerce...for your cracking pleasure
by nobody＠shell.portal.com 31 Aug '93

31 Aug '93

================================================= INDIVIDUAL INTERNET SUBSCRIPTIONS AVAILABLE! Have you ever wanted the convenience of the news delivered directly to your e-mail box? Now its possible! ...USA Today....The Moscow News is available for just $80 a year. For more information, or to place an order send an e-mail to subscrib(a)americast.com Include the e-mail the address to which you want the services delivered. In that e-mail please designate the paper and sections you would like to receive, give your full name, address, phone number, and the credit card (Visa or MasterCard) you will be using for the purchase, and the credit card expiration date.

1 0