December 2003 - cypherpunks-legacy

anonymous cash without blinding
by Wei Dai 17 Dec '03

17 Dec '03

With all this talk about Chuam patents, I would like to remind people that blind signatures are not absolutely essential to an anonymous digital cash system. You can combine a traceable cash system with an anonymous communication system in a fairly obvious way to get a fully (both payer and payee) anonymous cash system. Suppose a bank is running a digital cash system that works like this: it maintains a database of valid coins, and whenever someone presents it with a valid coin (string of bytes) it erases that coin from its database, and then either gives the person an equivelent amount of paper cash or a newly created coin(s) of the same value. Now if the bank allows this exchange of old coins for new coins to be done over an anonymous network (e.g., a remailer-net), then the system is anonymous as long as you don't move physical money in or out of the system. Maintaining anonymity when moving physical money in and out of the system is what blinding helps you to do, but this will be less useful in a fully digital economy where such movement will be infrequent. Wei Dai

1 0

Re: NEW Netscape RNG hole
by an402976＠anon.penet.fi 17 Dec '03

17 Dec '03

Phil argues this file handle being lost isn't a big mistake. He describes how the function is actually called, which does indeed show that it shouldn't be much of a problem. However, Netscape had not revealed enough information about their RNG to allow myself or other reviewers to determine how critical it was. If, for example, this seeding function were called once every time a secure connection were established, losing a handle would be a major problem. This seems like a good reason to ask for the code for SEC_RandomUpdate(). You show us from what sources you gather bits, but you don't show us how you mix them or, for that matter, stream out "random" bits. If you did have a description in your original published code that was better than "mixing is accomplished with MD5", I must've missed it. RingZero =========== From: Phil Karlton <karlton(a)netscape.com> Subject: Re: NEW Netscape RNG hole Date: Sunday, October 08, 1995 1:39AM RingZero wrote: > > Did anyone else notice a bug in the new, public Netscape > RNG code? It appears that on Windows builds, during the > RNG seeding, the function that hashes in file contents > (EnumSystemFiles) doesn't close a file handle (lFileHandle). I think you mean lFindHandle. I'm not a windows programmer, so I have no idea if the enumerator needs to be cleaned up, but I will forward your message to the appropriate folks here. [...] --****ATTENTION****--****ATTENTION****--****ATTENTION****--***ATTENTION*** Your e-mail reply to this message WILL be *automatically* ANONYMIZED. Please, report inappropriate use to abuse(a)anon.penet.fi For information (incl. non-anon reply) write to help(a)anon.penet.fi If you have any problems, address them to admin(a)anon.penet.fi

1 0

Re: Netscape 2.0 beta
by Raph Levien 17 Dec '03

17 Dec '03

I just got Netscape 2.0 beta for DEC Alpha. Here are my initial impressions: * Still no Java support (for DEC Alpha) :-(. * I saw no hints of S/MIME support. Did I miss something? Do the binaries for other platforms support S/MIME? I have access to HP, and will probably demand a Sun account from the powers that be here, just so I can run Java. I _really_ want to test S/MIME. * The mail reader facility looks pretty good. The smooth integration of MIME and HTML objects is really quite nice. On the other hand, there are a few UI quirks that will no doubt get ironed out over the coming months. * The .mailcap parser _still_ doesn't recognize %{variable} style entries. Thus, I wasn't able to test Michael Elkins' PGP/MIME format for compatibility with Netscape. The %{} syntax is in the spec (RFC 1524), guys. * To send mail, Netscape apparently connects to the SMTP server running on localhost. I haven't confirmed this. Thus, I don't see any way to integrate Netscape and premail :-(. Guess I'll just have to wait for Netscape script (from what I've seen so far, it's not possible in pure Java). * No user-defined headers from mail :-(. Thus, I couldn't put an "In-Reply-To: " header in this followup. * News is also not bad, but I find I can't zip through tons of low S/N newsgroups the way I can with trn. I tend to rate newsreaders in number of messages deleted per second. I think with a bit more tweaking on the UI, it could really sing. * One slick feature of news is its ability to decode binary postings, apparently in both MIME and uuencode formats. However, I wasn't able to figure out how to decode multiparts. * The UI render code has a number of cosmetic defects (possibly related to the DECness of my machine). One new one is that drag-and-drops leave the Bookmarks window in an inconsistent state, leading one to want to press ctrl-L :-). Overall, nicely done. Mozilla will clearly remain the standard in Web browsers, for the next few months anyway. Raph

1 0

Re: Netscape 2.0 beta
by schampeo＠imonics.com 17 Dec '03

17 Dec '03

Raph Levien wrote: >* To send mail, Netscape apparently connects to the SMTP server > running on localhost. I haven't confirmed this. Thus, I don't > see any way to integrate Netscape and premail :-(. Guess I'll > just have to wait for Netscape script (from what I've seen so > far, it's not possible in pure Java). You have to set this in the preferences dialog. The default is localhost, assuming that your UNIX box is set up correctly to send mail. I don't know about you guys, but my mailhost is called 'mailhost'. One thing that I found with the newsreader was that it doesn't do POP correctly. No matter what you do with your mail, it still doesn't check for read vs. unread articles. This is fine when you have a mailbox with one message. It can be annoying when you have 110 messages in your mailbox and don't want to 1) delete them or 2) download them twice. Not that I'm complaining. I'm still gigling from the introduction of frames and client-side imagemaps (which also don't work on the Solaris 2.4 version, running on Solaris 2.5beta/CDE)... Steve -- Steven Champeon (919) 469-7833 Technical Lead, schampeo(a)imonics.com Imonics Web Services webmaster(a)imonics.com

1 0

[NOISE] Caution! Netscape 2.0 and Linux
by Christopher J. Shaulis 17 Dec '03

17 Dec '03

Hi all. Just a warning to steer clear of Netscape 2.0 on Linux. It can't resolve FQDNs, its distributed as a staticly linked a.out binary *AGAIN*, and I'm hearing lots of reports sitting here on IRC that it locks up R6 versions of XFree solid on a random basis. Not to mention that there is no java support yet for Linux, and the Java support in the Sun/Windows version of netscape is based on the new (and imcompatible with earlier versions) beta version of Java which sun pimped out to Netscape and won't release the source code too. The part about not resolving FQDNs might be an undocumented feature. The most secure transaction is one that doesn't take place at all. ;) Have fun, Christopher

3 3

Re: Graphic encryption
by patrick＠Verity.COM 17 Dec '03

17 Dec '03

> From: Andrew Loewenstern <andrew_loewenstern(a)il.us.swissbank.com> > > > Is there anything particular in graphic encryption? I usually > > encrypt graphics and document images as regular files with regular > > encryption... > > I believe graphic encryption outputs a valid image file that is apparently > white noise until you perform the decryption transformation on it. An > obvious way to do this with a non-lossy file format is to encrypt pixel vales > with a stream cipher. Another way to make the image unviewable would be to > shuffle the pixels or rasters with a PRNG. > I thought it was the ideas presented in Dr Dobbs last year of encoding other information in a graphic image so that the image still looked the same, but the other information could be extracted...i.e. using the image as a covert channel. Does anyone know what they're really talking about? Patrick _______________________________________________________________________ / These opinions are mine, and not Verity's (except by coincidence;). \ | (\ | | Patrick J. Horgan Verity Inc. \\ Have | | patrick(a)verity.com 1550 Plymouth Street \\ _ Sword | | Phone : (415)960-7600 Mountain View \\/ Will | | FAX : (415)960-7750 California 94303 _/\\ Travel | \___________________________________________________________\)__________/

1 0

Re: subjective names and MITM
by patrick＠Verity.COM 17 Dec '03

17 Dec '03

> From: Scott Brickner <sjb(a)universe.digex.net> > > Patrick Horgan writes: > >Given the difficulty of finding primes, how likely do you think it is that > >given one of the well known methods and finding the first 1024 bit prime > >that pops out would give you an effective attack? > > I'm not an expert here, but I understand the "well-known methods" to > essentially use some formula that "tends" to generate prime numbers from > uniformly distributed numbers, feed it a "good" random number, and then > check to see if it's really prime. If it's not, pick another "good" > random number and try again. The entropy in the prime is the same > as in the random number generator. > Can someone give me pointers to literature on this? Books prefered. Patrick _______________________________________________________________________ / These opinions are mine, and not Verity's (except by coincidence;). \ | (\ | | Patrick J. Horgan Verity Inc. \\ Have | | patrick(a)verity.com 1550 Plymouth Street \\ _ Sword | | Phone : (415)960-7600 Mountain View \\/ Will | | FAX : (415)960-7750 California 94303 _/\\ Travel | \___________________________________________________________\)__________/

1 0

Re: cypherpunks digicash bank?
by cman＠communities.com 17 Dec '03

17 Dec '03

>Douglas Barnes <cman(a)communities.com> writes on cypherpunks: >> I'm afraid you may have somewhat misunderstood the motivation behind >> the Identity Agnostic paper. In no way is it intended as a way of >> not facing the music wrt regulators in the country(s) where such an >> institution has offices. The IA approach is intended as an possible >> alternative for an institution that might otherwise license from >> Chaum. > >So it's intended for avoiding patent issues only? Yes, and not really so much because of price. I'm not the only person who has found David hard to work with -- he has apparently decided to only license "respectable" organizations that will not, in his eyes, make his technology look "bad". I can tell you from painful, expensive experience that a band of cypherpunks does not qualify as "respectable" in his eyes. Alas for David, his technology is fundamentally cypherpunk -- there is little motivation for a "respectable" institution to be the first penguin on this stuff. So, if you're a bank, licensing from David is cheap and relatively painless, but there are powerful disincentives to deployment. Also, assessing demand is problematic. For a small cypherpunk startup, the demand is obvious, there are powerful incentives, but the cost of licensing and David's reluctance to license a slightly wild-eyed startup act as a barrier. There is certainly a middle ground, but nobody seems to be stepping forward. My understanding was that we were exploring the idea of a much smaller, non-bank organization going into the digital cash business. Note that while this is theoretically interesting to me, the fact that I'm talking openly about this stuff is largely due to my decision _not_ to get personally involved in this kind of venture except as a sidelines cheerleader, or as an absolute last resort if nobody deploys. Given that we're talking about a small, cypherpunk-motivated NBFI, I think the IA stuff will work fine. It should be presented to the regulators as if it were a fully-blinded system, since there won't be any guarantee that blinding isn't occurring. You raise the issue of using software not supplied by the "bank" -- if we're talking about a c'punks project, I'd assume you'd make the full source code of the system available. The change of a client from non-blinding to blinding would be about two lines of code. You've also discussed the issue of commercial uptake -- as I see it, one of the primary goals of a project like this would be to provide incentive to deploy to the allegedly numerous Chaum licensees that have not implemented a damned thing so far. And, with a nice interface -- or at least the sort of generic GUI interface you can get with Java -- it might do surprisingly well on its own merits. Realistically, as soon as someone large comes along, you're likely to get squashed. But it would be an interesting ride, and you might just end up getting bought out instead of squashed. As for a limit on transaction size, my understanding is that Mondex is using $300 initially as a transaction limit. This certainly seems reasonable, and it skates in underneath a lot of magic numbers.

1 0

anonymous cash without blinding
by Clifford Neuman 17 Dec '03

17 Dec '03

Date: Sun, 8 Oct 1995 14:20:27 -0700 (PDT) From: Wei Dai <weidai(a)eskimo.com> Now if the bank allows this exchange of old coins for new coins to be done over an anonymous network (e.g., a remailer-net), then the system is anonymous as long as you don't move physical money in or out of the system. Maintaining anonymity when moving physical money in and out of the system is what blinding helps you to do, but this will be less useful in a fully digital economy where such movement will be infrequent. See: Gennady Medvinsky and B. Clifford Neuman. NetCash: A design for practical electronic currency on the Internet. In Proceedings of 1st the ACM Conference on Computer and Communication Security November 1993. For a discussion of the tradeoffs for this weaker form of anonymity. You can find postscript for this paper as: ftp://prospero.isi.edu/pub/papers/security/netcash-cccs93.ps

1 0

Re: Two bugs in CERN3.0 HTTPD
by patrick＠Verity.COM 17 Dec '03

17 Dec '03

> From: Ray Cromwell <rjc(a)clark.net> > > Suffice it to say, the CERN code is *littered* with temporary local > stack arrays, and I'm guessing that many of them have faulty bounds > checking (or none at all) > > -Ray Three cheers for C++ which let's us make types without these problems:) Patrick _______________________________________________________________________ / These opinions are mine, and not Verity's (except by coincidence;). \ | (\ | | Patrick J. Horgan Verity Inc. \\ Have | | patrick(a)verity.com 1550 Plymouth Street \\ _ Sword | | Phone : (415)960-7600 Mountain View \\/ Will | | FAX : (415)960-7750 California 94303 _/\\ Travel | \___________________________________________________________\)__________/

1 0