Re: Hallam-Baker demands more repudiations or he'll write!

At 04:51 PM 9/26/96 -0700, Declan McCullagh wrote:
Anonymity and nonescrowed crypto are the linchpins of AP and its more general case, Maysian crypto anarchy. The withering of the nation-state. Whatever you want to call it. To prevent it, governments will ban both. A criminal law, passed in the wake of say a bombing this fall in Washington, DC, banning nonescrowed crypto. (Freeh will assert he has evidence the terrorists used PGPhone.) And another law banning online anonymity. What then, Mr. Bell?
While we should continue resist such developments, I (like many others) tend to believe that the government is going to have a great deal of difficulty implementing such restrictions, for reasons which have been discussed here ad nauseum. I happen to believe that the really crucial reason Clipper was proposed is so that it would exclude non-escrowed encryption using market forces, because they understood that banning encryption or non-escrowed encryption would be essentially impossible. Remember the VHS/Beta VCR wars? With a fairly equal market in about 1978, Beta died 10 years later because the market couldn't support two incompatible standards. It wasn't that one was dramatically better than the other, it was simply that having two standards forced the market to duplicate stocks, for the machines as well as tapes, particularly pre-recorded tapes. Notice, however, that VCR's are relatively "isolated": It doesn't really matter if you have one format and your neighbor has another, unless you want to swap tapes. But crypto telephones inherently require (in the long term) full intercompability. If you didn't have that, there'd be half a world of people you couldn't call! (This kind of situation is remeniscent of a recollection my mother has of a small town she grew up in, in which there were two competing telephone companies serving the same area, but customers of company "A" couldn't call company "B." The most popular girl in town was the daughter of a doctor, who as a consequence of his profession had telephones from both companies. That girl was constantly being called to relay information between people on opposite telephone companies!) The government understood this, and realized that the best way to derail the advent of a good crypto telephone standard was to produce one of their own, which they figured would pre-empt and kill off any competing system. It turns out that an essential element to this killing-off process is to ensure that the telephones aren't compatible, so I was a bit surprised to have to wait a long time to hear that one of the requirements for Clipper-phone approval is that they NOT co-operate with non-escrowed telephones. (I was surprised to have to WAIT to hear this, NOT that they said it eventually...) I think they realize that Clipper was their last opportunity to implement a de-facto market ban on good encryption, but it isn't working. As for a ban on anonymity I don't think that'll fly either. There are too many interactions we do today (off the 'net as well as on) which are already anonymous, even if we don't normally think of it that way. Visit a theater, pay cash, and nobody records your name, for instance. Buy groceries, pay cash. Throw coins in toll booth. Put coins in Coke machine, ride on bus, etc. It would be very hard to prohibit non-anonymous interactions on the net, with so many opportunities for unrestricted anonymity. There's even less reason to ID non-economic transactions too. Since most interactions on the 'net are not a financial transaction, there is even less reason to identify the people involved. So at the risk of being overly optimistic, I think the government isn't going to be able to pull off the kind of anti-crypto/anti-anonymity coup you describe. But nevertheless, I think that if they "progress" towards that goal against the odds, that could easily be an excellent justification for actions which you know they'll describe as "terrorism." Jim Bell jimbell@pacifier.com

At 8:22 PM -0800 9/26/96, jim bell wrote:
Remember the VHS/Beta VCR wars? With a fairly equal market in about 1978, Beta died 10 years later because the market couldn't support two incompatible standards. It wasn't that one was dramatically better than the other, it was simply that having two standards forced the market to duplicate stocks, for the machines as well as tapes, particularly pre-recorded tapes. Notice, however, that VCR's are relatively "isolated": It doesn't really matter if you have one format and your neighbor has another, unless you want to swap tapes. But crypto telephones inherently require (in the long term) full intercompability. If you didn't have that, there'd be half a world of people you couldn't call!
Be careful when drawing conclusions from the "VCR Wars." (BTW, James Lardner has a book out on this subject.) There are various interpretations of what happened. While I, for example, had both formats, I ultimately shelved my Beta unit. Some views often heard: * The longer recording time of VHS was more compelling to most consumers than the higher quality of Beta. Not surprising in the late 70s, early 80s, when most consumers had televisions incapable of showing the difference in quality. (Even today, most consumers are happy to rent VHS tapes which are rather "dodgy" in quality.) * VHS was "Pretty Good Video," and PGV was enough for its time. * Once VHS took a lead over Beta, the snowball effect took hold. Video rental stores started to appear in earnest in around 1980, and by then VHS had enough of a lead over Beta to cause the stores to stock VHS tapes for rental over Beta tapes. This dramatically widened the lead of VHS over Beta. (Jim's point: "Notice, however, that VCR's are relatively "isolated": It doesn't really matter if you have one format and your neighbor has another, unless you want to swap tapes." But in fact a large fraction of all VHS owners use their machines to _rent_ tapes, so the compatibility with what the many rental stores carry is paramount.) A tenuous link to crypto is that various VCR formats can still intercommunicate because they all use NTSC (or PAL, SECAM in some countries) as the "common language." This is analogous to the way various flavors of PGP on various platforms can communicate with other flavors because ASCII text is read and written by all. As I have argued many times, this was really the Big Win for PGP, that it did not use an odd or proprietary format that was platform-specific. Such basic ASCII operation ensures interoperability, and is of course inconsistent with the government talk of making sure that key-escrow products cannot interoperate with non-key-escrow or "alternately"-escrowed products. --Tim May We got computers, we're tapping phone lines, I know that that ain't allowed. ---------:---------:---------:---------:---------:---------:---------:---- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay@got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Higher Power: 2^1,257,787-1 | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."
participants (2)
-
jim bell
-
Timothy C. May