Re: sooner or later

[Wei Dai <weidai@eskimo.com>]

On Fri, 19 Sep 1997, Bill Stewart wrote:

> Also, "Later" might mean "Clinton's out of office, and some of the
> Republicans have gotten in the habit of pretending they like
> privacy as long as Clinton opposes it, even though it's traditionally
> been the Republicans' job to rip off our privacy."

I don't think it makes a difference who is in office. The TLAs seem
incredibly adept at converting/subverting politicians who initially
support crypto. They must have known about this ability (and by ability I
don't mean some kind of mind-control machine, but simple persuasiveness) 
for a long time, but seem to have started excercising it on a large scale
only recently. The pro-crypto lobby is pathetic by comparison. Has it
converted any anti-crypto politician to our side? This is why I think a
crypto ban is inevitable. We simply don't have the resources to defend
against this type of attack directly. A delay is possible, but not
one long enough to make the ban impossible.

As for the residual effects of the ban after it is lifted, I think you are
overestimating them. All (escrowed) crypto built during the ban should be
designed with the lift in mind. When the ban is lifted, everyone will be
able to upgrade simultaneously by simply plugging in non-escrowed crypto
and protocol modules. This can even be done without user-intervention,
similar to auto-upgrade of virus scanning modules. Compatibility betwen
escrowed and non-escrowed crypto can be kept during the upgrade period
with appropriate negotiation protocols.

The residual social effects of the ban is harder to estimate. It's
possible that it will be minimal, for example if digital bearer
instruments are so much more efficient than account based ones that people
will use them despite escrow requirements, then these can be quickly
converted to use non-escrowed crypto after the ban is lifted. But in any
case, I don't see how a short delay will make any difference.