DC-Nets and sheep

[rishab@dxm.ernet.in]

doug@OpenMind.com (Doug Cutrell)
> One possible way around this societal control of anonymous remailers might
> be the existence of very large DC-nets (or related technologies).  The idea

I've been arguing that DC-Nets are among the crypto protocols that we've not
exploited much so far. I was working on an implementation, till I got stuck
with the 'net' part of it. The basic protocol is very simple, with hooks to the
bit-flipping crypto routines. The problem is integrating the DC into an 
existing net protocol in a manner that makes it transparent and practical to 
use. Ideally a layer over IP or Ethernet, as DC-Nets share many of the basic
problems of net protocols - conflict resolution, random wait-before-resend etc.

Ideas? (Read section 13.4.8 of Tim's Cyphernomicon for an explanation of 
DC-Nets)

Note that the major problems lie in the _practical_ implementation, which is
not necessarily the speed - despite conflict resolution and anti-collusion
techniques, the basic operation in a DC-Net, assuming the presence of a stream
of random numbers, is the fast XOR. DC-Net implementation problems are more
severe than those in secure-IP (swIPe etc) as they have to handle lots of 
things at a lower level in the network protocol.

Doug goes on to suggest that to be immune from the "sheep^H^H^H^H^Hpeople"
DC-Nets will have to have millions of members. But if anon remailers were used
by millions, than they'd be immune too. All crypto is vulnerable to mob action
until it's widespread.

-----------------------------------------------------------------------------
Rishab Aiyer Ghosh             "Clean the air! clean the sky! wash the wind!
rishab@dxm.ernet.in                   take stone from stone and wash them..."
Voice/Fax/Data +91 11 6853410  
Voicemail +91 11 3760335                 H 34C Saket, New Delhi 110017, INDIA