exit node only server

Nick Mathewson nickm at freehaven.net
Sat Mar 5 00:21:12 UTC 2005


On Sat, Mar 05, 2005 at 01:39:14AM +0600, Vlad SATtva Miller wrote:
> On 05.03.05 01:22, clifnor at warpmail.net wrote:
> > Some people are almost always greedy
> > Some people are sometimes greedy
> > No one is NEVER greedy
> 
> And some just using GPRS and other low-bandwidth connection (mobile,
> especially with per Mb payment) prohibiting them to pass others'
> traffic. In tit-for-tat scenario all those people will have to abandon
> Tor network at all.

As has been mentioned, such a system assume a lower bound of service
that servers provide to everybody.  That way, low-bandwidth users, or
users who can't run Tor nodes, still get some service.  (More users
will help provide more anonymity against many adversaries, after all.
True, they'd get bad bandwidth --- but low-bandwidth users don't need
much bandwidth, and the choice may be between either creating
incentives for server operators by providing worse-but-existent
service to non-servers, or having too few servers, so nobody gets
anonymity.)

Of course, this isn't tit-for-tat anymore (at least, not the tit-for-
tat that Rapoport came up with and Axelrod helped investigate).  In
straight tit-for-tat, you respond with exactly the behavior you saw in
the round immediately before the present round.  I think that most
people who are using the term here mean it in some more vague and
extended sense.  (i.e., 'Y'know, like what Bittorrent does, or
something, except localized, and with bandwidth relayed instead of
chunks uploaded, but quantized somehow, and not broadcast, and...')

For more info, you might want to see the "Challenged" paper linked to
in my last email.

Yours,
-- 
Nick Mathewson
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 652 bytes
Desc: not available
URL: <http://lists.torproject.org/pipermail/tor-talk/attachments/20050304/a2ec0e98/attachment.pgp>


More information about the tor-talk mailing list