[mnet-devel] Anonymity and Performance in Peer-to-Peer Systems

[Some Guy]

This post is supposed to get to Nikita Borisov and Jason Waddle regarding:
http://www.cs.berkeley.edu/~nikitab/projects/p2p/proposal.html

Part 1:
"Research a formal metric for the anonymity of a system. We are currently performing a literature
search; if we fail to find a suitable metric in past research, we will propose a new one."

Here's a nice paper that analizes the annonymity of Crowds:
Probabilistic Analysis of Anonymity
Vitaly Shmatikov
SRI International
http://citeseer.nj.nec.com/shmatikov02probabilistic.html

"We use discrete-time Markov chains to formally model the behavior of group members
in Crowds [22], specify anonymity properties of the system as temporal logic formulas, and use a
probabilistic model checker to verify them. This analysis method is applicable to other 
probabilistic systems such as Freenet."


Part 2:
DOS Resistance: I've got a basic handle now on DOS resistance vs. peformance and I can almost get
a faint glimmer of what the a "perfect" DHT should look like.  I've posted my ideas on it to the
Freenet list and I'll post them to the MNet list today too.


If anyone wants to discuss how to build a better Freenet, I'd LOVE to help.




__________________________________________________________________

Gesendet von Yahoo! Mail - http://mail.yahoo.de
Logos und Klingeltöne fürs Handy bei http://sms.yahoo.de


-------------------------------------------------------
This SF.net email sponsored by: Enterprise Linux Forum Conference & Expo
The Event For Linux Datacenter Solutions & Strategies in The Enterprise 
Linux in the Boardroom; in the Front Office; & in the Server Room 
http://www.enterpriselinuxforum.com
_______________________________________________
mnet-devel mailing list
mnet-devel at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mnet-devel