--- [EMAIL PROTECTED] wrote: 
> I have an idea.
> 
> The datastore does pcaching on all data that goes through your node and does not 
> know whether or
> not it was you that requested the data. Then have a second datastore that cache all 
> of your data
> requests this should be encrypted. However because there is no way of guaranteeing 
> that the key
> does not get swapped out, advice people to do this at an OS level as well. (IE: keep 
> your cache
> on a rubberhose partition) However when you request a multipart file, you use premix 
> routing to
> make all the data requests come through a single node that does not know you. Then 
> it should
> store all the data that you requested in it's second cache too (and 
> probabilistically in the
> first). That way, even if someone is able to perform a timing analysis on your first 
> cache, they
> have no way of knowing whether it was you or someone that connected to you that 
> downloaded the
> data. (Also it should be quite hard to pull off because it is always probabilistic). 

A couple people including Toad and I seem to have this idea too.

On mistake up there: "(and probabilistically in the first)" don't do that at all.  
Even 10% of a
couple split files might be too much.

> Then they break down your door and steal your computer, and you are unfortunate and 
> at the time
> some of the keys were swapped out and written on the disk (any you weren�t prudent 
> enough to use
> a secondary encryption tool), they can read some of your cache, but they still can't 
> prove that
> it was actually you that requested the data.

"I was framed!!!" :-) I just like saying that.


__________________________________________________________________

Gesendet von Yahoo! Mail - http://mail.yahoo.de
Logos und Klingelt�ne f�rs Handy bei http://sms.yahoo.de
_______________________________________________
Devl mailing list
[EMAIL PROTECTED]
http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to