Crazy idea: How trust in darknets enables secure democratic censorship

Matthew Toseland <toad-EI5O+8PHWbJeeLb3ft/[email protected]> wrote:

> > When a node drops data have it remember the hash for a *long* time. If
> > somebody tries to reinsert the data just act as if the insertion was
> > successfull and ignore it.
> 
> What if they just insert it locally on their own node and then propagate
> it by requests? 

If my understanding of request routing is correct a request will be
routed toward the region of the net where the requested data *should*
reside. 

They will be able to insert their data on their own node.
To propagate the data they have to make requests to other nodes. Those
requests will be routed toward the region of the net where the requested
data *should* reside and fail there.

The only exception is when they control nodes that are right where the
data should reside. Then they can insert locally and propagate the data
with a few requests from random nodes.

Unless they operate lots of nodes they won't be able to do this for a
relevant portion of their data. 

> We can't interfere with requests propagating data without dire
> consequences, and only dealing with inserts would be impotent.

I don't suggest to change the behaviour of requests. 

If an insert is unable to place the data where requests will find it the
requests will fail and the data will not propagate. 
_______________________________________________
chat mailing list
[email protected]
Archived: http://news.gmane.org/gmane.network.freenet.general
Unsubscribe at http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/chat
Or mailto:[email protected]?subject=unsubscribe