> > Er, but if everybody did that Freenet wouldn't work.  A fundamental
> > part of Freenet's adaptive mechanism is the "path compression" which
> > comes from nodes using the DataSource field.  Removing this
> > functionality from your node would probably mean that the presence of
> > your node is doing more harm than good in the Freenet network.
> There is always a trade-off between security and efficiency. Talking to
> unknown nodes is a security risk. It is absolutely essential that you give
> people the option for as much security as they want. And Freenet certainly
> would work if everyone did this, but it would be significantly
> slower. That's the tradeoff. I'm not advocating that everyone do this, but
> that it be an option. The Freenet design works by being pulled
> simultaneously in the two directions of efficiency and security.

I disagree, it is not just a question of degrees, Freenet *depends* on
the path compression functionality, giving people the option of
switching it off would effectively mean that they are not implementing
the Freenet protocol, merely exploiting the network, gaining the
advantages without contributing back.

Besides, I am sceptical about the security benefits of this anyway. 
Do you seriously thing Joe public is going to go to the trouble of
identifying other Freenet users, confirming that they are trustworthy
(how?), and then getting their public key through non-electronic
means?  Even the most paranoid wouldn't be bothered to do that.

The best defense against corrupt nodes is to firstly ensure that it
is difficult for someone to deliberately get your machine to use a
corrupt node (which it is, it would mean that one of your requests
for data had to go through another node which they controlled, *and*
that no node further down the line changed the DataSource randomly). 
Secondly, even if you did connect directly to a corrupt node, you
could maintain deniability about who initiated the messages which
were sent on to that node, which you can currently do in Freenet. 
Neither of these precautions require any action on the part of the
user, where as your proposal would require a degree of effort on the
part of the user in terms of obtaining the addresses of trusted nodes
which even I would be reluctant to do (just think how difficult it
would be to do this living under an oppressive reigeme (in a pub:
"hey mate, you run a freenet node?...").

Ian.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: 
<https://emu.freenetproject.org/pipermail/devl/attachments/20000819/d3b627b4/attachment.pgp>

Reply via email to