On Tue, May 16, 2006 at 05:32:26PM -0700, Ian Clarke wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > On 16 May 2006, at 17:05, Matthew Toseland wrote: > > >Because of the literalist (and possibly stupid) way we have > >implemented > >the swapping algorithm, a lot of information on swap attempts is > >exposed. We can already do a size estimate on the local network from > >this. I propose to: > >- Include the size estimate in the node. > >- Produce a histogram of the path lengths *on the actual darknet*. > >- Produce a graph of the network topology on the actual darknet. > > > >The last item is somewhat harder than the first two, but the first two > >are fairly straightforward. > > Surely the last thing should be very difficult and somewhat defeats > the point of a darknet, how is that possible?
There is no end to end crypto on swap requests. There can't be, because they are routed randomly. Any intermediary can easily see the exchanged data, which consists of 1) my location, 2) his location, 3) my friends' locations, and 4) his friends' locations. Right now, the network size estimates are based purely on swaps which we are an endpoint of. We can send less information, however doing so would probably make spoofing harder to detect. Spoofing is trivial in any case; to secure swaps we need to expose the local network topology (which would also let us do premix routing), but this IMHO is not a priority until 0.8. > > Ian. -- Matthew J Toseland - toad at amphibian.dyndns.org Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: <https://emu.freenetproject.org/pipermail/tech/attachments/20060517/273df104/attachment.pgp>
