On Tue, Aug 15, 2006 at 11:48:35AM -0700, Ian Clarke wrote: > >Secondly, it is entirely reasonable to expect that a true darknet > >and an > >opennet will be two separate small world networks, which have > >completely > >different properties. Requests on the darknet would benefit from > >checking the darknet first, because data is often cached within the > >darknet (due to related interests, for example), and a darknet will > >often > >have very few links to the opennet, so there is a major performance > >penalty for the transition. > > I don't think there will be a distinct separation, I think most > darknet nodes will be just a few hops from the opennet.
In some places yes. In others no. I'd hope that there will eventually be a substantial chinese darknet, for example. Most of this would have to be some distance away from the opennet. > > >What I propose is that requests on the > >attached darknet are routed first within it, and then forwarded out to > >the wider opennet. > > We have gone through the trouble of coming up with an algorithm that > should be able to figure out how to route around the network - why > interfere with its routing decisions? Because in many cases the network we provide it with is not a single small world network (which is what it is designed for), but two loosely connected small world networks of different parameters. > > >On the other hand, opennet requests which get routed > >to an attached darknet will often be falling down a black hole, for > >the > >same reason: That there is limited connectivity between the two > >networks. Our location swapping algorithm will work if it is > >running on > >a small world network. I doubt very much that it will work well if > >it is > >run on two small world networks which are not closely connected. > > Firstly, I think they will be closely connected. Not in all cases. Any substantial true darknet will mostly be some distance from the opennet, especially if it is operating in a hostile environment. > Secondly, you are > second-guessing our entire routing algorithm, the whole point of > which is to figure out how to route around a connected network. Our routing algorithm is designed for a small world network. There is no reason to think that the combination of the chinese darknet and the opennet will be a small world network. And it doesn't matter if the attached darknet is small; the argument applies even more so then. The question is simply how closely it is connected to the opennet. > You > are proposing a complicated solution to a problem that we don't even > know exists, and IMHO probably won't exist. -- 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/devl/attachments/20060815/94446b78/attachment.pgp>
