On Thu, Aug 31, 2000 at 12:33:11AM +0700, Oskar Sandberg wrote: > On Wed, Aug 30, 2000 at 10:56:06AM -0500, david at aminal.com wrote: > > > > I don't understand what you mean by 'reversing the datastore vector'. If > > you have time, could you expand on that a bit? > > When requesting, the reference in the datastore is set back towards the > source of the data. This reference is towards a node that was achieved > after more sets of routing then the current node. > > When inserting, it currently sets the reference back to the node the data > came from. This node towards a node that was achieved after less routing > then the current node. Besides not being ideal for routing, I also see > this as a security issue, since it makes it easier to retroactively track > data back the inserting node. >
Ok, gotcha. (I think) Requesting sets data references to nodes that are *more likely* to have keys close to the key being requested. Inserting sets data references to nodes that are *less likely* to have keys close to the key being inserted. Hmmm - > <snip> > > Yeah, me too. It was my intention to get to that after finishing 0.3 but > before going back to school, but school starts on Monday, so I don't see > that happening. But the whole "fail out of University" thing never seemed > to hinder me before, so... > Boss - "What are you doing?" Me - "uhhh, er, um, testing the network infrastructure. Yeah, that's it. " D Schutt _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
