> -----Original Message-----
> From: Toad [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, October 22, 2003 4:51 PM
> To: [EMAIL PROTECTED]
> Subject: [freenet-dev] Theory
> 
> A theory:
> NGRouting is not working because it can't predict classic routing unless
> the nodes using classic routing specialize quite obviously, which they
> don't.
> 
> How to falsify: 
> 1. Prove that classic routing nodes DO indeed specialize in a fairly
> obvious manner.
> 
> Or
> 2. A very large NGRouting-only test network might do it...
> 
> any other ideas?

When and how does a node decide to keep a piece of data over something else
because of specialization?

If a node applies specialization only under certain limited circumstances
then we may have a problem. I would expect specialization to have happened
already since most established nodes' datastores should already be full. 

Consider this simple thought of mine and compare it to how things work now
(this is based on very rough assumptions of freenet's working, mainly by
Ian's lecture video and some other info I gathered from posts in devl):

A node is a mediator for the transfer of a piece of data found.
That piece of data is passing through on it's way to the original
requestor(?)
Information we have:
- The node that handed us the data
- The node that we will pass the data to
- The node that claims to be the originator of the data / a node who
certainly has the data stored at this moment (am I assuming correctly?)
Since we have the node IDs, we know their specializations, correct?
We measure our specialization for the piece of data against their
specializations and:
If we happen to rate higher than all three, we deem this file very worthy of
keeping and claiming "origin" (i.e. we should "probably" store it and claim
ourselves as source).
If we happen to rate higher than our "neighbor" nodes (the one we got it
from and the one we are giving it to), we deem this file moderately worthy
of keeping, and maybe mark ourself as "origin" for it.
If we happen to rate lower than all other nodes, then we don't particularly
care.

All this is theoretical (my theory) and may not apply at all to the workings
of freenet (2 cents).

Doc
_______________________________________________
Devl mailing list
[EMAIL PROTECTED]
http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to