Is the main thrust and incredulity in your argument because you think there CANNOT be a better solution than burning 100,000 CPU at 100% 24/7 and sending 100,000+ redundant messages per transaction?
No, not at all. Like I said when I jumped into this thread, I think using a DHT network to somehow distribute the work is a very interesting idea, it just seems to me any solution that partitions the network will be more vulnerable to attacks that insert malicious nodes. I think it would be fantastic to come up with less resource-intensive solution that actually works.How does Freenet generate node IDs? The only info I see in the latest Freenet paper is:
Quote
When joining, nodes choose an identity at random, and then connect to those peers with whom they have a pre-established trusted relationship.
.... and:
Quote
The perhaps largest advantage that the trusted-connection model offers over traditional anonymity networks is protection against the Sybil attack [13], where a single attacker infiltrates the network by pretending to have many identities. Since the network depends on out of band trust relationships, an attacker gains strength only through the number of people he can fool into trusting him, and gains nothing by presenting multiple identities online.
...which doesn't sound like it would work very well for Bitcoin.