I really like Bram, but what is revolutionary here?
I was pretty heavily involved in p2p field back in the early '00 and I've read extensively on the subject. Even back then there was plenty of research into peer-casting, including peer clustering by proximity metrics. The idea is bloody obvious, and it all inevitably boils down to constructing an efficient and resilient overlay networks, which is a very well researched domain. The reason there's not much of it implemented is because there is always a simpler (read - dumber) solutions that worked just as well in practice. Think YouTube vs. Joost.
Can anyone with a more recent exposure to p2p stuff comment on whether this is indeed an innovation or is it just a PR spin on a patent application?
I was pretty heavily involved in p2p field back in the early '00 and I've read extensively on the subject. Even back then there was plenty of research into peer-casting, including peer clustering by proximity metrics. The idea is bloody obvious, and it all inevitably boils down to constructing an efficient and resilient overlay networks, which is a very well researched domain. The reason there's not much of it implemented is because there is always a simpler (read - dumber) solutions that worked just as well in practice. Think YouTube vs. Joost.
Can anyone with a more recent exposure to p2p stuff comment on whether this is indeed an innovation or is it just a PR spin on a patent application?