Re: Gnutella 3

Michael Green wrote:
> Now, before we get ahead of ourselves, what is the underlying reason to 
> actually need a "Gnutella 3". Gnutella 0.6 has matured quite well and 
> has withstood a lot of the issues where other P2P networks have failed. 
> So I fail to see why there's a need for it to be superseded at this point.

The problem is complexity: it's too hard to write a Gnutella
implementation, too hard to understand the subtleties of how an existing
implementation works, and too hard to build new services on the Gnutella
network. Almost nobody plays around with Gnutella and builds cool stuff
any more. Occasionally someone tries, and they find out how hard it is,
and they build a Facebook app instead. (Credence and DeHinter are rare
exceptions - two in five years.)

So from my point of view, the question isn't what features we should add
to Gnutella, or what P2P network we should "rebrand" as Gnutella 3; the
question is whether we can imagine a very simple, very flexible P2P
architecture that incorporates the lessons from the first ten years of
Gnutella and provides a platform on which to build cool P2P services for
the next ten years.

Cheers,
Michael