On Sat, Mar 30, 2002 at 01:20:18PM +0100, Eugene Leitl wrote:
> > To resist 2. you have to be able to randomly probe IP addresses to find
> > a node.
>
> Yes, probabilistic headless node discovery vs. a centralist approach.
I never really found discovering a currently active node on the
network a problem even with original gnutella client.
google seemed to reliably give you a host cache file with at least
some remaining online nodes.
As an aside, it seems that google (and other search engines) are vying
for first place as replacement for URLs as first line content
location. I looked recently at the web logs on my web pages over the
last year or so and I think > 50% of queries came from search engines
(from referrer field). The search terms (also in the referrer field
with google) mostly looked purposeful ("hashcash adam", "perl-rsa"
etc, though there were a few ill-thought out or name-clash search
terms "making hash" who probably didn't find what they were looking
for ;-)
This seems like a success factor for a would be storage-surface with
ambitions to replace and subsume the web -- any such system has to
have a really good google competitive search mechanism, or have a way
for it's content to be indexed by google.
Adam
