You claim that a maximum of 100 nodes can be searched
on any query.

So if a new node inserts a new data, and another new
node requests that data, which has never been
requested before, there is only %0.1 chance that the
request will succeed on a 1,000,000 node freenet. That
is "unacceptable".

Let's make something clear: the routing technique used
in freenet (for 0.3; 0.4 is not documented) is only a
heuristic fucntion that tries to "guess" where to look
for data.

This will work good ONLY IF the data is moved to a
location it can easily be found. Otherwise, freenet
routing will only work for "fameous data", for regular
data it is a blind BFS which is pruned TOO EARLY.

Can Ian comment on this? I hope he knows much better.
I also hope there is already an "advertising"
algorithm in freenet. Otherwise what I said is true.

> Yes. Freenet doesn't query every single node, just
> the ones that are most likely to have the info. 
> Furthermore, the maximum number of nodes your
> request is allowed to pass through is limited by the
> Hops To Live (HTL) value. Nodes will default to
> setting the HTL value to 100 if it sees an HTL
> higher then that.


=====
Sukru Tikves,
  Hacettepe University
  Computer Engineering 
  Department, Ankara

  http://piskare.cjb.net

__________________________________________________
Do You Yahoo!?
Listen to your Yahoo! Mail messages from any phone.
http://phone.yahoo.com

_______________________________________________
freenet-tech mailing list
[EMAIL PROTECTED]
http://lists.freenetproject.org/mailman/listinfo/tech

Reply via email to