On Sat, 19 Feb 2000, you wrote:
> Brandon wrote:
> > You don't need to have a permanent internet connection to run a node. You
> > can run the node, run the client, and then shut down the node. Or, better
> > yet, have your script for connecting to the Internet automatically run the
> > node so that it's running whenever you're connected.
> is there a way that when my node goes offline the other nodes will not
> send requests to it, or will they then when they get no answer, assume
> it is down? basically what i'm asking is - is the connection status of a
> node irrelevant to the operation of the freenet?
As it is now nodes will just continue requesting and fail, after which they go
on to the next node. This should not be a problem as such, but we will still
implement a way for nodes to ask not to be sent requests, since people with slow
connection may not want to serve connections (and Freenet works better if they
> also thanks for the info, i'm downloading the sun JDK ATM so i can run
> my own node (and thus client).
> i have another question about freenet - it seems that the only way to
> find something is to specifically ask for it (due the search method
> described on the site), is that say that there is no "index" of
> files/data on the freenet in case someone just wants to browse? if there
> is how is it handled because i can think of ways that would harm the
> freedom of the freenet (ie a node keeping an index of what IT has, which
> makes the whole system counter itself)
You are absolutely right that there is currently no way to locate data on
freenet without knowing it's exact key. To what extent this will cause a problem
is hard to know, but we are hoping that, at least for now, using strict naming
conventions and links (in many ways hypertext is the ultimate decentralized
namespace) will be enough. I have yet to see an idea for implementing searches
that I think will work.
> Jay McCarthy (jay@...)
> "Question authority!" "Says who?"
> Freenet-chat mailing list