Re: [freenet-devl] What replaces inform.php in 0.4 (if anything)?

[Ian Clarke]

> Anyway, if many many websites list a couple of node addresses, then it
> will be easy for people to find one -- and it seems fairly likely that
> once the process is clear, many many websites will do just that.
> This is *good* :)

Again, putting words into his mouth, but Oskar's hypothesis seems to be
that any node can be compromised given sufficient attention, and thus
any node which Dr Evil is aware of (yes, Mr Evil now has a PhD), ie. any
publicly advertised node, is inherently untrustworthy.  Do you agree
with this?

> > Actually it seems to work ok, although far from perfect - but back to
> > the point. Scott researched this specific topic. He discovered that the
> > announcement protocol will have a significant beneficial effect.  There
> > is no evidence to suggest that routing announcement messages via a small
> > number of seed nodes will have a damaging effect on the network.
> 
> Well, depends on whether those nodes are compromised of course.

Of course, but the fact remains that network topology concerns are
probably unfounded where the announcement protocol is concerned.

Ian.

PGP signature