Re: Identifier/locator recap

[Iljitsch van Beijnum <iljitsch@muada.com>]

On Mon, 17 Mar 2003, Pekka Nikander wrote:

> > No, HIP is smarter than that. But what I find troublesome with that
> > approach is that the identifiers are a flat 120+ bit space which makes
> > it incredibly hard to create a distributed way to look up properties for
> > identifiers.

> Not really.  Distributes hash tables (DHT) should work.  Somebody
> just need to work out the details; they are not exactly simple, though.
> But not that complicated either, I would believe.  See the proceedings
> of recent peer-to-peer conferences and you will find multiple occasions
> where DHTs have been used.

Hm, I'm not very familiar with this technique and I can't find a good
explanation of how it applies to the problem at hand right now. But it
seems to me that such a distributed system is very vulnerable to
malicious nodes holding back information that they are supposed to
share. Also, the self-organizing could make unsuspecting nodes attract
more traffic than they can handle.