[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [freehaven-dev] Opening example

On Sun, 19 Nov 2000, Roger Dingledine wrote:

> Speaking of, we really haven't spec'ed out a detailed proposal for
> introducers for Free Haven yet. We criticize Publius for not having any
> real support for a dynamic system, whereas we don't have much of a system
> at all yet. :)

Well, that's partially because we don't have it up and running yet. 
We did discuss this at some point. As I recall, the introduction 
process went like

	0. Some entity E decides it wants to become a Free Haven server.
	1. E approaches a current node N. Somehow convinces N to 
	introduce it to the servnet. The "somehow" is out of scope for
	the protocol. 
	2. N broadcasts to all servnet nodes "E is a new servnet node
	wannabe." This broadcast by itself does not imply an endorsement
	positive or negative of E's capability.
	Now all servers know about E. 

	3. Some servers (brave souls) decide to trade low-value data
	to E. The results are reported in the form of trust broadcasts
	by these servers to the entire servnet. 
	Of course, if we actually were doing this, 
	how long would it be before someone "screws up" and has
	a servnet node software which trades high-value data to a newbie
	by mistake? :-)

> In order to describe that, tho, we need to use an example where files
> actually get moved to other computers. Such systems are not widespread
> in the public eye, currently. (True? Akamai doesn't count?)
> (Look at http://www.oreillynet.com/pub/q/p2p_category for a sample
> (incomplete) list.)

Akamai doesn't count because all the Akamai server farms are under the
control of a single entity, right? One of the neat things about peer to
peer is that it seems to involve large numbers of autonomous participants
in such a way that they really are autonomous...not merely passive or
atomised as in basic web serving.