[rrg] making a difference

[Lixia Zhang]

On Nov 30, 2008, at 10:57 PM, Michael Meisel wrote:

> Hi all, I've just been catching up, and I've noticed the  
> reappearance of
> a common theme that I think warrants discussion directly -- what
> difference we should make, how to do it, and if it's even possible.  
> Here
> are my thoughts on the issue, many pieces of which have been said by
> others before me (Tony, Lixia, Christian, Noel, Dave Conrad, and  
> Brian,
> to name a few).

thanks for the credit! :)
though it seems to be quite a distance between what's said this msg  
and my thought...

Let me just first focus on your 3 principles below

> There are many tiny changes happening constantly in the Internet on a
> small scale. These changes have the potential to add up to a radically
> different picture sometime down the road, as we've seen happen with  
> NAT.
> So, to have any influence on the network, we should be thinking about
> how we can tweak the small changes to push things in a good direction.
>
> I believe this means a focus on three principles:
>
> (1) The network will always be heterogeneous

agree in principle.
but some clarification here would help more: exactly along what/which  
dimensions of heterogeneity are you referring to here?
given this msg is about making changes, I interpreted "heterogeneous"  
here as measured by different interest, different reactions to any  
proposed changes.


> (2) Any large change will only happen as the aggregate result of many,
> many tiny changes at individual networks and hosts

I am not clear what is the definition of tiny and large here: what are  
you measuring?


> (3) It's not up to us (or any central authority) which tiny changes  
> get made

if you are saying the Internet has no boss, then I agree.
But I do believe that it is our job to understand what is driving the  
trend, so that we figure out how best to influence or facilitate the  
changes.

and just one more thing:

> (1) and (2) imply that we should push forward with some
> host-based scheme that separates the host's identifier from its  
> routable
> address.

not exactly sure what you meant by "host's identifier" here (e.g. LISP  
calls hosts' PI addresses as identifiers, and separate that out of  
global routing--is that what you meant here?)

> This is orthogonal to ISP-based solutions, and it is bound to
> be just what *someone* needs, and could make some difference in the  
> size
> of the routing table.

TO make a difference in routing table size: it may only take the  
action of a single or small number of parties to increase the routing  
table size; but to remove a prefix from the existing routing table  
requires every party, not just some hosts, knowing how to reach that  
prefix thru some new way.

Lixia