The IP Over NBMA (ION) Archive

["Eric W. Gray" <egret@nh.ultranet.com>]

Masataka,

You wrote (in part):
> 
> Yakov;
> 
> I can't see your points.
> ...
> > And if you fully align the topologies, there will be quite
> > a few other things that wouldn't be needed as well...
> 
> Yes.
> 
> So?
> 
> So what?
> 
> What's wrong with it?
> 
> While it's not the current reality, what's wrong to make it
> the reality of the future?
> 

Reality has a way of adjusting those who think they can adjust it.

Part of reality (needing adjustment most) is the existing network
infrastructure - including the Internet and all existing intra-nets.
Assuming that we all worked very diligently toward adapting the
topology of this infrastructure-reality to align with the "new guy
in town" ATM switching topology, other influences (such as changing
needs of users) would pull the logical topology (which is at least
cheaper to change) away from the real one (perhaps in new ways).  If
- as seems likely to me - we can never quite get to the point where
the real matches the ideal, than we might want to expend some energy
on making things work better when real and ideal don't quite match. 

That's so what.

>
> Or, do you want to continue to develop all the complex but
> unnecessary protocols forever, so that we won't lose our job?
>

Job security will never be an issue for people who want to solve problems.

> ...
> 
>                                                 Masataka Ohta

--
Eric Gray