[arin-ppml] Why are ISPs allowed?

[Ted Mittelstaedt]

> -----Original Message-----
> From: arin-ppml-bounces at arin.net 
> [mailto:arin-ppml-bounces at arin.net] On Behalf Of Joe Pruett
> Sent: Monday, January 26, 2009 7:51 PM
> To: ARIN-PPML at arin.net
> Subject: Re: [arin-ppml] Why are ISPs allowed?
> 

> 
> i'll fess up to being that partially-clued isp.  i didn't 
> think that multilayer nat is the right thing to do, but 
> unless someone builds a good v6<->v4 nat device, then 
> multilevel nat is probably what will happen.  it already 
> happens nowadays and people survive (think wireless gateway 
> behind dsl gateway).  my research into the v6<->v4 nat state 
> of affairs leads me to believe that multilayer v4 nat is much 
> better understood at this point.
> 
> 

  Every time I have got involved in setting up a multilayer
NAT at a customer site (mainly because the customer has
bought themselves a wireless gateway router, against our
recommendations, instead of an access point and is behind
a DSL modem (ie: M1000) that
doesen't work properly in bridged mode) I have observed a
significant drop in throughput.  I came to the conclusion a
long time ago that multilayer NAT was a dead-end and I'm
surprised that anyone is still giving it any credibility.

  I'm glad you don't think that multilayer NAT is the right
thing to do, but I disagree with your assumption that this
is what will happen anyway.  People demand reliability on
their circuits - it's bad enough trying to explain to them
that in a storm, their DSL line is lower-priority than a
voiceline down the street.  I've heard enough "I'm losing 
thousands of dollars of business every hours the Internet
connection is down" excuses to last me a lifetime.  The
idea they would put up with the unreliability of a multilayer
NAT is IMHO false.

Ted