Re: IP: Economist: Upgrading the Internet

[David Farber <dave () farber net>]

> To: farber () cis upenn edu
> Subject: Re: IP: Economist: Upgrading the Internet
> Date: Mon, 02 Apr 2001 07:48:25 -0400
> From: "Mike O'Dell" <mo () ccr org>
>
>
> gee, another nice IPv6 rah-rah piece
>
> the item such articles never mention is that IPv6 does
> *nothing* to address the fundamental problems of IPv4,
> but rather it makes them *worse*.
>
> true, IPv6 has larger addresses, but having lots more of
> a hard problem doesn't make it any easier.
>
> things that could sink the current Internet are not running
> out of addresses or lack of payments in packets, or even the
> non-existance of the mythical QoS beast, but a really fundamental
> matter: scaling.  scalability of the routing infrastructure,
> the ability to divorce abstraction boundaries from hardware
> boundaries, and the economic scalability of various product
> technologies, these will make or break future growth of the
> Internet.
>
> disclosure: i spent a lot of time working on IPv6 and was
> originally a booster, but came to the slow realization that
> we will not address these with a change to the IPv6 packet format.
> in fact, they have almost nothing to do with packet formats.
>
> another qualm I have with IPv6 is that i find it unlikely
> that deployment will be widespread before it starts running
> out of gas.  remember - the *usable* part of an IPv6 address
> is only 8 bytes long.
>
> i'll stop now, but i'm concerned that "happy talk" about
> the curative powers of IPv6 is an opiate.
>
>         -mo



For archives see: http://www.interesting-people.org/