[ppml] Proposed Policy: IPv4 Countdown

[michael.dillon at bt.com]

> Nice thought experiment, but 99.9999% of the routers in the 
> ARIN region (you 
> know, all those boxes consumers and SoHos are sitting behind) 
> don't support 
> IPv6 and the vendors have shown absolutely no interest in 
> adding it.  

If the boxes are based on a CPU or SOC such as ARM, MIPS, PowerPC, then
the software can usually be upgraded by flashing the box. It may be a
bit tricky for consumers to do, but it is NOT a forklift upgrade. These
days most such boxes are based on programmable hardware, and sometimes
even FPGAs.

> And, even if it were available, how do you plan on 
> reaching out to 
> millions of Joe Sixpacks and Grandmas and convince them to upgrade?

Cost/benefit comes into play here. How many of these have a significant
capital investment in their boxes? Very few. Most of the capital
investment was hidden in monthly rates that lose money for the first 12
months with an 18 month lock-in, or they are called a "setup fee". A lot
of these people will simply switch to the new IPv6 service with bundled
IPv6 gateway when there is something they want on the IPv6 network.
These millions are not the early adopters. They will tag along when the
early adopters entice them into it.

> An IPv6 core is irrelevant if none of the endpoints can reach 
> it.

Somehow I think MPLS cores are the more likely endgame.

>  Vista 
> finally has v6 on by default, and maybe 20% of users will 
> have upgraded by 
> the end of this year, but they still won't be able to reach 
> an IPv6 router 
> even two hops away at their ISP.

Once upon a time Vint Cerf and friends created something called IP even
though there were other perfectly good protocols around such as X.25.
Back then people scratched their heads and wondered why anyone would
connect to an IP network when there was no content on it. I used the
X.25 Telidon service in Canada back in the early 80s, and I remember
surfing to various sites all over the country with weather reports,
history of the telephone (with embedded photos), tourist info and
chatrooms. All of this was graphical using NAPLPS over an underlying
X.25 network. I remember when I first found the Internet in 1991 and how
clunky it felt in comparison. But in the end, IP won, NAPLPS is mostly
just history and X.25 is more likely to be found running on some
obsolete switch in a corporate basement that nobody knows how to switch
off.

At that time nobody could predict how and why people would migrate to
IP. There were plenty of people giving hundreds of reasons why IP and
the Internet was inferior. Many of those reasons were 100% right but
both IP and the Internet evolved, slowly and inexorable gathering mass
and momentum. IPv6 has been doing this for years now. It's not just a
quaint idea on the shelf.

--Michael Dillon