[ppml] IPv4 wind-down

[Tony Hain]

Leo Bicknell wrote:
> > ...
> > further extending the lifetime of IPv4 is a waste of everyone's time
> > and energy. People will need to learn the realities of the new
> version of IP, or find another line of work.
> 
> I strongly disagree with that statement.  No matter your thoughts on
> IPv4 and IPv6 one of ARIN's tasks is to be a steward of IPv4 space /for
> as long as people wish to use it./
> 
> There is a long history of new technology eroding existing markets, but
> rarely does it make them go away.  More importantly, technology that is
> not useful in one market may be huge in another.
> 
> Venezula uses lots of Ethanol, the US does not.  This despite a gas
> crisis in the 70's, and numerous alternative technologies.  Digital
> downloads have not stopped people from selling CD's.  VoIP has not
> forced AT&T and wireline phone providers out of business.  There are
> companies out there still making steam engines and buggy whips, if only
> for much smaller market segments today.

Yes, there will be a market for a few IPv4 experts for some time to come,
just as centuries after the height of its demand there is still a market for
hand-made masquerade masks in Venice. My point was that the market will not
sustain the number of IPv4 experts that it does today, just as there are
only 2 mask makers left in Venice. 

Assuming the IPv4 space is fully allocated/assigned by 2010, by 2015 the
number of people that can make a living by only knowing IPv4 will be
substantially lower than where we are today. By giving preference to IPv6
when it works, by 2015 the amount of IPv4 traffic will be effectively zero
from the perspective of today's giddy heights. Yes, IPv4 will persist for
decades within organizations (HP printers with embedded IPv4-only
controllers will not die), but those islands will be parked in cookie-cutter
corners of the network where no expertise is required. 

Tony