Noel Chiappa wrote:
I mean, it's now coming up on *11 years* since IPv6 was selected (i.e. back when the latest and greatest uSoft-ware was *Windows 3.1*, and the WWW had a grand total of about 3K sites), and still we hear the "oh, it will take off soon" - the same line we've been hearing for close to a decade.
That is to be expected.
As you know, the value of a network is roughly proportional to the square of the participants. Since adoption is influenced by the value of the network it can take quite a long time to acquire "take off", whatever that is. Is it just an arbitrary point along the geometric growth curve or is it some milestone like native access being widely available to the average home?
It took a very long time to get native IPv4 connectivity to the
average home. Even as late as 1993, long after IPv4 had taken off
in academic circles it was difficult (expensive) to get native connectivity, even dialup. At that time dialup shell/terminal
server access was more common but severely limited as it is
non-native. Native IPv4 wasn't widely available to the average home
user until about 1995.
So given the decades it took for IPv4 to be commonly available why do we expect an accelerated deployment for IPv6? Leveraging the existing IPv4 infrastructure does help deployment but that alone does not affect the value or desire to deploy. Just as government subsidies helped deployment of IPv4 but did not directly affect the value or desire to deploy.
I hear the exact same excuses for not adopting IPv6 as I heard for IPv4 in the early 1990's:
"I have no reason to connect" and/or "I can't justify the cost"
The solution to both problems is n^2.
If you really want to jump start n^2, find a way to convince the RIR's to require demonstration of IPv6 deployment for subsequent (non initial) IPv4 allocations.
Kevin Loch
_______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf