This is really the best part of the article: "In the meantime, the clock is ticking for the RIPE NCC, which handles Europe, the Middle East, and the former Sovi

This is really the best part of the article: "In the meantime, the clock is ticking for the RIPE NCC, which handles Europe, the Middle East, and the former Soviet Union. RIPE still has some 65 million addresses left—almost exactly the number it gave out last year. RIPE has a similar final /8 policy to APNIC, so Europe may be looking at a Christmas stocking with no IPv4 addresses in it by the end of 2011. Of course if ISPs rush to submit their last, big requests, it could happen sooner. LACNIC in Latin America and the Caribbean and AfriNIC in Africa dish out relatively few IPv4 addresses every year, so they are good for several more years. ARIN, which handles North America, does give out a lot more addresses, but somehow managed to get the long end of the stick: not only does it have 61 million regular addresses left, but it also "administers" 75 million unused legacy addresses (given out before the RIPE NCC, APNIC and ARIN were formed). So North America should be good, IPv4-wise, for a couple more years. Maybe even longer. " I haven't seen this type of analysis in any of the other IPv6 articles and is really the most important variable. Growth is important too, but I think it takes a back seat to availability and run rate. It's good to hear ARIN has some super secret addresses. If we keep going at our current rate we'll probably see some of the RIRs run out of addresses. However, I think between ISPs buying back addresses, B2B IP address sales, and other IP market dynamics will extend the life. If the Microsoft deal is a signal, IP addresses are currently around $11, if bought in bulk. If I could put stock in IP address price, I'd be long right now.