Danny, On Wed, Nov 12, 2008 at 01:15:07PM -0700, Danny McPherson wrote: > > On Nov 11, 2008, at 11:57 AM, David Kessens wrote: >> >> It seems that arbornetworks estimates are extremely low to the point >> where one has to ask whether there were other issues that caused such >> a low estimate. > > No, the methodology is outlined in the referenced report. > Given what we were measures and what data was supplied, those > were the results. The report as presented at the RIPE meeting indeed mentions the possibility of undercounting. However, it appears that there is an undercount of several orders of magnitude. At that point you really cannot claim that the report provides a perspective on Internet IPv6 traffic as it does. It is quite reasonable to conclude that something went wrong with the methodology, measurements or analysis. >> There is no question that IPv6 traffic is quite low in the Internet. >> However, many other reports that I have seen recently measure multiple >> orders of magnitude more IPv6 traffic (for an easily accesible example >> see: http://www.ams-ix.net/technical/stats/sflow/) > > Indeed, and multiple orders (less than two) of magnitude is still, > from this, only .1% on average. I believe the point remains very > much the same. The difference between something that is barely measurable and something small but measurable like 0.1% is huge. Basically, 0.1% on the scale of the Internet means that a very large group of people is using IPv6 today. There is no question that that group pales to the total number of Internet users but it sure is more than a few people in IETF experimenting with IPv6. David Kessens --- _______________________________________________ Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf