On Jul 10, 2015, at 2:01 AM, Nicholas Suan <nsuan@nonexiste.net<mailto:nsuan@nonexiste.net>> wrote: You should elaborate on some of these 'holes' then. Indeed. If there are “holes” in the methodology, then they are quite consistent holes, since Google, APNIC, and Akamai <http://www.stateoftheinternet.com/trends-visualizations-ipv6-adoption-ipv4-exhaustion-global-heat-map-network-country-growth-data.html#countries> are all reporting similar rates in US IPv6 end-user connection ratios (around 20% of connection attempts at present and growing rapidly) It’s possible that there’s some self-selection going on (i.e. that the subset of Internet users reflected in this percentage are disproportionate users of Google, Facebook, and Akamai-served content sites compared to the “other” Internet end-users), but given the widespread usage of the companies providing the information, there’s a fairly high burden of proof necessary if one is to assert that the metrics are somehow not representative of the US Internet end-user population as a whole. /John John Curran President and CEO ARIN