As long as about 2/3 of the IETF attendees are from North
America, 2/3 of the meetings should be in North America.
similar logic might apply to havana.
or, as long as 2/3 of the meetings are held in north america,
2/3 of the attendees will be from north america.
You're being silly. That does not stand a historical test: when 100%
of the meetings were in NA, it was not the case that 100% of the
One could argue that all the meetings should be located at the
feasible site nearest the centroid of the homes of the participants,
but I suspect that centroid is not a well-defined concept on a sphere.