Queueing/capacity theory is a bitch. To my knowledge the best you can get is a square law – which is an ugly obstacle in business. The networks could add massive new capacity that would be redundant for most of the time, just to handle the occasional peaks that happen once or twice a year for major events or emergencies. It wouldn’t be economical.

This overlaps with three other debates, network neutrality, network integration (comon carriers) and private/free market vs state/national infrastructure. Because collectively the networks have enough capacity to handle huge peak traffic bursts, but one assumes they do not share out their resources properly. The story deals with the failure of one network, O2, but if the SMS networks were properly interoperable with rapid dynamic bandwidth purchasing ability in place for the networks to trade their pipes and resources the problem would vanish.

In other words, it’s not necessarily the problem that there aren’t enough resources out there to go round, but rather that the individual operators won’t play nice with each other and cut off all their noses through mutual spite.