I read the original paper † and according to footnote 1 it's measuring iPad usage as network operations + network embodied energy + tablet embodied energy, igno

I read the original paper † and according to footnote 1 it's measuring iPad usage as network operations + network embodied energy + tablet embodied energy, ignoring data centers and end-use tablet charging. That's right, this calculation ignores the energy the tablet actually uses. "tablet embodied energy" is the energy from manufacturing the tablet, and "network embodied energy" is the energy to build the cellphone network. "network operations" is the energy to transmit data through the network which is being counted at 2kWh/GB. Since the average electricity cost in the US is 12.4 cents / kWh, that works out to data transmission costing 25 cents of electricity per GB, which seems crazy high. Needless to say, this paper is very controversial, see ††. TL;DR: The iPhone electricity usage in the paper is ignoring the actual electricity usage of the iPhone and counting lots of other hand-wavey things. † http://www.tech-pundit.com/wp-content/uploads/2013/07/Cloud_Begins_With_Coal.pdf?c761ac http://www.tech-pundit.com/wp-content/uploads/2013/07/Cloud_... †† http://www.huffingtonpost.com/2013/08/20/iphone-energy-refrigerator-controversial-study_n_3782211.html http://www.huffingtonpost.com/2013/08/20/iphone-energy-refri... which is more informative than you'd expect from Huffington Post. Edit: The paper is titled "The Cloud Begins With Coal" and is sponsored by the National Mining Association and the American Coalition for Clean Coal Electricity.