The model we reviewed (which I am happily writing to you from, using approximately 2 per cent of its processing power) is ...