Let’s not ask whether, in two years’ time, Amazon will deliver your last-minute Christmas shopping by octocopter. It won’t – the idea is a glorious publicity stunt.

The current objections to filling the skies with delivery drones are tremendous. The hovering robots are unreliable and likely to bump into things, and each other; the energy costs are high; and, of course, we look forward to Amazon fending off litigants after a miniature helicopter makes a hard landing somewhere inconvenient.

But while we tend to get excited and then disappointed by short-term techno-hype, the longer-term changes are often more profound than we expect. We typically lack the imagination to understand what they might be – nothing ever looks as dated as old science fiction. Regulatory obstacles are there to be surmounted: we already drive heavy chunks of metal around, killing people every day, so it is safe to assume that we will get over any safety concerns about drones.