The idea of an Internet of Things linking tens of billions of devices (50bn in one exuberant estimate) is now generating huge excitement.

The IoT combines multiple technologies – sensors, the internet itself, open data, cheap wifi, Bluetooth, the cloud, predictive tools and others - and points to a future where the web is once again only a small part of the Internet.

A few weeks ago Nesta hosted a gathering on the IoT, with the Chief Scientist and a bunch of very smart innovators and thinkers. Some of the issues discussed have been covered by my colleague Jess Bland in her recent blogs.

Some of the discussion reminded me that the IoT has been a long time coming. I did my PhD in telecoms 20 years ago and most of the applications discussed now were being talked about then.

Indeed, I recently came across an article I wrote in 1988 (before the Internet, if such a time is possible to imagine) that talked about smart houses controlled by voice commands, peppered with smart meters, and connected with broadband, and the many issues of privacy and control that would come with it.

The benefits of a truly advanced IoT would be potentially dramatic, and worth the wait: a jump in capacities to coordinate physical flows, such as energy, water, and traffic, allowing more efficiency, less waste, smaller buffers and downtime.

But most of these benefits are largely invisible. They’re more like advances in plumbing and electricity than smart phones and iPads that you hold in your hand.

The age-old cliché of the smart refrigerator which automatically orders more milk when you’re running low proves the point. The IoT infrastructure is impressive, but the consumer benefits are few. We can all see some benefit in remote control of security, heating, lights, fingerprint activated door locks or cool energy meters from companies like Nest Labs.

But these are hardly earth-shattering. And to the extent that there is a social imagination in play it’s neither very inspiring nor always very attractive.

If there is a social imaginary in the IoT it’s a very old one: the panopticon.

This was the idea developed by Jeremy Bentham, and implemented in Pentonville Prison: a design whereby the observer could observe many people simultaneously. Michel Foucault built part of his career out of thinking through the many meanings of the panopticon as a metaphor for comprehensive surveillance and control.

Today it remains a dominant theme in the IoT – symbolised in great centralised control rooms monitoring in real time everything from energy flows to pedestrians on streets.

But it’s not the only possible imaginary for the IoT. One alternative is to see the IoT as the enabler of nature, with trees fitted with sensors that can tell if they’re being chopped down, rivers that can sense and communicate when pollution levels go up or, for that matter, air quality sensors of the kind proliferating in China right now.

Another imaginary is the IoT as bottom up social sensing – symbolised by projects like Smart Citizen or Open Corporates - that help us, the people, monitor the actions of the powerful rather than vice-versa.

Then there is the vision of the IoT as hackable – the implicit ideal of Arduinos and the open hardware movement, in which DIY meets the digital age and we all learn how to hack our homes and cars, programming them to meet our needs rather than fitting into the neat corporate designs of a Nest or Apple, an ethos that’s also strong in the shared economy.

Then again there are also hybrids like the various tools for blind people to navigate the city, or Boston Streetbump with citizen generated data helping the city manage physical infrastructure – a potentially cheaper alternative to the state/corporate panopticon.

It’s the 21st century equivalent of the sewer and the water system, which enabled massive social progress indirectly but were appropriately invisible

These alternative imaginaries matter because so far the IoT, and its relation the smart city, has not done well in exciting much public enthusiasm. The showcases – like new Songdo – are now routinely disparaged for their lack of appeal. Consumer products – like smart toothbrushes – look like parodies of contemporary capitalism’s tendency to solve trivial problems and needs rather than important ones.

Some of the advantages of the IoT do depend on a great deal of centralisation and harmonisation, and some of the features of the panopticon.

The driverless car – which I see as a potentially huge step forward – requires all of these, and shifts in culture and behaviour we can only begin to imagine. So would a health service in which every body part had a URL. Or a truly smart energy infrastructure, constantly adjusting supply and distribution.

And in some respects the IoT is bound not to engage the public. It’s the 21st century equivalent of the sewer and the water system, which enabled massive social progress indirectly but were appropriately invisible. But a bit more imagination would not go amiss.

The panopticon isn’t a neutral idea: it crystallises relationships of power.  And the mark of any healthy society is the ability to challenge, to question and deconstruct offers of the future, the better to put the pieces back together in a way that really does meet our needs.

Photo by Eris Stassi on Flickr.