Skip to content


Lessons from the history of electricity generation

The time has come to take a lead in creating open standards for digital infrastructure suitable for homes in cities of the future.

It is the modern fashion to set up a “market” for all utilities delivered to customers. Bill payers are encouraged to use U-Switch and similar resources to seek out the best deals; the theory is that this will lead to steady downward pressure on prices. Flipping domestic contracts for utility supply from one company to another is almost trivial. The hardware underpinning that switch is taken for granted.

What is often forgotten is that all the traditional utilities (electricity, gas, telephony, water and sewerage) went through a phase in which public ownership was a major factor, if not the monopoly supplier. This was usually the period during which national standards were created for the infrastructure components, and when that standard infrastructure was rolled out universally. The standardized hardware was put in place long before competition was seriously contemplated.

In the case of electricity, for example, in 1925, the British government asked Lord Weir, a Glaswegian industrialist, to solve the problem of Britain’s inefficient and fragmented electricity supply industry. Weir consulted an expert called Charles Merz, and the result was the Electricity (Supply) Act 1926, which recommended that a “national gridiron” supply system be created. The 1926 Act created the Central Electricity Board, which set up the UK’s first synchronised, nationwide AC grid.

With broadband, we have not had such a period of nationalized ownership, nor are we likely to in the future. Whilst broadband was just a way of connecting family PCs to the World Wide Web, this was not a real issue. So long as home computers could plug in to the ISPs’ modem, all was fine. Now that we are starting to embed digital systems into the very fabric of buildings, the issue of standards becomes much more important. Fitting a new central heating management system is a bigger job than throwing away an old router and installing a ne one.

In order to create a homogenized digital infrastructure – and thus a “level playing field” for competing suppliers – we are going to need a slightly different approach from the course taken for other utilities. Not only has there not been the period of nationalised ownership, but also broadband is a much more complicated utility than anything else which is delivered to residential customers. To specify a connection to electricity, gas, or water or sewerage takes only a few parameters, and to test an implementation is also easy. By contrast, the full collection of standards applicable to data infrastructure already runs to tens of thousands of pages for the connection alone. And there are still areas which need more standards work. Even a simple move from one DSL supplier to another inevitably involves receiving and installing a new router; if you want to contemplate a switch to cable or WiMAX you’re into installation engineers crawling over your house. And that’s just for the connection to the outside world.

The word “Broadband” is of course a convenient shorthand for the entire data infrastructure and its ecosystem. At the moment there is a clear case of market failure in this ecosystem. The variation in the configurations and the options available make development a nightmare, marketing confused, and customers underwhelmed. As the “internet of things” grows, the situation is in danger of getting so completely muddled that market growth will stall completely.

What is most needed is an approach to residential data infrastructures which simplifies the problem to the point that it becomes tractable. I propose that we create an open, standard “core” for the domestic infrastructure, consisting of internal and external connectivity, a processing resource to run apps, and a standard display device that can be shared by all apps (yes, it’s an iPhone-equivalent for your home – except a totally open standard)

Electricity is delivered to the home through a “Consumer Unit” (which used to be called a fuse-box, until fuses were superseded by circuit breakers). I have called the equivalent device for delivery of digital services the “Digital Consumer Unit”, or DCU for short.

The DCU provides exactly the needed simplification of the digital infrastructure ecosystem, and creates a resource into which all the other parts of the ecosystem can plug, in a standard way. Creating that standard will kick-start the home control industry, expand the market for compatible add-ons, bring down prices and allow a panoply of smart apps to be created. Energy saving is of course the most important agenda but other services will also be part of the system: Telecare, security, smart lighting, and bus timetables to name but a few.

The objectives

The top level objectives are threefold:

  • Create a connection device that can be transferred between broadband suppliers, to create an open market for Broadband ISP services.
    A customer should be able to choose between services in just the same way as they choose other utility offerings. Services could include broadband to the PC, but could also include the management of the core home device itself.
  • Create an environment where standard off-the-shelf devices can be added to the home infrastructure without requiring specialized skills.
    The standard for the 240V/13A socket allows consumers to buy white goods from any supplier, knowing they will work when plugged in; the same should apply to data devices.
  • Provide a standard processing resource to allow software services to be added to the home at minimal cost.
    The capability of modern browsers has allowed many advanced functions to be delivered through the web at low cost, and independent of the platform on which the browser is run. A similar environment for home control add-ins would allow a whole new industry to emerge.

Britain is in a great position to lead this revolution. Let’s not throw the chance away.

Posted in Future Cities, Market needs, Politics.