The PC era is arguably over and the age of ubiquitous computing might finally be here. Its first incarnation has been mobility through smartphones and tablets. Many pundits, though, are looking to wearable devices and the so-called “internet of things” as the underlying trends of the coming decade.
It is tempting to talk about the internet of things as simply another wave of computing like the mainframe, mid-range and personal computer. However there are as many differences as there are similarities.
While the internet of things is arguably just another approach to solving business problems using computing devices, the units of measure are not so directly tied to processing power, memory or storage. The mass of interconnections mean that network latency, switching and infrastructure integration play just as much a part, for the first time divorcing a computing trend from “Moore’s Law” and the separate but related downward trend of storage costs.
Does it matter?
The internet of things brings the power of the information economy to every part of society, providing an opportunity to repeat many of the gains that business and government experienced through the last decade of the twentieth century. Back then there was a wave of new applications that allowed for the centralisation of administration, streamlining of customer service and outsourcing of non-core activities.
Despite all of the apparent gains, however, most infrastructure has remained untouched. We still drive on roads, rely on electricity and utilise hospitals that are based on technology that really hasn’t benefited from the same leap in productivity that has supercharged the back office.
The 40 year problem
It’s not hard to build the case to drive our infrastructure through the network. Whether it is mining, transport, health or energy, it is clear that there are mass benefits and efficiencies that are there for the taking. Even simple coordination at a device level will run our communities and companies better for everyone’s benefit. Add in the power of analytics and we can imagine a world that is dancing in synchronisation.
However, just as this new wave of computing doesn’t adhere to Moore’s Law, it also lacks the benefit of rapid consumer turnover. We’ve accepted the need to be constantly replacing our PCs and smartphones, driving in turn the growth of an industry. But it’s one thing to upgrade one or two devices per person every few years, it is quite another to replace the infrastructure that surrounds us.
Much of what we work with day-to-day in roads, rail, energy and all of the other infrastructure around us has a lifespan that is measured in decades or even half centuries. Even with a compelling business case, it is not likely to be enough to replace embedded equipment that still has decades of economic value ahead of it.
If we don’t want to wait until our grandchildren are old to benefit from the internet of things, we have to find a way to implement it in a staggered way.
Making a staggered implementation work
Anyone who has renovated their house will have considered whether to build in new technology. For most, network cabling is a basic, but centralised control of lighting and other devices is a harder ask. Almost anything that is embedded in the walls runs the risk of being obsolete even before the builder hands over the keys.
The solution is independent and interoperable devices. Rather than building-in security cameras, consumers are embracing small portable network cameras that they can access through cloud-based services. The same providers are increasingly offering low-cost online light switches that don’t require anything more than an electrician to replace the faceplate. This new generation of devices can be purchased as needed rather than all at once and raise little risk of uneconomic obsolescence.
Google’s purchase of NEST appears to be an investment in this incremental, cloud-based, future of interconnected home-based devices. By putting it in their stable of technologies, it is a fair bet that they see the value being in offering analytics to optimise and better maintain a whole range of services.
Just as companies focused on the consumer at home are discovering an approach that works, so too are those vendors that think about bigger infrastructure including energy, mining and major transport assets.
The car industry is going to go through a revolution over the next decade. After years of promise, new fuel technologies are bearing fruit. It’s going to be both an exciting and uncertain ride for all involved. Hybrid, electric pluggable, electric swappable and hydrogen all vie for roles in the future.
Just as important than as fuel technology is the ability to make cars autonomous. So-called “range anxiety”, the fear of running out of power in an electric car, is eliminated if the car’s computer makes the driving decisions including taking responsibility for ensuring an adequate charge or access to charging stations.
Arguably the most exciting developments are in the car software rather than hardware. Like smartphones, manufacturers are treating their vehicles as platforms for future apps. New cars are including technology that is beyond the capabilities of current software such as that needed to run an autonomous vehicle.
Early beneficiaries are the insurance industry (telematics), car servicing providers (who can more actively target customers) and a whole range of innovators who are able to invent solutions that as drivers we never even knew we needed.
Perhaps a tie-up between innovators like Tesla and Apple will show what connected cars and infrastructure could do!
Winning the marathon
Unlike previous waves of computing, the internet of things is going to take many years to reach its full potential. During this transition business opportunities that are unimaginable today are going to emerge. Many of these lie in the data flowing across the internet of things that is genuinely big (see It’s time for a new definition of big data).
These new solutions can be thought of as permutations of cloud services (see Cloud computing should be about new business models). Those wanting to try to imagine the world of possibilities should remember that the last twenty years have taught us that the consumer will lead the way and that incumbents are seldom the major innovators.
However, incumbent energy and infrastructure companies that do want to take the lead could do well by partnering with consumer technology companies. The solutions they come up with are likely to include completely different relationships between a range providers, services and customers.