A lot of ink has been spilled regarding the impending tsunami of Internet Of Things (IOTs). It is certainly an interesting topic, and not just for geeks. As the next generation computing platform, IOT will see an explosion of connected device (wearables, intelligent refrigerators, smart thermostats, etc.), currently estimated to be at least 10X the ecosystem of smartphones (this tech generation’s ecosystem). With prices of a basic sensor-equipped computer already down to sub $5 level, it will reach the sub $1 level in couple of years than then – boom, ubiquity.
However, lets first start by clearing some of the basic misconceptions regarding IOT, before we can get to answering the core question. IOT is not really a bunch of ‘things’ – its a software-based ecosystem (think ‘stack’) – that enables people to ‘plug into’ a universe of connected active devices, that, for most part, are communicating with each other, while occasionally providing data/inputs to humans and taking instructions/decisions.
The key to understanding when IOT will become real, is therefore, an exercise in predicting when the software stack will be ready. And that begs the obvious question – how do we identify the components of this ‘stack’?
The ‘Stack’
Lets take a simplified view – just 5 ‘layers’. Start with UI, then Applications, then Messaging/Communication, then Data and finally Infrastructure. And then lets look into the trends/trajectories in each of these 5 layers, towards that ‘software-based ecosystem of things’. Once we see where we are going with these components, we will better understand the progress of this entire ecosystem.
The Consumer UI is obviously the most exciting one, and perhaps the most critical one. Each computing revolution fundamentally changes the ‘UI’ – think mainframe to Mini to PC and smartphones (people old enough will remember leaving shoes outside an air-conditioned room to work on mainframes). IOT will need a leap in the interface capability and mechanism over the smartphone based UI design, and leading firms are currently experimenting with multiple approaches – augmented reality, VR, voice, hand gestures, etc. While it is difficult to accurately predict which one of these will win in next 5 years, or if something entirely new will emerge from a start-up, what is clear that the computing requirements, coupled with size requirements of the user interface, makes it a difficult challenge. This layer will probably be the last shoe to drop and complete the picture.
The Consumer (and Enterprise) Application layer is probably the easiest, as the current technology is enough – but it will be heavily influenced by the business use cases, which in turn, will flow from the user interface capability. Think Uber and Tinder to understand how application and business use case are driven by the UI capabilities of the system. And it is this layer that will then drive the future product cycle evolution, where most start-ups will be born in the next decade.
Messaging and Communication layer is also ‘there already’, for the most part. Of course, it is most advanced in retail commerce, and lot needs to be done for other ‘verticals’ like healthcare, automotive, etc – but as the business use cases develop in the above layer, this layer will keep pace in terms of new standards. Payments is an important part of this layer, which while very well developed, will still need new/innovative messaging, processing and business models to fit into the new use cases being developed – just like Payments had to evolve with the previous PC based model (think eCommerce). However, this is not a stumbling block – the basic capabilities exists.
Data Layer and Infrastructure (think cloud/IaaS) is, as always, most critical layer for this entire ecosystem to work smoothly, and deliver the business user cases that will drive the economics. Here, AI, Deep Learning, Machine learning and Big Data have made tremendous progress over the past 5 years, so it is ready for prime time now.
What then, next?
So, we come back to the last shoe to drop – the UI layer. With everything already ready from the ecosystem perspective (i.e. the remaining stack), we are probably looking at a lot of innovations over the next 3-5 years. Rapidly declining micro/embeddable computer prices, further improvements in the Data and Cloud Layer, and better stack integration will tip over the system to the next-gen very fast.
We are, in some sense, where mobile computing was in 2005 – lot of interesting action, lot of point innovations, lot of standards development, 3G/4G roll out for infrastructure, an impending sense of ‘something is about to happen’ and then boom – iPhone solve the Consumer UI challenge.
5 Expectations for the Memory Markets in 2025