次の方法で共有


Internet of Things: Gadgeteers, Gatherers, Messengers, and Controllers

2 trillion (Gartner), 4 trillion (ARC), 7 trillion (IDC, PwC), 14 trillion (Cisco), 19 trillion (SoftTech). U.S. Dollars. Cumulative value-add by “IoT”. by 2020. or 2022. or 2025. or over the next 20 years. Shiny CFO and potential investor eyes. Because dough.

There is quite the gap between 2 trillion and 14 trillion U.S. dollars. There’s also quite the gap between 2020 and 2034. The figures that are being thrown around raise a grand red flag that has the word “overscoped” writing in shiny letters across it. Because “Internet of Things” is a conveniently empty conceptual shell that everyone can project their own notions into. Not only is it “big”, it’s also whatever anyone wants it to be. And with large volume consumer devices becoming a little boring now that all phones and tablets effectively look the same and innovation has slowed, and the apps market gold rush having tapered off, everyone has started looking for where the fresh, green pastures are. Trillions seems mighty green. So the hype machine is doing its job trying to convince the wallets to dish out investment dollars by promising trillions in return.

And so, happy heartfelt congratulations, ladies and gentlemen, “Internet of Things” just made summit on the Gartner hype-cycle curve. The outline of Gartner’s depth hype-cycle analysis zooming in on IoT shows what a hodgepodge of stuff falls under that umbrella. Smart Dust and Quantified Self to RFID for Logistics and Transportation

The Gartner outline by itself is obviously informative on what they believe is worth a mention in the context; the list is dominated by horizontal platform capabilities and commercial use-cases. That is, interestingly, in stark contrast to where the tech press and blogosphere puts their attention. Since they’re slaves to the click, the latest network-aware gadget with the ability to post your achievements on Facebook is obviously promising more eyes than advances in data transparency on the mid-voltage portion of municipal energy grids.

Generally, “Silicon Valley” from where the brightest light beam beacons emerge in hope of attract grand swarms of capital (and young talent willing to take an all-or-nothing risk while they can) is fueling its built-in hype machine with consumer-focused products and services. Successful consumer-products and services are commonly conceptually straightforward to understand by both investors and prospective customers and users (there’s a difference) and they do scale well. Whether they generally monetize well is a different story. But hype is about attention and there are many examples of companies with a lot of attention dancing on the razor’s edge with respect to financial sustainability of their model.

What’s noteworthy is that a significant portion of the commercial “IoT” innovation on the Gartner outline is no longer emerging from or being led by “Silicon Valley”, meaning the consumer-technology obsessed part of our industry. And when you start digging, you’ll find – and this is not surprising when you think about it a little - that the epicenters of where innovative commercial or municipal technologies are being tried out and deployed at scale, and where also a lot of the R&D happens, is where “things” are made and where you find an intact manufacturing or manufacturing enablement industrial base: Germany, Sweden, UK, France, Netherlands, Italy, Japan, Korea, China.

You may be offended I left USA off this list. I’m on the fence about that myself. Large scale commercial applications of digital infrastructure technologies require that there is willingness to invest in infrastructure and its upkeep. It also requires public and political will to socialize some of the cost of such infrastructure investments, like traffic, water, waste, logistics, energy, and environment control applications, also in the form of subsidies to private infrastructure operators. I see interest in infrastructure investments and also in substantial subsidies in strategic technology sectors for the common good and to set up the societies for sustainability in the EU and also in the key economies in Asia. I also see it in many emerging markets elsewhere in the world. I’m not sure I see it in the US that much. 

It does not appear absurd to think of “IoT” as the way how the old, boring world of makers of things that aid us in coping with the basics of our physical existence are going to disrupt Silicon Valley. So, yes, the hype is a big deal. Also because I think the money is may not be going where everyone thinks it’s going.

As we now appear to have the wallets’ attention and as we’re strapping in for the slide down the backside of that hype-cycle summit into the dark trough of disillusionment (that valley, friends, will be deep and dark and I can promise dangerous and sometimes fatal encounters with the dragons of scalability, reliability, and security), it may be worth to triage “IoT” into a few more handy subdivisions that engineers (someone’s gotta do the work) can actually reason about. We can’t actually reason about “IoT”. Overscoped.

One way to slice the space up is by use-case scenarios. That’s the somewhat obvious and seemingly reasonable approach, but arguably yields too many sub-scopes. Here’s a candidate outline of scenario families, anchored largely on “smart” or connected product categories, which I’ll just list here without turning this into a book:

  • Mobility – Cars, Aircraft, Buses, Bikes, Trucks, Vessels, Trains, Traffic Management, Charging, Fuel, ..
  • Factory – Servicing, Manufacturing Integration, Automation, Predictive and Reactive Maintenance, …
  • Cities – Fire, Pollution Control, Emergency and Disaster Management, Law Enforcement, Waste, Water, Pollution, …
  • Energy – Grid, Distribution, Fossil Recovery, Fossil Exploration, Renewables, …
  • Building and Home – Automation, Lighting, Comfort, Safety, Security, …
  • Logistics – Letters, Packages, Containers, Tanks, Bulkware, ..
  • Agriculture, Entertainment, Retail, Healthcare, Professional Services, etc. etc. etc.

You can double-click on any of these keywords and find its own universe of scenarios and use-cases and, often, a multi-billion dollar weighing slice of the promised bounty of trillions. But if we all walked off into the verticals, it would be hard to achieve significant scale effects on the technology development side and it would also seem foolish as the solution patterns for many of these concrete scenarios will undoubtedly be similar – as much as we found that a fairly scoped set of web technology works across a very wide set of scenarios.

Another way to slice up the space is by observed formation of clusters of interest groups. There appears to be a sort of organic self-organization trend going on in the Internet of Things space. When I look at the engineers and companies engaged in “IoT”, I think of four broad groups:

Gadgeteers are enamored by the “T” in “IoT” and largely focus on the things. These are the “makers”, the folks who make light bulbs that can show all colors of the rainbows as you move a slider on your mobile phone, light bulbs with integrated Bluetooth speakers (yep). They make the fitness trackers, the intercourse performance trackers (yep) and emotional robots. This also includes the world’s hackathons full of smart folks doing amazing stuff with their Raspberry Pis, the Arduinos, Gadgeteer boards (sic!), and other hardware tinkering platforms or who are just making their own PCBs outright. Universally, Gadgeteers seem to focus on devices that are designed to improve the immediate personal space in and around the home, in the car, or on (and sometimes in) the body. The common complementary Internet capabilities focus on data collection and remote control of the individual gadget or a small and localized group of gadgets. They also make cars and airplanes.

Gatherers come to the Internet of Things high from riding the hype wave of “Big Data”. As they’re skidding down the lee side of the hype-cycle curve and are fairly convinced that gigabit and petabyte analysis technology will answer any questions they will ever want to ask, they’re now looking around how they can fill their gigabits and petabytes with purpose and what questions actually need answering. Signal words for this area are prediction, forecasting, trend-detection, efficiency analysis. The gatherers are in the game to collect and hoard data, sometimes even with nothing but a hunch that they’ll eventually figure out what to do with it.

Messengers take keen interest in the “I” of “IoT” and wonder how all those things will talk. How can the gadgeteers interconnect their things, how can the gatherers gather, and how can the controllers control if there were no communication routes. Of which there are plenty. The obvious incumbents, the mobile wireless network operators, are the ones who most fanned the marketing flames under the “IoT” hype so far because they’re hurting from saturated markets, sluggish recovery of spectrum investments, and eroding premium services revenues. Yet, many of the operators are badly positioned because they’re not creating the technology they run and the technology they bought into, mostly recently and expensively LTE, is well suited for interactive entertainment experiences, merely covers a sliver of the “IoT” communication needs. Tracking the health metrics of a cow on a farm can be done with much less energy and monetary expense. The Messengers build alternate networks on license-free or deregulated radio spectrums like the ISM band, analog TV frequency whitespaces. The develop ways to power devices while talking to them, or new low energy communication paths in the house or around the body. They also figure out how to connect the things over the internet in a secure and reliable fashion in order to enable the gatherers and controllers do their thing.

Controllers look at all of the above and go “if the gatherer finds that a hundred of these gadgets dispersed around a space just messaged a similar thing we might want to go and do something” – and then do something. Controllers build systems that take advantage of devices deployed in physical space, and of the data collected by them and the insights derived from that data, and build the logic that causes things to happen. They understand that fine-dust pollution in a certain city quarter is too high, will find the heavy trucks driving through the quarter and set them all traffic lights to green to they can get out quickly, and flip the digital road signs as well as the roadblock traffic information fed into commercial navigation systems such that all trucks are made to avoid the quarter until the pollution levels fall.

You could read this list as “Devices, Databases, Connectivity, and Compute/Control Systems”, but that would not be correct. Each of these areas of focus encompasses all those aspects. The focus of a gadgeteer causes a bias towards device concerns, the focus of the gatherer a bias towards database concerns. Depending on which of these stereotypes you talk to, the perspective on what “Internet of Things” means will be vastly different.

That’s ok. Because “Internet of Things” does not really exist. It is the merger of the operational technology space that interfaces with the physical reality with the information technology space that interfaces with the inside of people’s heads. The direct consequence is that, for instance, vehicle manufacturers and industrial automation manufacturers will start considering digital capabilities, services, and operational excellence increasingly as a core competency. It is the advent of the post-tablet, post-smartphone spread of digital intelligence and interactivity into everything. “19 trillion” can only make any sense if everything becomes a computer.

For realizing the digital infrastructure visions across the vast array of scenario families to come to life, we need to let the gadgeteers be gadgeteers, the gathers be gatherers, and the messengers be messengers. There will be great amounts of innovation and declaring victory on “a” standard or “the” winning technology for “the IoT” is not only premature it’s also foolish because of the breadth of the digital infrastructure space, which covers nearly the entirety of mankind’s economic activity. What we must all care about is that communication paths are interoperable, systems are uncompromisingly trustworthy and secure, and that data ownership and privacy is preserved such that private information stays truly private. Only then can we provide a sound and technologically diverse foundation for safely orchestrating slivers of the world in order to make them better, more efficient, safer, and/or more fun.

Oh, and, I'm a messenger.

[Also read: Terminals, Hosts, Peripherals]