Copy
 
 
IoT news and analysis you can use:
 
 
Can’t read this? Click here
 
 
 
 
 
Stacey Knows Things
 
 
 
 
 
Should connected products come with an expiration date?
 
 
by Stacey Higginbotham
 
Making a connected device requires more than just a chip or an app. The best connected devices are really services wrapped in a piece of hardware. The hardware is a delivery mechanism, but it's not the core feature.

For example, the Amazon Echo is the hottest IoT device around in the consumer market. But the hardware is a wrapper for the personal assistant, a smart home hub and music on demand. 

I've written about this concept before, but I've been spending a lot of time thinking about the ramifications of this shift.
 
 
 
Do connected devices need an expiration date?
 
 
One ramification is that consumers expect that over time the Echo (and similar devices) to do more. So developers are constantly working not just on getting a new piece of physical hardware in the world, but also on adding features to all of the other existing Echos in homes. 

The product development cycle shifts from one focused on getting a new device out the door every year or two, to a bifurcated product development cycle focused on new hardware as well as new software features for all of the existing devices.

This can be rough on manufacturers but also rough on consumers who may find their hardware becoming obsolete, making them unable to participate in new features. Tesla vehicles are a great example of this challenge. Cars from earlier years don't have the hardware for the autopilot features and Tesla Founder Elon Musk says that Tesla owners should just get used to this.

One day we'll likely get most of our products as a service instead of buying them upfront. Until we get there, should our connected devices have an expiration date? This may sound silly, but I think it's going to pop up more and more as a question that deserves an answer.

In 2013 I purchased a $250 Nest thermostat and wired it to my HVAC unit. This was the most I have ever spent on a thermostat and the first time I had actually purchased one. Before the Nest, it was a product that was just "there" and never changed until my AC broke down (and maybe not even then). 

Since I bought my Nest, the company has made several software upgrades and released two new versions of the thermostat.  I have to think that at some point the hardware inside my thermostat may not be able to handle an upgrade or a missing sensor might mean I miss out on a new service. 

In the computer and smartphone world this isn't an uncommon issue, and so we see vendors pledge to support their phones with the latest operating system and security patches for two to three years after the first launch of the physical hardware.

After that, the device may or many not keep working with subsequent updates and they may become vulnerable to security breaches.

Apple, Google, Samsung and others make sure these policies are fairly well understood and also have picked a time frame that matches the lifecycle of the phones or computers.

When it comes to smart home gear (or even things like enterprise gateways) the promise of support is less clear. Most consumer companies don't have anything on their sites or in their terms of service about this topic. Enterprises generally strike some sort of deal.

I asked Belkin about this idea of expiration dates for connected devices because the company makes everything from routers to wall switches that a user installs. Brian Van Harlingen, CTO at Belkin, explained that the company hasn't set any defined date for its WeMo products yet, but it expects to support every product for five years after it has been taken off the market. 

So barring some catastrophe, my four-and-a-half-year-old WeMo switch will still be supported for five years after Belkin stops manufacturing them. Honestly, that feels a bit generous given how much technology can change in half a decade. Especially since WeMo may still be making connected outlets a decade from now. 

So while I'm comforted that some vendors are at least thinking about this, I am concerned that we're not actively talking about how businesses plan to ease off support for an older generation of products. It may not be as bad as asbestos, but un-updatable smart switch in your walls doesn't seem like a fun thing to deal with in twenty-five years.
 
 
 
Are the right people developing and hosting your IoT software? (sponsored)
 
 
Developing secure, resilient and performant software is critical to any successful IoT product. That’s why organizations like Whirlpool and Ford Motor Company trust SpinDance to build and manage their IoT solutions. SpinDance’s integrated team of embedded, cloud and mobile engineers can collaborate with your in-house developers to get the job done right. Or, we can oversee your entire IoT project.

Accelerate your next IoT product today: Download our free whitepaper on building great IoT solutions.
 
 
 
Security stands in the way of Enterprise IoT
 
 
Enterprises are adopting the internet of things at a slower pace in North America compared to the rest of the world, and compared to the interest shown from North American corporations three years ago.

According to an index released this week by ARM and The Economist, interest in adoption of the internet of things dropped in North America while rising elsewhere. 

There could be several reasons for this. In a conversation with reporters about the study, Pete Swabey from The Economist Intelligence Unit said that the slowdown in adoption may simply be a result of North American firms having done the research on IoT without seeing an immediate need to push forward to a pilot or implementation.

The index also included data about what obstacles enterprises see in their adoption of the internet of things. As you can see in the chart below, the top issue is cost followed closely by concerns over security. Toward the bottom, though, is where the chart gets really interesting. Categories like the "absence of a business case" or "products and services do not have an obvious IoT element to them," make me wonder if perhaps we're not explaining well what IoT actually enables.

If you stop thinking about IoT as putting an app or Wi-Fi on everything and start thinking about it as a way to cheaply get lots of data about your business processes, your customers or your environment, then suddenly it becomes difficult to see where IoT wouldn't help your business.

A lot of companies talk about wanting to find a use case for IoT before diving into a project, but perhaps a better way to think about this is what about your business is the most valuable, and how can you better understand how to deliver that value more cheaply?

So if you sell appliances, perhaps you stop thinking about selling washers and dryers and start thinking about how to sell consumers clean clothes when they need them. To get to that point, think about the data you need to make this possible and the supply chain you have to build to control as much of the clean clothes service as you can. 

Plenty of companies are well on their way to thinking like this, and I bet many of them never had an "IoT committee or project" designed to inject some IoT into the organization.

For example, Tesla is already contemplating the future of autonomous cars and inserting language into the sales agreements for its vehicles that prevents a Tesla owner from starting their own ride-sharing business base on a Tesla autonomous vehicle. Will that hold up in court? Who knows, but Tesla is thinking ahead.

So as studies like this one come out discussing the adoption of enterprise IoT, consider where an infusion of data makes or breaks your core business, and what else might become a service provided by someone else.
 
 
 
 
A slide from the ARM and Economist report.
 
 
 
 
 
 
 
8 Tips to Help You Accelerate Your IoT Product Development (sponsored)
 
 
Manufacturers attempting to bring their first IoT connected product to market tend to make many mistakes and struggle mightily during the complex journey. Spending millions of dollars and years developing as well as launching first generation IoT products is not uncommon. In order to help you to avoid those pitfalls, we've distilled our IoT platform expertise into this eBook with 8 best practices to consider when planning, designing and launching your first generation IoT products.

Download this eBook on tips to help you accelerate your IoT product development.
 
 
 
It’s our 100th podcast, which would be a big deal if Kevin Tofel and I were a TV show hoping for syndication, but in the podcast world it means we’ve been at this for almost two years. YAY! We took a brief stroll down memory lane before digging into the week’s news covering new LTE chips for the IoT from Intel and Qualcomm as well as a report from ARM and The Economist that highlights slow growth in enterprise IoT projects. We talk about a few things to see at Mobile World Congress next week, discuss the Orbi router and also share our thoughts on Somfy motorized shades, female personal assistants and shopping from Google Home.
 
 
 
Google’s Home speaker and AI assistant.
 
 

For our guest this week, I speak with Jaoa Barros, CEO and founder of Veniam, about what happens when we treat cars and buses as roving nodes on a mesh network. Venian calls this creating the internet of moving things, and it’s a big, awesome idea. We cover everything from the connectivity needs to autonomous cars to how connected transportation makes cities smarter. You’ll like it.
 

Hosts: Stacey Higginbotham and Kevin Tofel
Guest: Joao Barros, CEO of Veniam
SponsorsAyala Networks and SpinDance

 
 
 
 
I don't really believe this
 
 
In five years 69% of of U.S. homes will be smart homes according to Ovum, a research firm. Ovum classifies a smart house as having a connected device and estimates that by 2021 1.4 billion connected devices will be sold. 
 

It also predicts that each smart home household will use on average 8.7 devices, bringing the total smart home active installed base to 4 billion devices. Ovum tosses everything form connected light bulbs to Amazon Echoes into this category so perhaps almost 70% of U.S. households will have a connected device, but it feels really far fetched to think of those homes as smart homes. Maybe we should just dump that concept altogether. 

 
 
 
The largest markets for connected smart home devices will be in the U.S. and in China according to Ovum. 
 
 
 
 
 
 
 
News of the week
 
 
Dear lord, it's another standards group! Ericsson and Intel  have created the 5G Innovators Initiative (5GI2) which aims to build some kind of standard for connecting devices to 5G networks. The challenge here is considerable, since 5G is a nebulous term, so the effort to build a standard for it should be fun. It's also a challenge because 5G networks are going to be mixed networks using different cellular, Wi-Fi and other technologies, while trying to make them a seamless whole. It will be a standard comprised of standards!  It reminds me a lot of the Industrial Internet Consortiums. The first target is the Industrial Internet of Things. With that in mind, Honeywell, GE and the University of California - Berkeley are the first participants to join the initiative. (Ericsson)

Disney's tech team makes wireless charging real: Engineers at the Magic Kingdom just tested the boundaries of physics (it's like magic) by showing off a wireless power solution that charged devices inside a specially designed room. Instead of devices needing to rest on a special charging pad, the specially designed walls of the room and electricity flowing up a copper pole in the middle of that room generated enough juice to theoretically charge a device. This isn't commercial yet, and there are a lot of caveats (there's a danger zone within 46-centimeters of the conducting pole) but it's pretty exciting research. I could deploy a lot more sensors if I knew I wouldn't have to wander around changing batteries every three months. (Ars Technica UK)

Are you ready for connected power tools? Actually, based on this article, why would you want a connected power tool? The added cost and the irritation of having another app on your phone make dumb power tools look just fine. Looks like so far instead of cool features, like the ability to better take measurements or plan a project, these apps are pretty basic. (Huffington Post)

Calling all SDN nerds: I used to write a lot about software defined networking, which is a way to make it easier to manage networks as software as opposed to hardware (that's the super short version). It's pretty game-changing stuff because it makes network management easier and scalable. Now Avaya is using SDN to provide security for the internet of things. Basically, each device on the network gets assigned a profile and a zone where it can communicate. Devices that violate that zone or profile by trying to talk to other devices they are not supposed to or to outside sites that are abnormal for it, get flagged. I like it. (Network World)

Where is the uncanny valley for personal assistants? Do you want Alexa calling you by your name? Or your nickname? (Medium)

A deep dive on IoT and insurance: Sure, we've all discussed how connectivity and sensor data change the insurance industry at a high level, but if you want to go really deep into the language used in policies then this story is for you. It's filled with jargon, but it also presents excellent questions such as, "If you incorporate another organization’s IoT technology into your product, how will your business interruption insurance respond if that technology malfunctions?" That's something I know contract lawyers for enterprises are working on right now. (The Indiana Lawyer)

Good news and bad news on IoT security: After the Mirai botnet attacked networked DVRs and routers, the internet of things took a lot of flak for being insecure. I hated this characterization because it lumped newer smart home tech with older insecure connected gear. This story outlining the threats posed by the internet of things gets it right. It lays out why there's a real problem and explains which devices people should worry about. (Internet Society)

Fitbit reported its first loss as a public company: Lackluster holiday sales pushed Fitbit to a loss for the fourth quarter of 2016.  The wearable maker this week reported a loss of $146.3 million down from a profit of $64.2 million a year ago. Revenue also declined, falling to $573.8 million, which was down 19% from the previous year. Analysts are worried that fitness trackers are more of a fad than a necessity for users. However, Garmin saw its revenue rise 10% on sales of wearables. Hmmmm (Fitbit)

Ready for Property as a Service? This blog from an architecture firm discusses how the nature of renting and owning property will change as sensors invade corporate campuses, apartments and hotels.  The idea is like office rental community WeWork, which offers users a subscription to office space in their locations around the world. What if we applied that model for vacation rentals or retirement homes? Hat tip to Open Sensors for the discovery. (Woods Bagot)

Idea for an insanely cheap Bluetooth radio: Cambridge consultants says it can cut the cost of Bluetooth radios tenfold to 7 cents by eliminating the analog components of the radio. It's using its all digital radio tech called Pizzicato to cut costs, and engineers at the firm anticipate using the same tech to shrink radios to such a small size, that a cheap "smart dust" becomes possible. I'm eager to see this, but I'm not holding my breath. (Cambridge Consultants)

If you want a cool podcast: This episode of the Smart Kitchen Show discusses how flavor can be considered a technology and how smart home tech can make even exotic flavors more accessible. It's a nice listen. (The Smart Kitchen Show)

Indiana just built an IoT Lab: It looks fine. 

A pest firm just built an IoT mousetrap: It looks pretty worthwhile.

More Stacey: My smart home column in PC Magazine deals with geofencing your garage door and an effort to find connected candelabra bulbs. (PC Magazine)

 
 
 
 
 
 
Stacey Higginbotham's weekly Internet of
Things news update and analysis.