This week could end up being a turning point in the almost decade-long effort to make the smart home a reality. When I started covering the Internet of Things in 2012, everyone was thrilled to take the power of smartphones and link them up to everyday devices as a source of remote control (WeMo outlets, for example) or as a way to access the everyday device’s intelligence (such as with Nest thermostats).
But whatever we hoped the smart home would become, it has not. Devices are still dependent on phones, third-party integrations, individual partnerships, and companies choosing which devices they will support. Products are too expensive, users are confused, and security is an abysmal mess. As a result, people are cautious about connected homes or, in some cases, actively turning away from them.
I have long blamed the sad state of the smart home on a lack of a standard. On Wednesday, I may have gotten my wish. Apple, Amazon, and Google all said they would support a new standard for the smart home called Connected Home over IP, or CHIP. So what will that mean, exactly?
To be honest, almost 48 hours after the bombshell announcement, I’m still not sure what it means. But I can make some good guesses, and I know what questions I’ll be asking. And based on what I do know, I believe it will be good for both consumers and developers, and represents a real threat to companies trying to build hardware for the smart home. It also opens up what I believe will become a new layer in the IoT “stack,” which is where companies should be focusing their efforts in order to distinguish themselves from the competition.
The CHIP standard will be developed under the Zigbee Alliance; a rough draft from the working group is expected in late 2020. While no one is making promises that your existing smart home products will work with the new CHIP standard retroactively, people I’ve spoken with who are involved in the various organizations that make up the alliance believe most of the hubs released over the last two or three years that have BLE, Zigbee, or Thread radios will be able to handle the conversion to CHIP.
The goal is to create a radio-neutral, IP protocol that has a set of pre-defined schemas for connected devices. That schema will let devices understand what objects they are “talking” to and what that object can do. So if it’s a thermostat, the schema will share heating/cooling/temperature/fan and other appropriate data and have a common way to tell the thermostat how to turn on and off and move the temperature up or down.
The result will be a way for companies to share data about devices and interoperate at the application layer. Another standards organization, the Open Connectivity Foundation, has been trying to do this for years. What’s different about CHIP is that it includes the biggest three smart home players in the U.S. along with notable smart home brands such as Samsung (SmartThings), Resideo (Honeywell), IKEA, Signify (Philips Hue), Somfry, and Legrand.
Will that be enough to make this standard stick? Because with big companies can come big company egos. But I think everyone is at the table because they understand that if they want to build a real business around the smart home that extends beyond mere home automation, they have to build the infrastructure first. The schema is the infrastructure layer.
The prize here isn’t owning the smart home. The prize here is building the digital plumbing for an era where most consumer devices are connected. If we don’t have the plumbing we can’t build light bulbs that talk to motion sensors, and we also can’t build sleep-tracking devices that share information with our cars to prevent us from driving while exhausted.
A standard like CHIP has the potential to become the equivalent of HTML for translating a real-world product’s functions to the digital world. It may start with thermostats and light bulbs, but if it becomes widely adopted in the home, products such as medical devices, appliances, and even cars would want to work with it. It’s no wonder big companies want to make this happen.
The future of the connected home is all about bringing the data our devices collect into a format where it can be used in ways that help people perform better, improve sustainability, and become more productive. Or at least that’s my hope. No standard will help with business models whose architects are hellbent on getting people to spend more and work themselves to the bone in order to boost a corporation’s bottom line.
But enough about my fears of unfettered capitalism and how it could steer the internet of things into “Black Mirror” territory. Let’s get back to CHIP. With such an ambitious goal at stake, and the fact that much of the underlying technology that will underpin it is already built, I think it has a good chance of moving forward.
But there will be politics involved, and various winners and losers will be deemed as the pieces get put into place. In the short term, we’ll likely see problems as companies dismantle existing integrations and discover that certain devices can’t support CHIP. Consumers will start to win midway through the process because they will have a real choice as to what devices they want to use and will no longer have to worry if, say, their doorbell will work with their preferred brand of light bulbs. But over the long term, the unfettered capitalism side of things—if left unchecked—may turn the home into a panopticon that most people do not want.
Manufacturers of connected devices should worry. While they will no longer have to support half a dozen platforms for one light bulb, it also will lead to what my friend Lee Odess, who writes over at LinkedIn, calls “a shift or squeeze opportunity for incumbent manufacturers.” Odess, who recently left his role at Allegion, writes:
It’s going to get harder for some companies to differentiate if they do not have IP, intellectual property, and a clear brand relationship with consumers. I fear we may see some weird flexes in the market by manufacturers where they hold back some features and make them unavailable unless you use their own app. Not sure yet but we will see. Not sure if this is a good thing.
By putting the standard at the application layer, it’s making the layers underneath look more commodified. Sure, companies with strong design, quality build, and great brands will still be able to score a premium for their hardware, but it’s hard to see places where they will be able to establish value at the software or services level. That said, for companies that have good services and direct relationships with customers, this is a chance to build deeper links to make their services stronger and more appealing to a consumer, or even to build a platform that lets a consumer tailor a connected home solution to their individual needs.
Developers are going to be happy in the short- and mid-term. In the long term, they will succeed if they are creative. Developers will now be able to build software for connected devices once and presumably have their software work with everything, which will make their lives easier when it comes to thinking up new features, as well as supporting and maintaining them. And that’s where it gets interesting. In the long term, developers will have to come up with features to build better user interfaces, design software that can pull data from different sources and make up new services based on algorithms.
So developers with big visions and new ideas will benefit in this world, but they will also have to fight against Google, Amazon, Apple, and other big names as the incumbents. Essentially, innovation has moved from figuring out ways to get gear to work together to figuring out ways to actually use such gear to improve people’s lives.
There is a lot about CHIP that we don’t know yet, and many basic things that the standard may never cover. For example, security isn’t really addressed here, although there are things that could be included in the standard that could help security. Nor does CHIP address problems with device onboarding and over-the-air updates. Moving homes and taking things off a network might still be difficult, too.
The FAQs associated with the CHIP project say some of those issues might be tackled with the standard, but they also might not. For now, I’ll be looking for more information about how parties are trying to safeguard their interests around getting access to consumer data (for example, by forcing someone to download the manufacturer’s app so the manufacturer can get email and other data), how much of the user interface the companies want to control, and even basics like how much processing and memory power the new standard will require. If it requires too much processing power or memory the standard will add expense and make it hard to deploy CHIP to sensors.
There’s a lot that can happen with CHIP. But to really evaluate the standard and its likelihood of success, it’s important to understand that if it succeeds it won’t just be a way to get devices communicating in the smart home, but a way to create the infrastructure that will underpin the next generation of services.