Google is building an environment to enter into the internet of things, just like it did with Android in mobile.
“MavHome [an acronym for ‘Managing an Adaptive Versatile Home’] operations can be characterized by the following scenario. At 6:45 a.m., MavHome turns up the heat because it has learned that the home needs 15 minutes to warm to optimal temperature for waking. The alarm goes off at 7:00, which signals the bedroom light to go on as well as the coffee maker in the kitchen. Bob steps into the bathroom and turns on the light. MavHome records this interaction, displays the morning news on the bathroom video screen, and turns on the shower. While Bob is shaving, MavHome senses that Bob is two pounds over his ideal weight and adjusts Bob’s suggested menu. When Bob finishes grooming, the bathroom light turns off while the kitchen light and menu/schedule display turns on, and the news program moves to the kitchen screen.”
Science fiction? Certainly, in 2002, when five researchers published the seminal paper The Role of Prediction Algorithms in the MavHome Smart Home Architecture, the paragraph above was beyond anybody’s wildest dreams. But no longer. On January 13, Google acquired Nest Labs, a company devoted, in the buyer’s own words, “to reinvent unloved but important devices in the home such as thermostats and smoke alarms.”
What does that mean? Why has Google bought a thermostat company?
The purchase of Nest is the second largest of the 146 acquisitions made by Google in its fifteen years of existence. Only the acquisition of Motorola Mobile, for $12.5 billion was more expensive. Google is paying the $3.2 billion for a firm that is only three years old and sells two types of products, thermostats and fire alarms, in three countries: the US, Canada, and the UK. And it is not paying in 100 percent in cash.
If your house is burning down you’ll now get gmail ads for fire extinguishers
— Sam Faulkner Biddle (@samfbiddle) enero 13, 2014
Nest can be the base that Google will use to build an environment to enter into the internet of things, just like Android did in mobile telephony. And, for society as a whole, Nest’s impact could be similar to that of smartphones. Nest is a first step towards the Smart Home, but also risks taking us closer to a dystopian future in which machines—and the companies that own them—know what we do… and what we are going to do next.
To fully understand the implications of Nest for the firm and for society, we must answer three questions: What does nest do?; What does the acquisition mean for Google?; and, what does it mean for the rest of us?
What does Nest do?
Nest was founded by Tony Fadell (one of the fathers of Apple’s iPod) and Matt Rogers (also from Apple) in May 2010, although its first product was not in the market until September 2011. It makes two products: fire alarms and thermostats. The company has around 200 employees, of which at least half of them come from Apple. Its finances have not been disclosed.
Nest works on home automation, using machine learning. Its devices do data collection and activity prediction. They employ self-correcting algorithms to predict the house’s inhabitants’ movements, patterns, tasks, and interactions. Based on that data, the devices program themselves. Therefore, as described in the academic paper cited above, the thermostat self-programs to minimize energy consumption. In practical terms, it is the Smart Home made into reality.
The most interesting piece of equipment of Nest seems to be the thermostat, since it has the most advanced self-predictive elements, and also an API available for developers to sync up with their devices. In practical terms, that implies a technological platform open to apps created by third parties.
Nest’s thermostat needs to be connected to a Wi-Fi network to get automatic software updates. Paradoxically, this has been a problem in terms of the user experience. As one customer recently posted in the site’s app site (for iPhone): “Once they [the thermostat] lose network connectivity, they keep trying to connect. As soon as the net is disconnected, which happens daily, even for just 20 seconds, it is offline and requires you to physically reconnect the device.” In order to make Nest’s products function, the buyers must install the company’s app in their cellphones.
What does Google get from Nest?
Nobody knows the answer to that question, but here there are some educated guesses.
The most obvious benefit for Google is getting the expertise of around 100 former ex-Apple employees (50 percent of nest’s staff), including former heavyweights in the firm, such as Fadell and Rogers. But this does not justify the price it has agreed to pay for Nest.
Nest’s two critical elements for Google could be the data compiled by the company, and the platform’s possibilities of becoming something like the Android of the internet of things.
In terms of data, the Nest’s devices can obtain an amazing array of information about the habits, interactions, and activities of anyone in his or her own house. They combine a physical component (the device), an information element (the gathering of data for decision making and self-readjusting), and communication (software to re-route the information). That trove of data can give Google unprecedented access to what people do when they are not connected to the internet.
The platform’s possibilities are also vast. Due to its communication component, Nest’s devices can be linked to all the elements of the Android platform, and, in the future, to other Google products, such as the driverless car (if that project succeeds), Google Glass, or other hardware. Nest also allows the development of apps, so, in the future, it can become an independent platform for Google in the internet of things. So, the Nest platform could be the base to interconnect, for instance, Google Glass to Google’s driverless cars and other products. As Rogers wrote once the purchase was made public, “I’m betting that there’s a lot of cool stuff we could do together.”
The problem is Google’s strategy in the internet of things is far from clear. Two weeks after buying Nest, the search engine sold the biggest part of Motorola Mobile to Lenovo, only two and a half years after buying it. Even after excluding the amount that Lenovo paid ($2.9 billion, and the income from the previous sales of other assets of Motorola), Google has lost $6 billion in the whole operation, although it retains most of the 17,000 patents of Motorola, and part of its R&D division. In the case of Nest there seem to be no significant intangibles to profit from.
Google has also been on a shopping spree of companies that develop self-correcting algorithms (read: artificial intelligence) and robots, such as DeepMind, Schaft, and Boston Dynamics. Following its tradition, Google has been anything but clear on explaining its plans. On January 30, in the conference call that followed its last quarter results, Nikesh Anora, Google’s senior Vice President and Chief Operations Officer said, when asked by an analyst from Crédit Suisse:
“As you know from the Nest acquisition, we keep to continue to innovate, we continue to be committed on hardware, in areas that are promising new frontiers.”
Google’s achievements in hardware have been so far modest, to say the least. Setting aside small gizmos, such as ChromeCast, the companies biggest product, Google Glass, is still in an experimental phase.
Google, Nest, and the rest of us
Google is a data-driven company, that obtains 85 percent of its revenue by selling ad space tailored to its users. With Nest, it can have access to data of what its customers are using while they are not on the internet, or when they do not carry their Android cellphone with them. Nest can find out the behavioral patterns of the individuals at their home, including the time they spend doing different activities, and even how many people are in a room. Empirical research shows that, contrary to what intuition suggest, people’s behavior can be surprisingly repetitive, so prediction algorithms can live up to their names and effectively predict what we are going to do.
It is true that Nest has said that it will not share its data with Google.
However, what would impede Google to change Nest’s terms of service in the future? Any gmail user that accesses YouTube without registering is asked to write down his name, and the screen display is designed to make him effectively put his gmail name in the ‘right’ (for Google’s interests, of course) box. Will this happen one day with Nest? Will Google be able to know, and to predict, where will we be at certain hours?
As a way to imagine that future, something out of “The Jetsons”, let’s go back to the paper The Role of Prediction Algorithms in the MavHome Smart Home Architecture:
“During breakfast, Bob notices that the floor is dirty and requests the janitor robot to clean the house. When Bob leaves for work, MavHome secures the home, and starts the lawn sprinklers despite knowing the 70 percent predicted chance of rain. Later that day, MavHome places a grocery order for milk and cheese. When Bob arrives home, his grocery order has arrived and the hot tub is ready.”
Is Bob going to rely on one big company—Google—to take care of his life in that way?