Perhaps the very best way to signal the seriousness of your intellectual efforts is not to begin with a children's television cartoon.

In fact, much of it is actually here, from aircraft by SpaceX to the Roomba by iRobot. A key aspect of making gadgets into robots is by making them "smart", by naming them with an IP address and creating a way for communication. The network of such devices is the "Internet of Things" and has been hailed for decades as the next frontier for engineering innovation. Companies like Cisco and IBM have worked at an institutional level to create smart sensors in city grids and industrial sites, pulling out data for use by government and private operators. Production lines can self regulate. Roads know the traffic overhead and can visualize the data. Military drones know how to fly even when their connection to the pilot cuts out.

If that's the future, it's almost entirely trivial. Most people wake up to an alarm programmed in advance. If they really like drinking bad coffee, there's hundreds of models of coffee makers they can already buy that can work on a timer set up the night before. Stupid people have televisions in their kitchen, and all such idiot devices have an on button, usually several. Inability to enter a room in which some bullshit box is not already yammering at you is not a trait we should wish to cultivate in humankind.

One of the persistent drawbacks in your writing is that you tend to approach the future of technology at the intellectual level of a General Electric commercial. It may be that you don't really believe that capitalism makes inferior technology because rent-seeking distorts conception, design and implementation. But reflection on your own choice of illustrations tends to provide support for the position you are ignoring.

We had all this bullshit forty years ago, actually, from the moment there was X-10. When I was a teenager we built all this low-voltage home automation stuff, and used it long enough to get completely tired of it. Having carried it through my life for more than 35 years, I ripped it all out more than eight years ago, and haven't missed it since. Once again, even when it comes to sensor-based technologies, your effort to convey what's going on, not by reading the literature but by absorbing advertising crap from self-promoting meaningless companies, plays you false. Want to know what the sensor revolution is about? Read the work of Deborah Estrin. Interested in "the Internet of Things"? Read the late Michael Dertouzos. Want to understand the approach of current technology-makers to the "physical graph" (which won't be physical)? Study Qualcomm's AllJoin? code.

Once again, trying to learn about a subject of complexity and (supposed) importance by reading company promotional material is not a defensible choice. You know you can't acquire an academically respectable knowledge of automotive engineering or the future of the car business by reading the General Motors consumer website.

Technical nonsense, unfortunately. It's not the "hub of the home" that counts. It's the freedom of the software that runs the sensors and routes their traffic. The router that defines the perimeter of the home network (which might be running privacy-reinforcing software we might call "FreedomBox") is a small, however important, part of the story. But all of this has been written about extensively in writing you haven't found because you were more interested in the websites of snake-oil purveyors.

---

The route to a better revision is familiar: get your head out of low-quality sources. There are very interesting and important things to say about the device cloud training people, the sensor revolution, and the effect of pervasive technology for studying, predicting, and shaping human behavior. There's even wonderful social psychology literature, thanks to Sherry Turkle, helping us to understand the effects of these technological changes on human psychic structure. One can't purport to make the three-card monte patter of business promotion a substitute for serious intellectual effort without making oneself sound ridiculous in the process.

Click Like To Share The Data From Your Home

-- By AlexeySokolin - 25 Apr 2013

Internet of Things

Fifty years have passed since the airing of The Jetsons in 1962. Where is the automated home of tomorrow? Or flying cars, metallic maids and consumer space travel? In fact, much of it is actually here, from aircraft by SpaceX to the Roomba by iRobot. A key aspect of making gadgets into robots is by making them "smart", by naming them with an IP address and creating a way for communication. The network of such devices is the "Internet of Things" and has been hailed for decades as the next frontier for engineering innovation. Companies like Cisco and IBM have worked at an institutional level to create smart sensors in city grids and industrial sites, pulling out data for use by government and private operators. Production lines can self regulate. Roads know the traffic overhead and can visualize the data. Military drones know how to fly even when their connection to the pilot cuts out.

But, this innovation is not just fueled by large companies. Nor does it have to focus on large scale projects. Many garages of hobbyists and enthusiasts today are the playground in which the technology infrastructure for the connected home is being built. It is one thing to have a coffee-maker that can connect to the internet and tell you your coffee is ready. It is quite another to have an script/application that (1) reads information from your sleep sensor to know that you will wake up in 30 minutes, (2) activate the coffee machine at that moment so that the coffee is ready when you wake up , and (3) turn on the TV in the kitchen as soon you leave the bedroom. This is one of millions of potential relationships that can be programmed between physical objects once they learn how to talk code.

Building the Physical Graph (Consumer Web of Things)

The implications for consumer robotics and connected devices, once at scale, are staggering. We are, however, just in the beginning stages of this project. The first step is to create devices that are sensors of the physical world and can output information. Example: a magnet that can tell you if a door in your house is open or closed, or an accelerometer that knows whether you are sitting in chair. The second step is to give this information a place to live. Companies that are focused on a single sensor, e.g., the Nest thermostat, will provide the user with a single dashboard that allows the visualization of data. Imagine this at scale, where each intelligent physical world object can communicate its status--and those signals are collected and standardized through a single platform. And lastly, allow developers to develop software based on this data, and provide write access to physical objects. A smart plug could turn lights in your home on or off based on whether the top lock of you apartment is locked or unlocked. A more complex use-case for the same device could be the cycling of lights in your home when you are away on vacation to resemble a regular pattern of someone moving between rooms, so that your home appears to be occupied and is less likely to be burgled.

The company SmartThings does exactly this: (1) it provides simple, open source sensors to the consumer that can be installed in a home, (2) creates an API where the data for those sensors can be accessed by developers, and (3) enables developers to build applications that tie together real world objects and behaviors. It has sold $1.2 million on Kickstarter in pre-orders, and raised $3 million from early stage venture funds and super angels. Not all companies are targeting the full value chain. Sen.se is an open source attempt at standardize the different data in our lives, from physical to digital, and providing analytics and calculations for people to make actionable decisions. For example -- one can automate an SMS message to be sent at the end of the day updating the user on what their daily electricity usage was in $, not watts. A third company, Cosm, has been in business since 2008 and focuses primarily on the API between devices.

Data and Privacy Concerns

Sensors create data, and if placed in the home, very private data. The companies discussed above are consumer-focused and are storing personal data themselves. They know where you are, when you open a door, the temperature in your house, what applications you use to run your life. They know not just who you are, but all the algorithms that make up how you behave. People will inevitably love the services and will give this data over to pay for them, in the same way that Facebook has swallowed our social behavior and Mint.com knows our financial livelihood. There is a distinct and real danger in private companies storing personal data, and particularly in those claiming to be a personal lockbox. No amount of marketing posturing will prevent them from being used by a powerful government as a source of information on its citizens.

As this technology is still in its relative infancy, there may be a way to still pre-empt how protected the data can be. If the Freedom Box will really be the internet hub of the home, providing a constitutional protection to sensitive personal information, it is every bit as imperative to build into the Freedom Box the ability to map the physical graph from day one. In fact, it may be the killer app by which the Freedom Box can spread into every home across the world.

Sources

• #Set ALLOWTOPICVIEW = TWikiAdminGroup, AlexeySokolin
• #Set DENYTOPICVIEW = TWikiGuest

Note: TWiki has strict formatting rules for preference declarations. Make sure you preserve the three spaces, asterisk, and extra space at the beginning of these lines. If you wish to give access to any other users simply add them to the comma separated ALLOWTOPICVIEW list.