I strongly agree that the home automation industry has a history of failing to deliver, spanning the early 1990s to the early 2010s. In the late 1980s, during

I strongly agree that the home automation industry has a history of failing to deliver, spanning the early 1990s to the early 2010s. In the late 1980s, during the time I was in primary and high school, my parents had X10 lamp and appliance modules throughout their home. The main benefit of this was the ability to remotely turn on lamps, which was especially useful for those that didn't have wall switches and/or had their switch in some hard to reach location. But a secondary benefit was scheduling. Since my parents both worked and I as a negligent teenager couldn't be bothered to remember to turn on outdoor lighting in the evening, there was an appeal in having the system simulate a "lived in" home by turning on key outdoor and indoor lamps at random times. Initially, my father set up a simple schedule using a programmable X10 controller, but we found this fairly limited. I ended up putting together a menu-driven application that signaled an X10 controller and automated the system from my computer. Why develop the discipline to turn on evening lighting yourself when you can make a program to do so? The controller used a fairly simple serial protocol. Using the computer gave us a great deal more control. We could give more randomness than the controller's "1 hour window" and account for vacations by also automating weekends, among other things. Fast forward to today and in my own house, I am using... wait for it... the same X10 controllers to allow me to remotely turn on and off lamps and appliances such as our attic fans. I no longer have that old application I built in the 1990s, but I also don't really care about the automation as much as simple remote control. Here are my beefs with the automation industry today, and where I disagree with the author: * I feel Nest is a meager representation of where the industry should be heading. It's thoroughly uninteresting to me: I already have a wireless programmable thermostat that I can use in any room of the house. It doesn't interface to the Internet, but frankly, I would never use that. I live in Los Angeles, so I don't ever need to kick-start the heater to warm the house up before I arrive home (or whatever else you can do with an Internet-connected thermostat). Maybe that's more interesting in other parts of the country. I've heard in some places, people remotely start their cars. Moreover, I turn on and off and fine-tune the temperature as my comfort merits. I don't want an energy-saving algorithm interfering. * Meanwhile, X10 has not changed a bit since the late 1980s. The modules do the same thing they always did. The modules even look basically the same. They still send their signals over the electrical wires. I suppose they've been comfortable doing what they do and leaving it at that. * I've recently converted 75% of my household from fluorescent and CFL to LED lighting. Now I have periodic X10 signaling problems. Superficial research suggests the X10 power-line signaling can get garbled when mixed with LED lighting. Great. However, based on the above, I am considering replacing my X10 modules with Insteon modules that augment their power-line signaling with wireless signaling [1]. I don't yet have any experience with Insteon, but it looks like a promising (if expensive) successor to X10. The author's chief complaint would appear to be that systems like Insteon are too complex for a non-technical consumer. That is probably true. Making systems like this easier to use is good. However, I fear that today's knee-jerk solution to complexity is remove options to simplify user adoption, gather as much data as possible, send data to central repository for analysis, control system from corporate headquarters. In other words, in today's world the plain "cloud" will solve all your problems as long as you surrender yourself. Perhaps unfairly, I have boiled down to the same fundamental matter as many technical debates: how much control to give the consumer? It's the underpinning of the plain-cloud versus (re)decentralized Internet debate. As a decentralization partisan, I am emphatically in favor of giving the consumer control—of their data, their house, whatever. I don't believe that the only viable path to simplicity is to absolve the consumer of any thinking. One can conceive of an Insteon system with the ethos of Nest. Plug in the controller and all of the modules, and the system will figure out how to control your devices, lamps, and security on its own by sending usage data to Insteon Inc. (Don't worry, we don't share this information with anyone but our most trusted advertising partners). Sure, you may have a lamp turn off while you're reading, but that's just the system telling you it's time to sleep. The convenience and energy savings are worth it. I'm not saying it's gonna be Electric Dreams in your house, but I am saying I don't want your Edgar playing music on my radio. [1] http://www.insteon.com/2635-222-on-off-module.html http://www.insteon.com/2635-222-on-off-module.html