Was It Too Soon to Be Sustainable?

Any two of these things might have been reasonable, though I feel that the sheer size of the house is ultimately the problem - it's going to be a lot of work to maintain a house on this scale in the best of circumstances.

And with the blinds, not to automate this is foolish, because this is something which has to be done constantly. Unlike snow and dirt which only has to be removed occasionaqlly.

But there isn't. Maintenance costs are still costs. TANSTAAFL.

Back in '01 I went to a job interview at a home automation company. At the time, there was some major hype surrounding the industry and I was very interested in the topic. The meeting was in a fancy conference room showcasing their flagship product: a touchscreen built into the wall that controlled some lights and such. More than anything, it was internet access built into the wall, showcasing things like Netflix and videoconferencing.

I showed up to the interview completely unpresentable: jeans, tshirt, sneakers. Everyone there wore suits and ties. The CEO interviewed me and told me upfront that I was not the right person for this job. He still went through the motions of asking me some questions and then gave me advice on my future endeavors. Just before kicking me out the door, he hesitated and then asked me what I knew about home automation.

I told him about my window shade controller that I built using some NPN transistors, a free H-bridge sample that I got from National Semiconductors, and a worm gear salvaged out of a dead VCR. For a U-joint connecting the motor shaft and the shade hook, I used a small length of duct tape. Wiring was CAT-5 (for signal and power), rotational position sensor was a bent contact switch and software was written in C running under Linux. Hardware interfacing was done via parallel port, since you can have 8 outputs and 2 inputs via any printer port (memory little rusty, but I had a few spare ISA cards that gave me additional I/O ports). A cron job opened the shades in the morning so I can wake up to sunlight, and closed them in the evening for privacy.

He chuckled, disbelieving that I could have pulled off such a project. So, I gave him a live demonstration -- using his product, I pulled up the control panel webpage (you guys remember days of dyndns?) where he could see a live webcam with timestamps pointed at a window, and demonstrated the "close" and "open" shade operations while he watched. He hired me on the spot.

Fast forward to a year or two later: "home automation" as an industry was apparently never into automating homes. Instead, they went for very expensive home theater setups with glittery control panels and a couple of remotely controlled light switches as an afterthought. We had so much technical capability, but instead settled for very artificially-priced gimmicks ($10k for a "base model".) I asked my boss why our price sticker was so shockingly high for something that does next to nothing, and his reply made all the sense in the world: "Because that's not the market we're aiming for."

In other words, the home automation industry of 2000s was aiming for the comfortably wealthy segment, who could afford these prices built into their mortgage for a prestige label ("home automation"). The companies built up partnerships with home builders, but only cooperated on the cosmetic level of home automation without doing any actual home automating. IMO, if you are doing any serious home automation, the best time to build the technology in is while the house is being constructed, and these industry behemoths ruined the name for anyone doing serious house integration.

Nest seems like a great leap forward, but mostly the leap is in the interface. Arrays of confusing buttons got replaced with a knob. Instead of entering the secret handshake to select Tuesday morning, you can just turn the knob down on a Tuesday morning. However, that's not home automation quite yet: that's airconditioning automation. For a whole system to work in concert, a thermostat would work in a designed unison with window shade controllers, with temperature sensors, with wall sensors, with air pressure sensors, with charge level indicators from the solar array and ... things I never even thought about. Designing homes with all of these subsystems in mind would be a forward stride.