I know there are dozens and dozens of open source initiatives to build 'the house of tomorrow', and even Microsoft is investing big on it's luxury robotic house.
Which begs the question: what do you think of home automation as a business? Is it still a far-fetched idea for the rich and famous?
The industry and market-base is still way too fractured.
One of the biggest limiting factors is that in order to be very useful and functional, there has to be a moderate degree of programmability in the device that is acting as the server/controller for the home. This limits the effective market to geeks and tinkerers, who are also the most common build-vs-buy crowd.
Homeowners, in the general sense, do not want to pay monthly fees for service and are leery of installing a system that controls their home that they feel they have little or no personal control or understanding of.
Where the HA industry has seen some decent business is in specialized lighting and HVAC control systems.
BTW, what do you think of the microsoft home initiative?
Even back in 1995 when I got started in HA there were several choices for relatively cheap/flexible controllers.
The problem has been that the stereotypical homeowner does NOT want to program anything. VCR's blinking "12:00" was the joke of the 90's, and the concept is still true today, even if VCR's don't really exist anymore.
Can you imagine the average person wanting to program an HA system? Telling it when to turn lights on and off, how to factor in shifting sunrise/sunset times, occupancy sensors and the like? It's very very difficult to make a system that is both powerful and simple, unless you break it down to a very very basic function.
Most people would love an automated home experience, but the setup is daunting. Especially when you consider all the edge cases... Turn on the interior lights at 30 minutes before dusk, but not if we're away or out. Shut off the HVAC system or zone if a window is open, unless the door to that room is closed or the humidity is very high (like in a bathroom), and so on.
I have personally found that if you don't carry the system logic far enough, it is just annoying and perceived as "stupid". In order to make the system very intelligent it needs to be rather complex and incorporate a lot of environmental (temperature, humidity, occupancy, seasonal, etc) inputs.
Heave on top of all that there is no one system that covers all the basic sub-functions really well: - Phone
- Security
- Lighting
- HVAC
- Irrigation
- A/V (multi-room or single room).
- User Interface (remote controls, keypads, input/control panels, etc.).
There are a some systems that do a couple of these well. HAI OmniPro alarm systems do a pretty good job with security, HVAC and lighting. Their audio is passable, but not great, and their irrigation is kind of lacking.
The JDS Stargate is kind of outdated, but does a good job with lighting and basic I/O (relays, TTL inputs, etc.). It has no security, but integrates with Interlogix alarm panels. But, you pretty much have to program both devices independently.
Russound makes great multi-room audio, and they have a robust serial interface, but not much direct plug-and-play capabilities.
All of the various sub-systems seem to almost, but not quite, talk to each other.
I've personally moved to an OSX based system (Indigo) and do apple-scripting to handle a lot of things.
In the past, I had a JDS Stargage, HAI OmniPro, and PC running an application I wrote to handle email, web and IM interfacing to the Internet, plus some PIC-based stuff I build for other one-off sensors and interfaces to things.
My stuff works well, and most people can use and understand it, but this is the result of no small bit of effort and many custom bits...
Not exactly joe-consumer friendly.
As for MS-home, they've been in and out of this space for 10 years. I don't expect much from them, and like most of their solutions, I find the approach too closed.
I'm curious, do home users go for automated blinds, room lighting, or TV controls?
The one time I thought about controls was when I had to replace a leaking skylight in a cathedral ceiling. I could do it myself for $300-400 or I could get one with electrical controls plus a carpenter plus an electrician for $1500-2000. I went with DYI and a long pole to open and close it.
The problem comes from trying to integrate the 3 things you mentioned, as an example.
If I am going to watch a movie on TV, I probably want to control the blinds and turn the lights down or off. If your blinds are IR controlled, instead of a keypad and/or RF control, you have a decent probability of being able to turn on the TV and close the blinds, as long as the IR target for the blinds is close enough to the TV so that the user holding the remote control doesn't have to wave it around the room.
But, there really aren't any lighting controls with direct IR interfaces, and very very few multi-remotes have IR libraries for automation devices. So, you need to install the room lighting, setup an IR interface, most likely either program the remote or the IR interface so that one can interact with the other, and then setup a macro on the TV remote.
Most people would rather just do things manually.
As the 12:00 blinking VCR phenomena shows, the majority of people are just not into doing this sort of stuff for themselves. Only a full service solution will meet their needs and to do that well is very expensive and people change their mind when they find out how much it is going to cost.
In 2009, it still is for the Rich and Famous and the DIYers.
otherwise, if its just frivolousness, you'll stay inside the realm of the people with the money to toss around.
What if people WANT frivolity? Look at the web: the most popular websites are all basically focused on chatting with people and watching silly videos. Maybe something like Boxee (who takes the usual appletv thingie and turns it into a 'social' stuff) would tickle?
(can't think of anything interesting enough though - comparing your fridge with your neighbors is booooring at best!)
Are you planning on installing a pre existing product line (eg cresteon, amx, control4) or creating a product like linux mce or misterhouse?