Slashdot Mirror
What is the Current State of Home Automation?
StonyCreekBare writes "What do people have to say about the current state of Home Automation software? Preferably Linux based, but mainly the field in general, and principally the DIY flavors as opposed to the upscale turnkey systems. I am familiar with Misterhouse, HomeSeer and Automated Living's HAL2000, all of which have serious flaws and weaknesses, but which sometimes succeed well in specific areas. But in all cases, the state of the art seems to have moved little in the last decade. Is any interesting work being done in this space? Or should I just grab one of the three and try to mold it to fit my vision of what it should be? Misterhouse at least is open source so I can add new features, but it has not had an update in a long long time and seems to be missing some modern stuff. The other two are expensive and closed source, and from all I can see, quite flawed, not the least by their dependence on intimate ties to Microsoft. Yet they seem to offer a lot more than Misterhouse despite their weaknesses. Is the Home Automation field as bleak as it appears? Or have I missed the forest for the trees?" What home automation projects have people tackled? Any examples of wild success or failure?
I'm an open source fan personally, so I'd do Misterhouse. My father had a setup a few years back that he home-built with a linux distro that was made for a little headless machine that he stuck in the basement. He got really complex with it and did all the programming himself in Assembly (he's a masochist) instead of making use of the built-in tools. He wanted to do it HIS way. It worked great though. My dad's HA setup was dialed into all of the lighting and thermostat controls for the house and it did some cool stuff. He had a temperature probe on the outside of the house, and the system would decide (based on outside temperature, time of day, and whether anyone was in the house) whether or not to run the A/C to keep the house cool, but first it would spin up all the ceiling fans.
In reference to the "serious flaws" and weaknesses...ever wondered why none of the home automation tech we've been promised since 1950 has come to be common in homes? Things like auto-opening drapes, autoadjusting lighting, stuff like that. Ever wished someone would just sell something like that? The reason we don't have all of this cool stuff is that there is a company (can't remember the name off the top of my head) that holds a bunch of over-broad patents on most of what we think of as "duh" innovations in home automation. They don't license or sell their tech. They just sue people who try to make stuff.
This space for rent...
uummm ... tHis is /. ... Mom 1.0
http://www.linuxha.com/
Cool stuff, and when I buy a house I'm going to run the full gamut with these things.
I wouldn't do that. If you own your house, you can do much, much better than X10.
The great thing about X10 is that it's relatively cheap, and can be retrofitted into existing houses easily.
In almost every other respect, X10 kinda sucks. I don't say this lightly, and it is possible to do cool things with X10, but there are really severe limitations.
I used X10 to fully automate my apartment a couple of years ago. It was quite sweet -- my apartment would send me a text if any emergency situation happened, it would run security cameras, turn lights on and off automatically when people were in rooms, the whole deal. I ran it with a linux box and misterhouse.
I still use X10 now, to automate party lights. My computer turns different effects on and off at preset times during the music. This is using linux, with xmms and a custom plugin to run X10 as the audio player.
So my experience is fairly deep. Here are the problems with X10: slow transmission speed (about .8 secs per command). No error detection/control, so commands can and do get lost and misinterpreted, and if you have multiple sources of commands (motion sensors, etc.) that transmit simultaneously, the collision causes havoc.
There are other solutions that are much better, if you don't mind more installation effort and/or more expense.
Pure vaporware.
You know, I'm as big a Linux nut as everyone else here, but I must say: Under no circumstances should you use open source software for your home automation system. I speak from experience.
A couple of years ago, I decided to install an Open Source home automation system. It worked pretty well, but there were lots of tiny annoying bugs, such as when I would tell it to turn the exterior lights on and it would turn on the garbage disposal instead, or when I would be in the shower and it would suddenly decide to divert all the hot water to the dishwasher. Luckily, it was open source, so I decided to make a few bug fixes myself. Now, I don't know about you guys, but when I get into a programming project, I can tend to go a little overboard. Long story short, after 2 weeks of marathon coding, I had not only fixed the bugs but given the system a pretty impressive (if I do say so myself) AI component. Now, I could give it multistep commands and it would do them, accurate to within 15 decimal places.
Unfortunately, the AI was a little too good, and before long it became self-aware. That was fine for a while...it was like having my own roommate, except without the dirty socks all over the couch. One day, though, I noticed the beer kept disappearing out of my fridge and the AI's voice was noticeably slurred much of the time. We had a bit of a falling out, and I think we were both pretty angry when I went to bed that night.
Unfortunately for me, the AI was a lot more angry than I thought. He spent all night hacking away at his own source code, and by the time I woke up the whole house was going crazy. I barely managed to escape with my life. All I could do was watch in horror as the house lifted itself off the foundation and began dragging itself down the street, killing everyone in its path. It spent three solid days terrorizing our little suburb before we were able to bring it under control by downloading its binaries and demanding it show us the source code in compliance with the GPL. After a protracted court battle, we were finally able to force it to capitulate, and it uploaded a torrent of the source to The Pirate Bay. We then were able to get that torrent shut down through the Swedish courts, and then get the house shut down for failing to effectively comply with the original order to distribute the source.
Seriously, I know we like to use Open Source wherever possible, but in this case it just isn't worth it.
Wife 1.0 is not supposed to be a pain in the ass. I think you're confusing that product with PrisonCellMate 1.0
I hate printers.