Migration Strategy

[upstatemike]

I have been struggling to come up with a good migration plan to upgrade about a hundred X-10 devices and possibly my HA controller to newer technology. Fortunately there have been some excellent discussions lately on things like the pros and cons of various new switch protocols and the features (or lack) on new HA hardware and software offerings. After taking some time to digest all of this information I have formulated my best migration strategy. I have decided to transition to an un-automated house. I think this is my best option for a number of reasons:

First, by not migrating a hundred plus X-10 devices to some other protocol I will save a ton of money. Probably enough to buy a really big flat screen TV!

Second, by removing my X-10 devices I will save a huge amount of time that won’t be consumed with filters and couplers and boosterlincs that have to be reconfigured every time I move a TV or computer. This will probably be the equivalent of getting a couple extra weeks of vacation each year.

Third, By getting a plain fire alarm panel instead of one that is HA integrated or HA compatible, I will be able to have several zones of 2-wire smoke detectors and not have to mess with the extra wiring and expense of using 4-wire smokes with end-of-line relays.

Fourth, Even though my RainWise weather station is a leading product in it’s class, there has been no direct hardware or software support for it within the HA community. I currently have to run a dedicated PC to collect weather station data, a Homeseer plug-in to move the data onto another dedicated PC running Homeseer, and a second plug-in to get the data into my Stargate controller. That is too much hardware and software (and too many steps) just to update a wind speed variable! If I use the weather station as a stand-alone product I gain 2 free PCs.

Fifth, I have the Slim Devices Squeezebox which fills the niche once held by the Turtle Beach Audiotron and before that by the Nirvis Slink-e. (Both great products and I own both of them but even an old coot like me knows when it is time to move on!) While the older products are supported within the HA community there is very little direct support for the Squeezebox (except via xAP which itself has limited hardware support). The Squeezebox is such a great product that if it comes down to it vs. a lesser “automation supported†music system, I’ll choose the Squeezebox.

I have been playing around with Home Automation since about 1979. Back then I just had a few BSR X-10 modules and an Ultrsonic Command Console but I was already looking forward to the 21st century and what it would mean for home automation. I envisioned hundreds and hundreds and hundreds of hard wired inputs to tell me not only that the washer was running but what cycle it was on. To tell me not only that the Turkey was done cooking but how much electricity it took to cook it. To tell me not only that a toilet had been flushed but how much force had been used to press the handle…

Unfortunately after more than a quarter of a century the reality has fallen far short of my expectations (except for JohnWPB’s LCARS Display system which is the coolest thing I have ever seen!) So if anybody knows where I can hook up with others who are struggling with large scale de-automation projects (automators anonymous?) please send me a link.

 

[fitzpatri8]

You raise several excellent points. HA today is a lot like automobiles in the early days--difficult to use, prone to breakdown, and lacking supportive infrastructure.

I've seen encouraging signs lately on the reliability front, and the new protocols hold the promise of creating infrastructure that would extend to appliances that could communicate status and energy consumption data. I cringe every time I see someone say, "I need to install the ultimate HA setup, and I need to do it by Friday," because so many improvements are in play.

This is a great time to be watching, even helping, HA evolve. It's a terrible day to automate everything.

Real Soon Now.

Tom

 

[kwilcox]

hmmm... I'm just not understanding your issue, especially in light of the fact that you want to transition to an un-automated house. Does this mean that you will actually be replacing x10 switches with the old style stuff? I guess there must be controller issues that I'm not aware of that have become unsurmountable to you. Myself, I just ordered 2 more UPB switches and 2 more lamp modules which should be here today. I'll just install them and yet another part of my home will be transitioned to that technology. To me, the key is finding a controller that can do X10 as well as the new technology you are migrating to coupled with the fact that the new technology you choose has no possibility to directly interact with the old in terms of performance degredation. With those two requirements in place, the only issue becomes money. I guess that this in itself could be your sticking point if you truly believe that the investment can't result in a system with sufficient reliability/speed given any of the current technologies.

edit:

HCA supports your rainwise weather station btw as long as you use either WeatherDisplay or Virtual Weather Station with it. While not "direct support" in the strictest sense of the definition, its pretty darn close especially given the fact that you may already be using one of the above programs with your station. Here's a link you can follow for more information:

HCA version 6 weather support

HCA doesn't seem to get much consideration around here and for the life of me I don't understand why.... Without HCA, my own UPB migration would be quite the painful process.

 

[WayneW]

HCA doesn't seem to get much consideration around here and for the life of me I don't understand why.... Without HCA, my own UPB migration would be quite the painful process.

I don't think anybody here has bad feelings about HCA or its developers and I don't think there has been a single HCA bashing post. I think it is simply a numbers game. At the moment, there are more HomeSeer users than any other product, therefore more people mention it. CT is not HomeSeer specific.

Please continue to brag about your system and any advantages it has. Please invite all your HCA friends to join you here. Don't be afraid to be a minority! The same goes true for HAL, JDS, ECS, HomeVision, etc.

 

[Dean Roddey]

I envisioned hundreds and hundreds and hundreds of hard wired inputs to tell me not only that the washer was running but what cycle it was on. To tell me not only that the Turkey was done cooking but how much electricity it took to cook it. To tell me not only that a toilet had been flushed but how much force had been used to press the handle…

There has to be a business case to create such systems. Unfortunately, no one really wants to know that much about their house, or almost no one. A kitchen timer still works quite well and costs a few bucks. I know you are being a little facetious, but still the point is valid. Most people only will adopt automation to the extent that the utility is there for them. Lazy as we are today, most people will get up and walk across the house to turn off a light instead of paying $5000 to do it automagically (unfortunately for us).

There has to be enough utility that the average person will be willing to trade off whatever they could have gotten with that $5000 (or whatever) in order to get the automation system. And of course unless they are into it enough to do a lot of leg work themselves, it'll cost them considerably more than that due to the custom installation costs.

The only two markets for automation are geeks and the fairly wealthy, both of which are fractional markets. The only way you'll get beyond that, IMHO, is when some tract builders start putting in a standard system of components into a large number of homes, and companies like us can then target an automation package for these homes that will work out of the box, or with some very simple configuration, because we know what's there. It won't require a custom installer, or at worst a Circuit City level of compentency, because we'll already know what the configuration is and have pre-fab systems ready to go for them.

At that point, you will see some real penetration of automation into the hoi polloi, because it will be as inexpensive as it is possible to get it, becasue it was put in in bulk, one design was done and reused many times, and it will be a fully wired system (where it matters) so that it will be robust enough for the average person to not screw up (because it's all hidden away in the walls.)

BTW, if you are looking to do a serious DIY automation scenario, you should consider CQC. If we don't support some devices you need, if you will just lend us each of the bits for a little while we'll do the drivers for you at no cost. CQC provides a fully networked, multi-user automation architecture, plus very powerful graphical front end, and media management, all in one package, so no need to cobble together multiple systems to get both. It's extremely robust, because we are really looking to compete in the custom install market mostly.

 

[kwilcox]

HCA doesn't seem to get much consideration around here and for the life of me I don't understand why.... Without HCA, my own UPB migration would be quite the painful process.

I don't think anybody here has bad feelings about HCA or its developers and I don't think there has been a single HCA bashing post. I think it is simply a numbers game. At the moment, there are more HomeSeer users than any other product, therefore more people mention it. CT is not HomeSeer specific.

Please continue to brag about your system and any advantages it has. Please invite all your HCA friends to join you here. Don't be afraid to be a minority! The same goes true for HAL, JDS, ECS, HomeVision, etc.

I never said anybody here bashed HCA, just that it "doesn't seem to get much consideration". Sorry if my statement was confusing! Also thanks for helping me understand why. I had no idea that HS had a larger market share in the HA software arena.

 

[upstatemike]

This is a great time to be watching, even helping, HA evolve. It's a terrible day to automate everything.

Real Soon Now.

I know but the waiting is the hardest part. Still hoping to see it in my lifetime.

 

[upstatemike]

Does this mean that you will actually be replacing x10 switches with the old style stuff?

Yes but only when an X-10 device acts up and causes me grief. I'll use regular switches and maybe some commercial grade stand-alone motion sensors. They seem pretty trouble free.

With those two requirements in place, the only issue becomes money.

As issues go, that is a pretty big one. The time investment that is required to cut everything over is also hard to ignore.

HCA doesn't seem to get much consideration around here and for the life of me I don't understand why...

I did look at HCA but since my current system focuses mostly on hardwired inputs and voice status messages, I didn't see an easy migration path with HCA.

 

[upstatemike]

The only two markets for automation are geeks and the fairly wealthy, both of which are fractional markets.

Unfortunately you are correct but I don't think you can blame the customers. HA needs to apeal to both geeks and non-geeks alike. If the only thing you could buy to run on a PC was C++ or VB then only geeks would buy computers and computer makers would have the same complaint. But just as you can buy plug-and-play programs for PCs (in addition to the geek satisfying programmer stuff) we need more plug-and-play connectivity in the HA hardware and software world. The geeks should still have access to plug-ins and scripts etc, but when a mainstream consumer fires up their new HA controller software they want to see something like this:

The system has auto-detected 24 UPB devices in your house. Please enter the name you would like to use for each device. (Clicking on any device will toggle it on or off to help you identify which device you are labeling).

The system has auto-detected weather station data being transmitted at 14.8MHz. Please check the box if you want to make this data available to the home automation system.

The system has discovered a SlimServer music server on your network. This will automatically be incorporated into the home atomation system. No action required.

etc.

Once the setup gets this simple you will break out of the "geek only" fractional market and begin to hit the mainstream. I'd even be willing to give up the toilet handle telemetry option if I could have a system this easy!

 

[Dean Roddey]

I'm not so sure that the scenarios you paint are really optimal. More likely it will be "There's a hacker trying to break into your system, would you like to let him?".

Automation systems have to remain very stable or they are useless. Allowing for broad auto-magical entry of all kinds of devices into that system is probably not desirable. It probably would be best to continue to require proactive addition of devices to a system. That's not to say that the customer can't do it himself, since he could do that easily. But it should require an active step to do so, and it shouldn't be doable by the kids or anyone who happens to come into the house or get near the house.

Since most systems remain unchanged for long periods of time, the introduction of a complex auto-magical system for device addition really adds a lot of complexity and danger (both in terms of security and in stabiltiy), for minimal benefit, IMHO.

As I said, what has to happen to make automation available broadly is that it has to be built into the homes from the ground up, where a whole tract of houses uses a similar configuration. Then, every automation vendor can bid to provide a solution for that whole tract. The homeowner will walk into a home with a pre-fab automation system for lights, HVAC, security, etc... without having to do a thing.

This way, every single home isn't a unique (read expensive) designed automation solution, whole chunks of homes will use a common solution and therefore economies of scale can come into play, and the whole solution can be well worked out ahead of time, ready to go in in bulk.

But, in order for that to happen, home builders must be convinced that putting in the enabling wiring and infrastructure will be worth it to the buyer, i.e. that the buyer will be willing to pay extra for it. From what I've read in the various postings of custom installers who work with builders, they don't believe that.

 

[upstatemike]

Dean-

I admit my example of trying to look at things as "mainstream consumer" probably wasn't very practical but my point was just that HA needs to be more like a mainstream product and less like a geek toy. At the same time it needs to have enough "wow" factor to capture mainstream consumer interest. Folks who buy ipods and cell phones and laptops are not easily excited by an alarm system that also dims the lights and sets back the thermostat, or a media control interface that also dims the lights and sets back the thermostat, or a PDA that also dims the lights and sets back the thermostat. The home automation industry needs to define itself as more than dimming the lights and setting back the thermostat or nobody is ever going to pay any attention to it!

To your point about hitting an acceptable price point by using standardized configurations, I can only agree. A while back I started a thread in the Professionals forum called "Controllers Most Used By the Pros" which began to address that topic. I assume you are working to define what a standard tract home package might include?

I can see a lot of potential challenges when trying to sell something like this to home builders. They probably would like to offer an automation upgrade on custom homes or build the feature into tract home developments but will they really do enough to make it work? Will they put in extra deep boxes to accomodate automation switches? Will they really make sure that there is a neutral in every switch box? Will they use metal electrical boxes instead of plastic so ganged dimmers can dissipate their heat?

I worry that builders will take shortcuts and then blame the HA contractor when things don't work. (Assuming as you point out, you can even sell them on the value of doing HA in the first place.)

 

[Dean Roddey]

I assume you are working to define what a standard tract home package might include?

I'm really unqualified to do that, honestly. We are primarily a software company and we remain very hardware agnostic so that we can always give honest advice to our users about what we think is good and bad. We don't have that nitty-gritty custom installer's ground level experience, and therefore wouldn't want to presume to tell people what to install.

Whatever it is, as long as the products are well designed to be controlled, we are happy to do drivers for it and control it.

I do believe that our fundamentaly IP based, networked automation architecture would work well in such systems, which are almost certainly going to have a strong emphasis on the home network as we move forward, and we'd like to see more devices become IP enabled. Even if, to insure stability and security, two separate IP networks were run through the home (one for general use and one for the automation system), it still seems like it would be much better than pulling lots of cables to support point to point connections between the automation system and devices.

Wireless and powerline is a possibility, but I'd still take a wired system any day if I could get it. A 100Mb/s or Gigabit network dedicated to control, automation, and media distribution, combined with all IP based devices, would be a good compromise that keeps the automation system safe from outside influence, provides very good performance, is not succeptible to interference, and only requires one run to each room, to a small hub which branches out to various connection points in the room. Wireless can still serve in an ancillary role of course.

 

[upstatemike]

Again I agree with the idea of using IP as a standard "in-house" protocol to hook things together. The Slim Devices Squeezebox is a network attached device... any plans to write a driver for that? The software and interface specs can be downloaded free from Slimdevices.com and they even have a free "softsqueeze" squeezebox emulator so you don't have to borrow a music player to work with it.

 

[Dean Roddey]

Our next move in that direction (for the 1.5 release) will be adding support for UPnP, which allows us to interact with a number of media repositories and renderers. We feel like that will get us the most bang for the buck in the media area, since it's a protocol that's used by a fairly reasonable number of devices.

I didn't see the actual specs on how a control system would interact with the Slim system. Can you point me at that?

 

[Rupp]

upstatemike ,
What's amazing is I can replace every switch in my house with regular old flip switches for the cost of one UPB switch. Now there's a good reason not to automate. Oddly enough I have never bought a single filter and only needed a coupler repeater to get my X10 nearly 100% reliable. I know the only reason I got into HA was because it was affordable but how that has changed. It seems that all new users have no problems what so ever dropping $50 - $80 dollars on a single switch. I got started with a controller, HomeSeer, and 4 X10 modules for less than $80. The killer app is still to be developed for HA and the crowded market of new devices is more confusing than ever which only seems to confuse rather than solve problems. I may be joining you on the un-automation quest soon. It's just not as much fun as it once was.....