upstatemike

With no plugin modules I wonder how Ninsteon manages bridging across the two power legs? I suppose you could hope to get lucky and have two switches end up on opposite legs and be in RF range of each other to provide the coupling but that seems iffy at best.

I don't disagree. I'm just pointing out that there may be a workaround for the lack of an API if somebody really wanted to incorporate Ninsteon devices into their system. Without plugin modules or motion sensors I don't see much use for these products standing on their own. They just don't do anything particularly useful.

I wonder how much control these devices will support through a voice assistant? If Alexa can control it Ninsteon must allow her (it) to create devices to work with. If these devices are able to be put into Alexa routines than they are available to be manipulated via an ISY. I use this method to interface LoRa and other things that do not offer an official API and it works surprisingly well. I also notice there are no motion sensors, or sensors of any kind from Nokia. Apart from schedules or sunrise/sunset triggers what kind of automation can you actually do with this system? I guess we will know soon. "Shipping late September" means the preorders should start going out no later than Thursday!

I understand that thinking but still not 100% convinced because there just aren't any hardware controllers that include the majority of things I use Home Automation for. Lighting and audio and thermostats and cameras and motion detectors are all things that I am actually disengaging from my HA system to some extent because I can find stand alone solutions that do the same thing more simply. Motion sensing light switches, switches with built-in timers, programmable thermostats, and security cameras are all things that can be installed using stand alone products with only minor additional advantages if you incorporate them into a central controller. Things that really benefit me from Home Automation from my perspective are the integrations you can't find out of the box solutions for. Some examples from my system: Having the music mute in a room (and only that room) when the phone rings or is picked up to make a call or there is an intercom announcement or a critical system warning. Verbal alerts when significant unexpected events happen with key equipment like the generator, well pump, furnaces, etc. So I guess I don't need HA to do most remote control or automatic control functions so much as I need it to keep me apprised in a timely fashion of what is going on in areas I cannot see or to create new functionality that is not available in any existing product. If I were to go with something like Hubitat I would not use much of what they offer because it is not that useful while at the same time I would miss a lot of functionality that they do not support. Not picking on Hubitat as the same is true with pretty much any single box solution.

Not challenging anything you say but I will note my experience with respect to troubleshooting is the opposite. Having discrete modules makes it easier for me to track down problems bcause I can instantly see if a problem is only occurring with Z-Wave devices (for example) so it is a problem in the Z-Wave subsystem v.s. sometheing that affects multiple technologies and therefore a problem in the controller or another shared sytem. Your example of the 994 being reliable is true but the technologies a 994 can handle is so limited that without a Polisy it is really just an interface, not a controller. What I can do with just Insteon and Z-Wave alone represents such a small percentage of the total Home Automation picture as to be almost insignificant in the grand scheme of things.

I will tell you why I want that even if nobody agrees with me. I am not advocating this or criticizing other views and am open to being persueded differently but my thinking at this point is: * I tend to avoid single points of failure. This is not some crazy notion of theoretical situations that will likely never happen but rather the product of long experience where too much centralization has gotten me into bad situations in the past. * I don't have the financial or time resources to replace an entire HA system if it gets outdated. I need to be able to tackle upgrades incrementally. For example I can easily change from 500 series Z-Wave to 700 series by just changing the associated "dongles" without disturbing anything else. Sort of a "future proof" architecture. * I like to experiment with new stuff so it is much more practical to have a framework that lets me tack on something new without disturbing the rest of my "production" configuration. * I have an environment that is not RF friendly and requires widely distributed radios to work properly (I have 11 Wi-Fi Access Points for example). I know this scale is an uncommon situation but I am more comfortable following this principle than trying to blast everything from a single radio source. * Some technologies have hard limits (Hue) or best practices limits (Z-Wave) to being scaled out so a modular approach is optimal (or required). I like using network attached interfaces for Z-Wave or Zigbee and place them in the most optimal locations from an RF perspective. *There is (and never will be) a single hardware HA controller that does everything I want it to do so I don't see any way of avoiding the "dongle" architecture anyway. Having a controller with a few built-in interfaces like Z-Wave and Zigbee just seems silly since there are so many other things that need to connect to the system. If you have to connect to other interfaces anyway you might as well make that the standard architecture for all interfaces. * Sometimes different devices, protocols, or companies get discontinued or fade away or piss me off so I like having the freedom to excise them from my system without disrupting everything. * All companies have resource limitations and have to prioritize what they work on. An HA Controller manufacturer trying to be a jack of all trades but master of none is not going to produce a "best in class" logic engine or user interface. An architecture where the manufacturer can focus on those things and let external partner resources handle the expertise in each of the interfaces seems like the most practical path to getting the best overall system possible. * A corollary to the above is speed of innovation. If an "all in one" system has to maintain regular updates for every interface they support, how much time and resource will they have left to innovate? At what point does that interface maintenace become or exceed 100% of resources and things start to actually stagnate or degrade or their product become finacially unviable?

I guess I don't understand what the advantage is to having radios in a Home Automaion Controller. Much the way a better quality stereo will consist of discrete components or a better Loacal Area Network will have separate Router and Wireless Access Points and Switches, so too a quality Home Automation System will have a Controller and separate interfaces, HA Controller= Logic Engine, User Interfaces, Connectivy Channels. Examples=Polisy, Homeseer, Home Assistant. HA Interfaces provide the gateway between the HA Controller and individual HA devices via various HA Protocols. Example=ISY994+PLM, Homeseer Z-Net, Hubitat Hub, Various Lutron Bridges, Hue Bridge, Various Zigbee and Z-Wave sticks, MQTTT, IFTTT, Amazon Echo/Alexa, Google Home, Yolink Low Ra Hub, Elk M1 Security Panel, Brultech Energy Monitor, etc. etc. Some Interfaces are complex enough to do some rudimentary logic or UI but that doesn't make them HA Controllers. HA Devices and Sensors connect to an HA Interface and are at the level that has the most churn and frequent replacement. (Sort of disposable like cell phones) I just don't see any value in combining the Home Aoutomation Hardware layers any more than I would want an ISP combo router = switch = WiFi Radio or a Walmart all-in-one stereo unit.

Hubitat could be thought of as a good Zigbee interface to other systems. It could be used with some sort of Master Home Automation system if that system had something like a Node Server to communicate with it.

One hub to rule them all! I guess the Wink is like a hardware Node Server platform with separate Node Servers to talk to the native hubs for each individual protocol. Novel idea!

Looks like they are still around. https://www.wink.com/products/wink-hub-2/

https://www.lutron.com/TechnicalDocumentLibrary/Clear_Connect_Technology_whitepaper.pdf

The Smartlife app will let you code all the automations you need.

Caseta uses Lutron's proprietary "Clear Connect" protocol. Clear Connect Type A uses 434MHz. Clear Connect Type X uses 2.4GHz. I think Caseta uses type A.

The real future is Tuya Smartlife. If you use a voice assistant you are already dependent on the cloud anyway so you might as well get Tuya Smartlife compatible stuff and stop agonizing over all the competing technologies.

One restriction is that you have to identify what you will be replacing them with.

X10.com is still selling stuff... I guess they aren't affected by the parts shortages. X10 may return to being a market leader in Home Automation because they are the only ones with products to sell.

I don't know that I would be comfortable deploying Z-Wave, or any other routed mesh protocol, on the same scale that I have done for Insteon. A broadcast mesh with all devices repeating just seems a lot easier to maintain and troubleshoot. Routed mesh networks have not lived up to their promise of instantly self healing if there is a problem. They can take hours (or even days) to re-optimize themselves and sort themselves out after a device failure or other routing change. They also don't seem too smart about discovering better routes when new devices are added unless you manually force them to re-optimize. They will happily struggle along with a marginal connection even when a much better route exists unless there is a complete communication failure or some other event to force them to find a different path. They also are too chatty which means they don't scale that well and start to have more issues as your system gets bigger and there is more traffic to deal with. The most stable large scale Z-Wave installations seem to use multiple controllers with each device within 1 or 0 hops of the controller servicing it. Then there is the problem of adding a remote device in a Z-Wave installation. With Insteon you can add a device most anywhere but with Z-Wave you have to carefully build out the mesh until it can reach the remote location even if you don't actually need any devices in the intervening area. If the remote device is an outside lamp post with no place between it and the house to put a repeater then you are pretty much just out of luck.

Any chance HUE motion sensors and switches will be added in a future nodeserver update?

Unfortunately Home Automation systems are not yet as smart as subject matter experts at work so you can't just give them a mission and let them figure out the best way to accomplish it. It would be great if my Home Automation system could distinguish between an old person who is cold and could actually go into hypothermia in a room you or I might think is too warm vs a child or spouse who decides to kick up the heat because they just felt a draft. If I am informed I can mitigate those different situations. Alternatively I could try to incorporate sensors and programming to address every possible scenario but that is where it gets to be way too much work for me.

To me an automated home is like an aircraft with "fly by wire" controls and instrumentation. It is a ratio of 90% informing you of what is going on and 10% actually controlling anything. Even when you are cruising on autopilot you want to know what all systems are doing and you want to know immediately if they do something unexpected. If you do have to react and take manual control you want the response to be instantaneous just like you have a physical connection to the thing you are controlling.

Yes I want a voice confirmation announced, but I don't ask for it, it just happens immediately when there is a change. Usually I am not the one changing it but I hear what is going on and can monitor behavior. If you live with old people you know that small numbers on a display are sometimes not read correctly and the voice prompt provides a reality check for what they think they just did. Also if you live with old people you know it is not practical to simply have a program manage the setpoint and expect nobody will need to change it. Also I like voice interaction. Most of my downstairs rooms greet you when you enter them. So of course I want the thermostats to talk.

I disagree. I use voice responses to confirm manual changes to my thermostats. Any response latency at all means the person making the change has already left the area by the time the confirmation is announced. I would say the maximum acceptable delay for a thermostat response is closer to 2 or 3 seconds but I prefer much less.

I have some locations where I use Hue Motion Sensors that are not linked to any Hue lights to trigger announcements via Alexa routines. I now want to add some Insteon actions when the motion sensors are tripped via Polysi but the motion sensors don't show up in my ISY network... I just see Hue Lights and Hue Groups. Is there a way to use Hue motion sensors with Polisy?

I was using a single state variable for multiple motion detector inputs by having each sensor set the variable to a different value. This was used to have each room greet me as I enter. It worked as long as I did not move too fast between rooms but the enforced timeout prevented me from walking quickly from room to room and have it keep up. I went to dedicated state variables for each room and now it works perfectly. If things gets too chatty I use the suppress feature within the Alexa routine to dial it back.

I got this email this morning... I of course am NOT going to click on the link. I assume this is a scam of some sort? Universal Devices Your account has been deleted.All your personal details and any uploads have been permanently removed. If your account has been deleted by accident please contact us. Thank you! The Universal Devices Team