Have we changed in the last 30 years with automating our homes? Non-interoperable technologies are still what they were in the 90s.
Alarm systems that have their own door and occupancy sensors and designed to notify monitoring centers.
Thermostats with room sensors that detect occupancy and they do it well for pre-heating/cooling and follow me temperatures to turn AC/Heat off and on.
Automation hubs which can turn lights on and off on doors, occupancy, etc. designed to be a platform to bring it all together.
Absolutely "minimal" integration where the vendor locks down their implementations and provides a small subset to interact with their platforms.
If I look at the marketplace I need a sensor on the wall for security, one for the thermostat for occupancy/temperature, and one for home automation. That is three sensors in each room or two sensors on doors, two water sensors one for ADT or SimpliSafe and one for SmartThings or Hubitat. Many only use their apps for notification other use IFTTT, others use SMS. The list goes on and on.
Until this sensor and platform lockdowns end in the marketplace we are never going to have a really solid integrated platform. I just received to communications from a major security company and a thermostat company that states that they are not going to re-open up their platforms or support back door connectivity. Thus making the platform closed for other use cases. Yes we can choose to go with only companies that support open platforms but the usually aren't the "best of breed" and more than likely the are in business to be bought out by Google or Amazon to be made part of their platforms and shutting out the rest of the world.
What can we do as home automation enthusiasts and professionals when it comes to these types of issues? Writing letters to companies don't seem to do anything. Creating another closed platform doesn't solve it. I am at wits end that this cannot be resolved were everyone gets a equitable solution.
Are we doomed when it comes to providing a simple solution for ourselves where functionality can't be leveraged across device manufactures and their implementation without jumping though huge barriers?
I don't think we'll ever see alarm systems fully integrated. Reliability and security concerns will limit that. And I think that's ok.
Everything else, however, should be interoperable.
Even with alarm systems, I'd like to see the system publish information (e.g., via MQTT) so other systems could consume the data and use the sensors for their own purposes. That would cut down on duplicate sensors to some extent.
Information going the other way, from non-alarm systems to alarm systems; I would not want to see. That's where security and reliability issues would prevail.
I think the answer is really dependant on how much effort you are willing to put into selecting the right hardware to achieve your home automation goals.
With the proper Hubitat hardware and automations there is no need no need for dedicated thermostat hardware to achieve your stated goals. The difference between the 2 is w/ Hubitat you’ld have to fiquire out and setup the logic yourself and with the (i assume ecobee) thermostat that part is done for you.
I dont see how it couldn’t be, to some extent adding Hubitat/Smartthings makes it more robust. Right now i use Smartthings presence senors to set Home/Away on my thermostat. I use Homekits presence detection to backup my presence sensors, thus making my thermostats presence detection ridiculously robust. I also have an automation that turns on my ceiling fans when the HVAC fan is on eliminating any hot/cold spots in my home making my HVAC system more robust. I have a 2 story home so i added smart vents and an automation to force more air downstairs when the heating is on so that is is equally comfortable up/downstairs. I let my thermostat handle its own scheduling because it does that well, most thermostats do. I have enough automations based around my thermostat that my power bill during a central TX summer is roughly $100 and there is no best of breed thermostat that can do all of that w/o some serious help.
Ultimately when you choose good gear and base your automations arround what your gear does best your system should be more robust with little to no duplication of hardware. It reallly sounds like you are over thinking it.
Well if you look at Ecobee for example, smart recovery, circulation, how would you program the house to be at temperature at wake up time? Ecobee builds a rules outside temps, home heating/cooling efficiencies along with weather forecasts. Then uses them to determine the best time to start heating/cooling for wake up or return to home.
You have done a lot of work to make your automation very controlled as I did when I had a StarGate and a controllable thermostat. I even built in a temp compensator at one time. If you invest in the hardware yep you can do almost anything as long as the software is also capable.
Well... I'll tell you that I think Ecobee's rules/algorithm is junk. But that is just one opinion, and not a statistical analysis of the system.
I LOWERED my electric bill the month I replaced Ecobee and just set the system the way I knew made sense - and was MORE comfortable. So in my case it failed on all fronts - energy and comfort. That was after being a rabid Ecobee fan and using it in multiple homes.
Now, I do have a pretty predictable schedule - so time based controls tend to work well for me. That might not be true for everyone.
But back to the original topic on sensors.... Different sensors are tuned for different use, and good at different things. There is no "one size fits all" sensor that can be used for all situations. Some are tuned for speed, some for detection area, some for reduction of false positives, some for pet immunity. Etc. Etc.
So I don't see it as "doomed", as there are few times I would use the same sensor for multiple needs anyway.
Further, it simply is not in a company's best interest to open up their platform/system to allow others to use their data and sensors. Why WOULD they? How do they benefit as a company? It definitely costs money for them to do so (testing, documentation, API development, support), so unless there is a return through higher sales it will never happen.
As you’ve clearly stated in your post the ecobee is very capable of scheduling itself and i would consider it to be one of the devices strengths... so that is exactly how i use mine w/o any extra sensors. I set Away (Temp), Home (Temp and time) and Sleep (Temp and time) on the device and thats about it. Then i use Hubitat and homekit to subliment the home/away status because they handle presence better which allows the temps to float when no one is home. With all the fan, curtain, window, and vent automations to help manage the comfort of my home i have no need to mangage my HVAC beyond that.
In my experience trying to micromanage my HVAC does not lead to more comfort or power reduction so I lightly manage it’s primary functions, so it can focus on doing it’s job and I support it with complementary devices (fan, curtain, presence, window, and vent controllers) so it can do its job as effectively as possible.
