Is there a list of openhab cloudless hardware?

Hardware Home Automation
21

I’m sure with just a little bit of googling (I’d start with “isp6x API”) you can find out how it works. And if you don’t like how it works, switches like these are not that expensive, especially if you are willing to wait for the slow boat from China and flash a new firmware like Tasmota on it. Most of the cheap wifi relays and switches are EPS8266 based which means they an often be flashed to free you from their cloud.

We did here in CO too, but it didn’t make the news. Not as cold as you have but more snow than usual. 60 accidents by noon in Denver alone. And this is CO, not some place in the south that never sees snow.

So why not move to Zigbee and keep some of money you’ve already invested valid? Hue bulbs and devices are Zigbee and I believe the Zigbee binding will work with most if not all of them. Then you can expand to third party makers like Tadfri and the Xiaomi.

Xiaomi might be a good choice to look at. Not all of their devices are fully Zigbee compliant so you might be stuck with their hub. But I’ve not heard that the hub requires Internet. I could be wrong. But I do know they have temp, humidity, and soil moisture sensors.

MiFlora are a soil/temp/humidity sensor in one which I believe can be flashed or work without cloud out of the box. I know that openHABian ships with a small service that can receive readings from these devices. I think they are BT though so range might be a bit of an issue. But you can put an RPi 0 with the service in the greenhouse and bridge that to your main OH install over wifi.

But if you have a background in electronics anyway, why not DIY? The only real hard part these days is the case. Microcontrollers, especially the ESP8266s are dirt cheap as are the sensors and relays. With Tasmota and ESP Easy and MySenors you hardly even have to write code any more. Just wire it up, make the settings and go.

Depends on where you are. In extreme rural areas and very urban areas where they are trying to switch everyone to fiber it’s not so good. Any infrastructure like that requires upkeep and the telecoms are not spending enough to keep it up.

But personally, despite my career in cyber security, I’m an optimist and I refuse to live my life in fear of what might happen. I try not to over value the benefits of any new thing but I especially try not to over measure the likelihood of the negatives either. One of the great lessons I learned getting my masters degree in security engineering is how to calculate risk. Risk isn’t “if X goes wrong it will cause y worth of damage.” That’s only part of Risk. Risk is actually: Risk = Impact * Likelihood.

The impact of a meteorite hitting my house is devastating, let’s say $500,000. But the likelihood of it occurring is 0.00014285714% (1 in 700,000) chance of happening in a person’s lifetime so the risk over a lifetime is about $72. If we assume an 80 year life span that means $0.89 a year. So what do I do with that? What does that mean?

It means that to mitigate the risk of my house being hit by a meter I should not spend more than $0.89 per year to mitigate it. Spending any more would be a waist of money.

So if you take the same reasoning and apply it to the risks of your home automations, you are able to make better decisions about what is and is not reasonable and how much effort you should spend to mitigate the risk.

Having the dogs freeze because of a loss of internet is a pretty big impact. But what is the likelihood that the internet will go offline long enough where the dogs would be in danger? Assign some fake $ numbers to the impact (i.e. how much would you spend to save the life of the dogs) and look at the historic reliability of your internet and you will come up with how much it is reasonable to spend to mitigate the problem. I bet paying for a Sonoff and your time to flash it with Tasmota will come in under the cost, unless you have really reliable internet (I personally do have really reliable internet but everything here is buried underground).

Replacing a device with one that doesn’t require cloud isn’t the only mitigation. For example, I don’t have to worry about my house freezing if I lose Internet because my Nest will continue to function as a thermostat even without Internet. All the Internet gives me is the ability to control the thermostat from other devices (e.g. openHAB).

22

i will have to read your response about 5 times to take it all in. who needs college? i can learn more from generous smart people like you than any semester in college. my best friend lives in edwards, co and i hope to visit him there this summer. he got his masters at perdue and worked as a middleware mgr @ cummins for many years. it makes me realize how much i enjoy spending time with people WAY smarter than i am. i have what i consider to be a rare privilege to say that my 3 best friends are approximately 2x smarter than i am. people that i meet sometimes accuse me of being a genius, and i always think to myself… you haven’t met my friends :grinning:

so does cyber security always put a $ value on risk or an asset? in corporate america (where you would be paid for that skill), money is all that matters, but in real life most things have intrinsic value that can’t be put in a spreadsheet. i have repaired many MacBooks for customers that when i told them it would be cheaper to replace it than to fix it, they said “fix it”. they are bonded with a machine, lol, but i get it.

very little i think, but i would be devastated if any harm came to my wife, the dogs or my stuff because of my neglect to spend a few hours and / or a few dollars that could have protected them. they trust me to keep them safe, and i want to make a reasonable effort to be worthy of that trust.

again with elon musk… at the end of the interview with joe rogan, he said that the most important thing in the world is love, and he looks for ways to make the world a better place. om gosh, he’s good at that… he knows that we all think he is either an alien or a robot, but deep down he is the same as the rest of us.

every day i wake up with a goal of finding a balance between protecting the people and things i love while striving to create better things and more genuine relationships.

sorry for straying off topic.

23

The saying goes “If you are the smartest person in the room, you are in the wrong room.” I’m lucky to work for a company and have friends where I’m almost never the smartest person in the room.

It’s the simplest way to do the calculation. And for businesses it is the one thing that will matter. Even human life has a $ value (just ask an insurance adjuster). For those things that have other values, like sentimental value (pets, family photographs) consider the extremes you would take (in money) to prevent their destruction (Give me $X or the dog get’s it!).

Also, don’t forget that your time has value as well. So even if you are doing things yourself, your time adds up to a cost as well. Make sure to assign an appropriate value to your time as well when assessing the costs of mitigation.

Finally, and this one is the hardest, is dealing with lost opportunity costs. For example, the “no cloud services” rule comes with opportunity costs as it decreases the range of solutions that may work for you that do require a cloud.

It’s not that everyone needs to go through a full paper exercise to evaluate risk like this all the time. But when you are going to spend a lot of money on something, or spend a lot of time on something and that something is to address a threat, at least think about whether the cost and time involved makes sense.

Or the military and government which is where I mostly work. Loss of life is a real danger and some of the threats are state actors so the losses can be huge. But there has to be a way to normalize the risk and come up with a meaningful number. Money is a convenient unit to use because everyone intuitively understands what it means and it gives you an immediate maximum amount to spend to mitigate the risk.

In which case the monetary value of those Macbooks would be much greater than market value. Pets are a great example. For the most part pets are worthless in market value (unless they are babies). But many owners would spend thousands of dollars to keep them alive. So the value, in terms of risk, would be the maximum amount one is worth spending to keep the pet alive.

The money value is not intended to actually work with market values of things. It is a way to come up with how much impact would be realized should a threat become realized. I probably couldn’t sell my dog but if she got hit by a car it would probably cost me $3000 in vet bills. So the impact of that threat becoming realized is $3000. That doesn’t mean the dog is worth $3000 on the market. But it also means that the cost of vet insurance is right at the point of being worth it so we carry insurance on the dog.

For a less concrete example. Let’s say you are a company and you are worried about being involved in some sort of scandal or controversy. Essentially the risk is a loss of reputation. What’s the monetary value of a reputation? That’s really hard to measure. But what’s the impact? There could be boycotts and protests and a resultant loss of sales. So we would make some educated guesses about how much we would lose in sales and that becomes the $ value we work with.

And calculating risk in this way gives you a way to figure out whether you are being negligent or over zealous. It would also help you decide between multiple options.

And I don’t do this sort of thing for every decision. But for the big stuff (cars, pets, family members, house, privacy/identity theft risk, basically anything that can financially devastate) it provides clarity and rigor to the decision making process.

That is actually my fear. He is pushing himself and his companies beyond what is healthy. I do not have a lot of optimism for the long term health and life of Tesla. I laud his goals and am impressed with his accomplishments, but I can see he is already burning out and driving his workers to burn out as well.

It’s your thread. :wink:

24

… Great thread.
But back to the original question: It does seem to me that some kind of reference or tag for bindings to indicate cloud requirement would be helpful. I’m beginning a search for a thermostat; a simple, heat-only thermostat that simple detects the current temp and turns on a boiler if required. My aim is to not create my own thermostat but to buy a relatively plain one and have it report to / managed by openHAB. I’m finding that all the ones I’m interested in (Bosch = “Home Connect”, HoneyWell = “EvoHome”) have bindings that begin with creating an account on the manufacturers cloud. But I imagine there are bindings of ‘wifi’ devices that by-pass cloud requirements; they just aren’t obvious.
Then again, I’m at the beginning of what seems to be a pretty shallow learning curve (by that I mean a steep domain complexity curve)

As it stands I’m looking at Z-Wave t-stats feeling confident I wont be needing cloud accounts.

But Rich has mentioned flashing devices to meet needs and suddenly I’m wondering if jailbreaking a Bosch BCC50 is a possibility (or something it would be worth wading into).

25

Lists are great when they’re maintained regularly, but we don’t have people dedicated to that function. So, I personally think it’s best to have conversations about hardware that are effectively timestamped to ensure that no one’s picking up two-year-old advice and thinking it reflects the current reality.

Personally, I’d recommend going with Z-Wave as you know your thermometers will always be in your control. Even if a WiFi thermostat currently works without constant cloud access, the manufacturer could easily change that in the future. There’s no risk of a Z-Wave thermostat ever requiring cloud access.

If you search for “Z-Wave thermostat”, you’ll find lots of discussion on exactly what you want to do. There are some decent-looking devices that have appeared on the market over the past year.

Edit: Welcome to the forum! I forgot to say that at the start of my post. :wink:

26

One caution about Z-Wave is the devices are region-specific. There are different frequencies used and permitted.

1 Like
27

Very true.
Logitech tried doing that last year/s to the Harmony Hub. It worked without a cloud, but since it needed a cloud to setup and adjust settings it can be bricked if a CEO at a company has the idea to discontinue and switch off the cloud to force people to buy the newer product.
I always look for Opensource DIY whenever I can as often now days you can buy a ESP8266 for a few bucks and flash a ready made bin file and the project is done.

Openhab’s mantra is to stay vendor neutral and not to recommend which has its merits.

1 Like
28

Do not forget the non-cloud standards Z-Wave and Zigbee that do not need DIY flashing. They are designed for smart home devices and have much lower overhead than Wi-Fi, extending battery life.

29

By the way, there are catalogs of such hardware:

  1. Tasmota (this devices can be flashed by open-source firmware): https://templates.blakadder.com/

  2. ZigBee (this devices can be driven by open-source GW in zigbee network): https://zigbee.blakadder.com/index.html

30
  1. Z-Wave - Our current device database https://www.cd-jackson.com/index.php/zwave/zwave-device-database/zwave-device-list/