Stacey on IoT | Internet of Things news and analysis
Internet of Things
by 32 Comments
After two-and-a-half years of primarily using HomeKit, now called Apple Home, I think it’s time for a change in my smart home. Prior to settling on Apple’s platform, I used Google Home for a while. And before that, I was a Samsung SmartThings user. Moving forward, I think it’s time to transition to Home Assistant for several reasons.
I’m sure most people prefer a “boring” smart home platform. You know, one that just, or mostly, works to meet their needs. These platforms from Amazon, Apple, Google, and Samsung, are fairly stable and proven at this point. And any changes to them are generally incremental, with Matter support being a recent exception. But I don’t like boring. I prefer watching a platform evolve and grow. And watching Home Assistant grow is very interesting to me.
How could it not be as it also is integrating Matter support, while tweaking or adding features by the dozen every month. The July update for Home Assistant, for example, added the following interesting items:
I could go on as even more changes arrived in Home Assistant last month, but you get the idea. It seems like there are quality-of-life improvements each and every month. And again, aside from the benefits of those improvements, I like to watch the process unfold.
To be honest, since nearly every smart home ecosystem I’d consider using is or has added Matter support means I could easily choose to use any of them. But as I mentioned, Home Assistant is also supporting the Matter standard. I’ve only taken an early look at it and it is a work-in-progress. However, it’s less of a “closed box” than other implementations as I can view more of the data and actions my Matter devices are taking. It’s not perfect yet; then again, I’m not sure any Matter implementation is as the rollout is slow going.
Regardless, support for the few Matter devices I have means I have the freedom to make the switch to Home Assistant with little or no loss of functionality. So I won’t have to replace much, if any, of my current hardware when it comes to Matter. As for older devices? I haven’t run through the entire list, but I know most of them integrate with Home Assistant.
The one pain point I do foresee is with my video doorbell.
I use a Logitech Circle View Doorbell and it works so well with Apple Home. Even better, it fully encrypts my video feeds using HomeKit Secure Video. Not only will I lose that service but also that doorbell with the switch. Home Assistant works with some Logitech devices; my doorbell isn’t one of them. So I’ll have to replace that device with another one, preferably a model that lets me keep the video content locally stored rather than using the cloud.
Over the past few years, I’ve been migrating away from “big tech” products and services in favor of open-source software. As I’ve been learning more and more about programming, I appreciate and understand more of the code behind these products. MacOS and Windows have been gone from my life since 2020, for example, in favor of Linux. I migrated off of Gmail to Proton Mail around the same time. And even though I use a Chromebook for my other job, I use the Sidekick browser on it which keeps even more of my data out of Google’s hands.
I fully realize that most people don’t think like I do when it comes to the larger tech companies vs. open source. And that’s fine. People want smart home devices and services that make their life easier, often as it comes with the cost of giving up data.
I think you can have a great smart home experience without giving up all of that data or using a “big tech” brand to tie those devices together. Home Assistant, as well as Hubitat, are examples of this. And they attempt to rely less on the cloud with more emphasis on local control too.
Speaking of Hubitat, I did consider it going forward. However, I simply prefer the overall interface, feature set, and app experience that Home Assistant provides. And I appreciate the sheer amount of progress the platform makes on a regular basis. I should also note that I installed the Home Assistant application on my wife’s iPhone and… surprise: She loves it! She says she prefers it over the Apple Home app that we’ve been using for the last few years.
Additionally, moving to Hubitat would require the purchase of a new hub. I already have a few Raspberry Pi devices in my house, one of which has Home Assistant installed. There’s no additional expense for a hub to make this transition.
Again, I’ll reiterate that most people wouldn’t do this thought experiment and come up with Home Assistant as their answer. That’s the beauty of today’s smart home landscape: You have plenty of good choices that will suit your needs. There’s no universal “best” smart home ecosystem for everyone.
Given my requirements and my migration to transparent, open-source services, I think Home Assistant is the right platform for me at this time. I’m looking forward to sharing the experience as the journey begins!
Want the latest IoT news and analysis? Get my newsletter in your inbox every Friday.
Filed Under: Analysis, Featured
Enter your email address to receive notifications of new posts by email.
Kevin- I’m not at all sure you have to give up your existing smart lock. Because Home Assistant has integrations to work with HomeKit devices. There is an officially supported “HomeKit Device” integration, which allows HomeKit devices to work with Home Assistant. (Not to be confused with “HomeKit Bridge”, which goes the other way –allowing HomeKit to work with Home Assistant devices.)
Not every integration to Home Assistant provides full fidelity of all device capabilities when coming thru the integration, so you would need to explore which sensors and controls are available to Home Assistant from your smart lock.
Umm, I meant your doorbell, not smart lock. (But same applies to most any HomeKit device.)
You likely have to un-pair it from HomeKit before it’ll pair with Home Assistant. And if it’s Bluetooth based, you’ll probably want to setup one or more ESPHome Bluetooth Proxy devices.
A quick search shows that it will work with Home Assistant, but it doesn’t support secure video.
I looked At HA after coming from another platform but decided to go with Homeseer instead. Much better platform IMHO.
Home Assistant is broad and deep, and the professional staff at NabuCasa and thousands of volunteer contributors and content creators form an amazing ecosystem. My favorite feature is ESPHome which allows me, in literally a few minutes, to create a smart device from a ridiculously cheap ESP32 board and some discrete electronics.
I thought the same thing just a week ago, with so much flexibility and access to everything, what is there not to like?
I’m now tearing my hair out on a daily basis because an integration stopped communicating with devices, automations failing to run with no useful log messages, and just general instability that constantly forces me to restart the server.
I’m an open source developer myself, and I can say with absolute certainty that it is rarely a good idea to entrust mission critical things to it. When your family can’t control a smart home gone haywire because you’re asleep, there’s a serious problem.
I use my 20 years of experience to try to resolve the issues myself, and all I find are unresolved threads from years ago saying the exact same things. You start to realise if the problem has been going on that long, then it’s never going to be fixed. It rarely ever is in open source because people don’t want to work on the hard problems they have no interest in.
I’m not saying the commercial alternatives are any better. You sacrifice control, privacy, and constantly have to make compromises.. but at least they work without you having to do damage control on a daily basis.
I love the idea of Home Assistant. I just hate the lack of support and general instability that plagues the ecosystem with no hope it will be fixed. I want something that just works so my family isn’t terrorised by the home they live in.
I’m constantly starting and stopping with HA as I have a need to have spousal approval and that comes with consistent and reliable solutions. I feel like getting to the point that I have enough radios setup on my RPI I’ll get to HA as I agree, it’s the best tinkerer solution out there.
Home Assistant is the best there is in my opinion. I’ve been using it for several years at this point and you can’t beat the unbeatable depth of it compared to other platforms, it’s open source nature, and that virtually anything can be integrated into it. When I first installed it on one of my Pi’s, its very first purpose (and add-on) was to become a network-wide ad blocker (AdGuard Home)…it has since grown from just that to doing all manner of things, from controlling lighting, HVAC, outlets, energy monitoring, a mailbox sensor/notification system, a garage door monitor/notification system, reading/reporting from BLE temp/humidity sensors around my house, notifying me of calendar events and much more.
I think you made your expectations very clear, and I think Home Assistant should work great for you.
I will also let you know about the HomeKit Device integration that has a chance of working with your doorbell. I’m using it to talk locally to my MyQ HomeKit bridge that talks to my garage door opener, and my Budget Blind shades via there Pluse 2 hub. This done without the HomeKit hub.
Home assistant is not great for the average user, but if you like to tinker she spend a couple hours a week making things work, then it’s for you.
Now come take a listen to the home assistant podcast, to get some inspiration on how everyday people are using home assistant.
We’ve got over 5 years worth of episodes going through the releases and interviewing developers and users from around the world. Its amazing what ideas people come up with.
100% agree. I did the same and love it. No regrets
Different things work for different people. Choice is good. I’m glad you found something that works well for you. 🙂
One factor you didn’t mention: many people, particularly younger people, don’t have a PC/laptop these days unless they use one for gaming.
A platform which only needs a mobile device to set up and run is the minimum requirement for quite a few households. This was one of the big advantages that smartthings had when it started out. Apple Home, Alexa, and Google all offer home automation systems that only need a mobile device.
If you can’t imagine life without a PC/Mac that’s fine, and that does give you more candidates to choose from. But about 1/3 of Americans only have mobile devices, and that number goes up for people under 30. It’s also higher for people in many other countries.
In China, a large home automation market, less than half of households have a laptop. But almost 3/4 of individuals have smart phones.
Plug and play app-based home automation systems will remain the top choice for many consumers, and home assistant doesn’t compete in that category.
FWIW
Just to confirm here. Smartthings have the smartthings hub. Philips Hue have the Hue gateway. Xiaomi/Aqara in China have their own hub. HomeKit requires an Apple TV or iPad to always be at home (at least last time I checked). All of these replace the PC/laptop you mention.
Likewise, Home Assistant have their own hub offerings with the Blue and Yellow. So I think Home Assistant is pretty close to those categories too
That’s a good point: home assistant yellow, which is a new hardware offering for them, can be installed with just a mobile device if you get the pre-assembled model (assuming you can find one). But Kevin says in the article that he intends to use an existing raspberry pi, and that will require a computer to set up Home Assistant.
https://www.home-assistant.io/installation/raspberrypi/
And Home Assistant yellow isn’t cheap: by the time you get all the necessary pieces, it’s at least $175 and maybe a little more. And it doesn’t have WiFi, just Ethernet, Zigbee, and thread.
In contrast, the Samsung SmartThings station has WiFi/Bluetooth/Zigbee/thread and costs $80.
If you already have an iPhone, you can add thread, Bluetooth, and WiFi home automation devices with a HomePod mini smart speaker for $99. (if you want to use Zigbee devices, you’ll have to get ones that have a bridge like aqara, Tuya, and Hue.)
Amazon offers the Echo 4th generation smart speaker at the same $99 price, with WiFi/Bluetooth/Zigbee/Thread. And it’s frequently on sale for a bit less.
Google Nest Hub is another $99 smart speaker with thread, Bluetooth, and WiFi, this time with a display screen. No Zigbee, so again you’d need a bridge configuration for that.
All 5 of these intend to support Matter. And the 3 smart speakers also offer music streaming, podcasts, weather, timers, reminders, and other popular features.
So lots of choices for easy setup inexpensive home automation platforms these days, depending on which features are important to you. Sure, if you are technically skilled Home Assistant offers a lot of versatility, but that’s a very specific market.
I was also using HA for some time, but switched to Homey Pro when the new one just released.
It’s more of a sleek finish and got a bigger WAF.
It supports what I got so, so if you are in the same boat (and is where they sell it), I can only suggest taking a look at it here https://homey.app/en-gb . 🙂
The best way to use home assistant is to actually use multiple smart assistants together. I use homekit, Google home, and home assistant all together. Google home for the voice assistant, home kit for the store home app which is the best app so far for quick access, then home assistant is where all of my devices reside. The smart assistants just talk to HA to turn things off and on which I find is the best utilization imo. Now I don’t have to worry about Google or apple having support. As long as HA supports it they will and I can make them all talk to each other.
Welcome to the family!
Home assistant is powerful and lets you use the best devices regardless of standard. It does take more tinkering but you can do exactly what you want.
Home Assistant works with Home Kit devices so you shouldn’t need to give up your doorbell. Look at the Homekit Controller integration.
Kevin,
I know you like data, so one of the reports I’d be very interested in following as you work with home assistant is a time log for every week with the three following categories:
1) time spent exploring and adding new devices and automations.
2) time spent on regular maintenance
3) time spent identifying, researching, and fixing problems
As you may recall, I myself am quadriparetic with very limited hand function. So I have to pay someone else to do even simple tasks like popping a battery or moving a plug. For that reason, stability and reliability are very high on my personal priority list. This is a big change from the days before I got sick when I was a typical engineer who enjoyed the tinkering and troubleshooting aspects of cutting edge systems. But it does give me more sympathy for people who want “set and forget” systems because they have other demands on their time.
Habitat. I’m actually looking into starting a company that serves thoseb with less than great mobility. My friend is in a chair and it lead me to realize the complete deficiency in the world when it comes to being less than acrobatic.
Yes, HomeSeer is a far more advanced and requires less work to achieve the same functionality. You can use a wizard or dive into the details, all the way to scripting if that is your thing.
I used Homeseer 1.0 back in the day. Switched for Wink and iphone app support. Fell in love with Alexa support the moment I plugged my first echo in next to the wink hub. Migrated to Not-so-Smartthings. Then I migrated to a snowman of hubs to get my zigbee devices into Apple Homekit. I bought a Hubitat on black friday but can never seem to get over the sheer amount of custom drivers needed for many top brands of zigbee devices. Home Assistant is great, but have to pay a ransom like Wink to get the most of us.
I’ve been going back and forth to the detriment of my marriage between Hubitat and Home Assistant. I think my Home Assistant though is finally getting to the point where I may actually pay the kidnappers…
Matter is perfect in Apple Home, Home Assistant is getting close. My home now has smart clutter. The Matter device added to apple home then Home assistant for great automations resulting in multiple copies of the same device. Here’s my Eve energy added via matter to apple home, here it is added through Home Assistant’s Homekit bridge. Same issues with Alexa and Google.
The $5 a month is almost worth it to control which devices get exposed to voice assistants from HA.
The nice thing is if you put in some effort you don’t even have to pay for the cloud service in home assistant. It’s fairly easy to do without a subscription. I have been using home assistant for 5 years now. In the meanwhile I have around 3500 entities and over 7 million events and state changes recorded by the database in 5 days. There is absolutely nothing that can’t be integrated in home assistant with some effort I must confess
For your doorbell, you may still be able to get it to work by night required some hacking (big surprise). If you can connect to scryptd, scryptd can then post a homekit proxy that supports apple secure video. You then connect home assistant to one of the other streams on scryptd and voila, you have everything working.
Full disclosure, I haven’t EXACTLY gotten everything to work because I’m missing a key component (probably a certificate or something), but a lot of people on the internet have gotten it to work.
Another advantage of HA is that if will run on almost anything that can compute. Rpi, ODroid, any windows/mac/Linux box that can run a VM, a NAS, in a docker container, or on a cloud server.
I’m just starting to delve into the Smart Home world and would appreciate any suggestions on how to really move forward with it in my 125 yr old home.
@TimaReams
As far as advice on starting out with home automation, Different things will work for different people and there are many options out there, so I think you’ll probably get better feedback in a forum more suited to long and detailed conversations.
I’ve been quite active on home automation forums for about 10 years, have owned a number of different platforms, and previously worked for IBM as a field tech. But now I’m retired due to a profound physical disability which left me in a wheelchair with limited use of my hands. And that took me to a bunch of new connections with other people with similar issues. So all in all I’ve had literally hundreds of conversations with different people on getting started with Home Automation.
ARE YOU A TECHNOLOGIST OR A PROBLEMSOLVER?
In my experience, most people fall into one of two categories:
A) technologists. These are people like Kevin, who enjoy getting deep into the technical details of their systems, like being the first to try new products and features, and don’t mind spending an hour or more a week fiddling with the system. In fact, they usually enjoy that. They don’t typically have a strict money budget, and also, again like Kevin, often have stacks of equipment around the house that they either haven’t used yet or used to use, but haven’t gotten rid of. They are the ones who will come home from a sale at Home Depot or Best Buy with a trunk full of devices that they haven’t thought of a use for yet, but they will. 😉
b) problemsolvers. These are people who have a specific problem that they want to solve, whether it’s kids who never remember to turn the lights off, needing a smart lock to let the dog walker in, wanting the lights to come on in the house as you pull up in the driveway, or having a way to find the wallet or keys that they’re always misplacing. Or they might have a household member who is disabled and they want home automation to make that person’s life easier. They typically have some idea of a money budget and they’re often looking for “set and forget” systems because they have other things they want to spend their time on. They want reliable, predictable systems and think of home automation as similar to buying a new dishwasher or washing machine.
——
Of course, some people are a combination of these, some people are something different, but most people start out with one of these two primary perspectives. And which perspective you start from will that affect everything else in the kind of research you do and the system candidates you consider.
The age of your house is also a factor, and particularly whether the electrical systems have been updated in the last 15 or 20 years. Also, the building materials and type of construction can affect smart devices. If you have plaster walls with chicken wire inside or a lot of leaded colored glass, that can affect the kind of devices and how many of them you will need.
Also, the country you live in, will affect the device selection.
So…
My suggestion to people just getting started is to first decide if you’re a technologist or a problemsolver.
TECHNOLOGISTS
If you’re a technologist, you’re in luck, because most home automation blogs and websites are by and for people in this category. Lots of technical details, very little budget planning, lots of discussion of devices which haven’t even been released yet. Google “get started smart home” and you’ll find dozens of technologist discussion sites and articles. From there, you’ll pick a primary platform, and then just jump in one device at a time.
The Reddit Home Automation subcategory is a good place for beginning technologists to start, and Tom’s guide and the-ambient are good websites if you think you’ll want to eventually get deep into all the details but you don’t know the terminology yet. Also, Paul Hibbert’s YouTube channel is a can’t miss destination for technologists.
PROBLEMSOLVERS
If you’re a problemsolver, the research is more challenging because most of the people who like writing about these devices don’t like thinking about them from a “set and forget“ point of view.
My own suggestion for problemsolvers is that you start by coming up with a budget.
Define your budget looked at three different ways: initial outlay, replacement set aside, and per month cost. I have a budget of up to $500 per room and up to a total of $5000 for the whole house including the hub and voice assistants. I expect each individual device to last for three years. So after the initial Outlay, I have a replacement budget of about $140 a month. But that has to cover not only replacement Devices, But also paying other people to install them. (I am quadriparetic and cannot, for example, replace batteries in a sensor myself. But even the more able-bodied may need to pay professionals to do electrical work.)
Once you have a sense of your budget, List no more than three highest priority use cases that you want to solve in the first phase. (My highest priority use case was a totally handsfree smart lock with the ability to set temporary schedules for up to 15 people. Yours might be a smart thermostat or that welcome home lighting or voice controlled television.)
In my own case, I set a budget for my pilot phase of $1500 and assumed it would take about a year to try everything out and learn from that experience. I was prepared to throw out everything from the pilot and start over after that year. I know not everyone will want to work that way, but I learned so much just from those first three use cases, not just about available home automation options, but about how our own household would use home automation, and what our particular frustration points were, that it was really worth it.
(we are three adult housemates with lots of friends, family, and healthcare aides coming and going. But every household has its own challenges unless you are one person living by yourself and never have any guests. It can take time to figure out what will work best for your own situation.)
Now start your research by looking for discussions by people who have solved that particular problem and are writing about it from that perspective.
Of course, you can be flexible on all this: if your first solved problem goes great, and you start getting a lot of ideas for what else you’d like to do next and you find that you’re very happy with the initial platform you chose, there’s no reason you can’t speed up the next phase.
But if your overall perspective is one of problemsolver, working on a specific use case or two in the beginning with a set budget will help you save time, money, and frustration.
JMHO, of course!
(Btw, almost all the “get started“ articles out there will tell you to start by picking your voice assistant: Alexa, Google, or Siri. Or Bixby if you have a galaxy phone. But I think even that is a technologist question. Some people don’t want to use a voice assistant at all. Some people can’t. And a lot of people, including me, use multiple voice assistants. Almost all the devices I buy work with both Alexa and Siri. Let’s be honest: most of the home automation blogs get their money by referral fees when you buy a gadget that they featured. So they tend to talk about gadgets, first, rather than use cases. The sooner they get you clicking, the sooner they get paid.
But, yeah, OK… If you’re a problemsolver, and you have an iOS device, Apple Home is a good place to start. Shane Whatley has a great YouTube channel on setting up a Smart Home based on Apple HomeKit. The information is helpful, thorough, but also very practical for “set and forget“ people.
If you’re a problemsolver, you have an android phone, and you do want to use a voice assistant, then, sure, begin by choosing between Alexa and Google Assistant and then go from there. But again look for articles that talk about the problem being solved, not just the gadgets that are available.)
Sorry, didn’t mean that to all be bold. Dictation error.
This is not the proper place for such an in depth question. This is just an article that was published and people commenting on it. All these comments do is bolster the authors presence. I highly recommend you join one of the many active forums out on the web.
Habitat. Not the greatest interface, but it’s stable and just always works. I forget that my lights rely on it and they never fail
BTW, among my engineer/technologist friends one of the top reasons for switching to home assistant is street cred with other engineers/technologists. Seriously. They don’t want to be left out of all the “fun” conversations, they want to solve hard problems in hard ways, and they want to share those solutions with people who will appreciate them. (For some of them, that’s also the reason they became engineers to begin with.)
I recognize this because that’s exactly who I was before my illness and exactly the kind of platforms I was attracted to. Now, though, limited energy, limited ability, and a limited budget have pushed me firmly into the problemsolver camp. Give me simple, easy, and reliable, as long as it solves The problem I need solved.
So I don’t think either perspective is better than the other, and you may even shift between them as things in your own life change. Choice is good. 🙂
Your email address will not be published. Required fields are marked *
This site uses Akismet to reduce spam. Learn how your comment data is processed.
