The smart home is a crappy idea because appliances, switches and fixed lights are supposed to last for decades before needing to be replaced. Companies have clearly shown close to no interest in maintaining and upgrading even reasonably recent devices - adding smart features to previously dumb devices is nice at first, but what about in five years when the software will start to rot and stop working, and probably become a serious privacy and security liability?
Until this gets solved somehow smart features will always remain a marketing gimmick. There's no way around it.
People accept replacing ageing phones and computers because their connectivity features are a large part of the reason they exist. This isn't true for a light switch, for instance, which is mostly seen as a way to turn lights off.
Even the value proposition is unclear. A simple mechanical light switch Just Works (TM). It's cheap, reliable, and nearly universally understood. Adding network connectivity, scheduling, voice recognition, etc. to these things sounds cool, but that novelty wears off very quickly, and you're stuck with features that probably won't matter to you day-to-day.
It reminds me of those 3D user interfaces in the movie Minority Report. Very cool-looking and futuristic. Who wouldn't want that? The answer is anyone whose arms would be tired after a couple minutes of flailing around like that. Stop making things overly complicated for the sake of marketing copy.
I think there is value proposition in a smart home, but I believe every manufacturer has missed the mark on it.
My smart home is entirely app controlled (HomeAssistant), generally scheduled, and at no point does anything notify me unless an actual exceptional event happens. Every smart home device is controllable via a physical switch. My command and control centers are old android tablets acting as digital picture frames until you swipe on them.
And it's really convenient and it works well. No voice anything, no needing five different apps, and the UI is stupidly simple.
Most smart home devices that are advertised are a pain in the ass, the apps are intrusive and terrible, they are far too voice heavy and they usually don't play well with normal switches/remotes/etc.
I've got a similar setup that I started originally so I could turn off lamps in my apartment with mock lightswtches on the wall. This enabled me to have an otherwise normal experience since redoing the electrical wiring in an apartment is generally a good way to get in a lot of trouble. Being able to schedule or turn iff all the lights in a simgle action at night was just a huge plus
Thermostats, lights, entertainment center, cameras (kind of, working to update to a better system), presence via network awareness. Currently using a weather api as well.
Lights and entertainment center get the most usage of course, though thermostat control is nice. I do have humidity data from the thermostats but I'm not using it currently.
I built and did an alpha test of garden monitoring but haven't messed with it since.
I think richer control of lighting, including color and brightness, is useful, even if 99% of the time you interact with it via the wall switch. I had Philips Hue lights at my last house and enjoyed the level of control the app gave me. I used the simple scheduling capability it had and would have used more power if they offered it. (I did wish I had more control at the wall; I only had on/off switches and would have liked to have a dimmer and a button to toggle different programs.)
I can't say I found the ability to control my lights and thermostat from beyond the local network to be compelling. I changed the Nest thermostat for my mother once while she was staying there since she couldn't figure out how to do it, and I think that was the only time I found it useful. I never set up remote access for the lights and never missed it.
Remote security will be a useful application, though. Being able to buzz people in for repairs and deliveries and other services, and know that they know they can't steal anything without being seen, is an attractive possibility.
Fine-grained control of sound is another feature I would use. I often wish I could play audio in the whole house, just one room, or certain rooms, depending on the circumstances.
Also, something I find myself wishing for (less often since Covid, but still occasionally): cameras inside my refrigerator and pantry, so if I have a dinner idea while I'm out, I know what I need to pick up on the way home.
> (I did wish I had more control at the wall; I only had on/off switches and would have liked to have a dimmer and a button to toggle different programs.)
For what it's worth this is possible now with their Smart Button
> Even the value proposition is unclear. A simple mechanical light switch Just Works (TM). It's cheap, reliable, and nearly universally understood. Adding network connectivity, scheduling, voice recognition, etc. to these things sounds cool, but that novelty wears off very quickly, and you're stuck with features that probably won't matter to you day-to-day.
As I mentioned down below, I'm currently on crutches with a broken leg. Voice-controlled lighting is a big convenience in my scenario. Mechanical switches don't "Just Work (TM)" for disabled people.
Yeah I find that a couple of the 433Mhz remote-controlled sockets suit me very well.
I can lay in bed and turn on/off a fan, a pair of lamps, and the coffee-machine (which I pre-fill manually the night before).
More complex things would be nice, but I just don't trust most of them.
The only other thing I do is record the temperature in my living-room, balcony, and sauna. The sauna monitoring is useful because I can make it ping my phone when it reaches 80/90 degrees and is "ready".
(That just uses an off-the-shelf battery-powered temperature/humidity sensor. Every 30-60 seconds it broadcasts the current readings and I have an SDR-receiver plugged into my desktop. That logs the values to influxdb, and visualizes with grafana. Nothing amazing, but fun and a little useful.)
That’s not a strong value proposition though. There are many specialized devices for disabled people - doesn’t mean that they offer great value to the able-bodied. Eg a special watch for the blind is a great idea… but not something that would attract many non-blind people.
There is a value proposition, but the tech just isn't there yet. It requires too much manual setup, pieces are not very standardized, and they also don't work well.
Imagine asking your phone to watch a movie in your family room, then automatically, the lights dim or change color, the curtains are pulled if it's dark, the tv and receiver are flipped on and the movie is loaded and ready to go on pause. Many people would find value in this. This is all technically possible and some people have actually created this setup. But it probably took them several days and fiddling to make it work. I can barely get my Yamaha receiver and LG tv to work together and they are top of the line new components.
There is no value proposition. This tech is designed by software engineers for the purpose of harvesting the maximum amount of data about a a given person while giving him/her the thinnest possible veneer of value proposition. That's why mechanical wonders like the million-function wonder toilets of Japanese fame are not pushed - they do not allow companies to harvest as much data about its users, while taking actual engineering skills to design - a no-no proposition for these smart device companies.
Umm, I was speaking about myself as well when I talked about this setup. There is nothing that phones home here. Almond for voice recognition, Zwave for the switches, movie served by Jellyfin, TV and Receiver blocked by OPNSense firewall with VLANs.
It's valuable and useful to me, and if it could be setup up easily, I'm sure others would love it to.
Ironically I also have two japanese toilets in my house.
I'd argue that this isn't true. Installing a light switch is a HUGE pain in the ass compared to a "smart plug" that can turn a lamp on and off via your phone or any "smart" switch that can connect to it.
As annoying as setup procedures can be they don't take nearly as long or involve nearly as much sweat as cutting holes in walls, running wires, potentially installing additional circuit breakers, installing wall boxes, installing the actual switch, and making sure it's all according to code.
Furthermore, changing which outlet the wall switch controls is also a pain in the ass compared to just relocating the little smart plug or changing the device it controls in the switch configuration software.
It's possible to make a smart device that lasts for decades without security issues but device manufacturers just aren't willing to put in the effort for that sort of thing. It's much, much more profitable to make devices that only last 5 years before the end user needs to spend money on a new one.
This exactly. I had this routine set up but then updates and bugs made it difficult for others to use. Even now turning on the lights in one room mysteriously turns off the lights and closes the blinds in another room. There are no logs to help debug it.
Z-wave switches are the key. They don't have any "smarts" within them, they require a controller for that. So the switch itself is never outdated, and the controller can be continuously updated or even fully replaced without having to replace the switches themselves.
Being able to say "Hey Siri, good evening" to switch to my evening scene, or "Alexa, turn on the lights" to turn on the lights in whatever room I'm in becomes a huge convenience, not a novelty.
I use Z-Wave and the key for me is I'm not beholden to some external IoT service that is a potential privacy issue, can update firmware without my consent, or simply goes away. I feel I own my Z-Wave network (aka right to repair).
For light switches I use Z-Wave modules that use the wall switch as a low voltage switch to trigger the module to switch the A/C circuit. This way the existing switch acts just as it did before. My family really doesn't notice, yet somehow that porch light is always on around dusk and off by dawn.
I don't use Alexa or the like but as a developer I easily integrate with DiY equivalent.
That last point is probably a key gap generally though -- as a developer this home automation stuff comes easier to me, potentially.
Having a couple of Z-Wave devices, in my opinion they are not the solution.
Combability wildly varies between manufacturers, you have have to know what works with what.
The mesh capabilities are kinda wonky, where you have some kind of "heal" network function you use when devices for some reason stops talking to each other.
Electricians are not really pulling cable to put your switches in convenient locations anymore. They’ll place a smart switch wherever requires least effort, and then give you a remote panel for it wherever you like. The remote panel looks like a normal switch but actually sticks onto intact drywall, no cutting required.
Not quite “IOT” but the bridge between that low power network and the internet is a natural add on.
I confess slightly more complicated light switches would be convenient. We have some that have an LED in them when in the "off" position as an easy night light/indicator of where to turn lights on. That is quite nice.
Would also be cool if they were more "push toggle" style so that if I have two sets of switches on the same lights, I don't have to accept that four switches all in different "positions" are different states. (That is, Up Down Up Up, but all off.)
All of that said, most of that does not need any sort of "smart" additions. Basic circuitry has come a long way.
Can program double triple or more taps to run scenes or events, and can also sync multiple switches (wirelessly and on different circuits) to control the same lighting, or use cheaper (wired) remote switches that lack status lights, but are functionally identical.
I have single press resume previous light level, double up goes to 100% double down goes to preset dim level, triple up or down turns on/off the entire room/area of house.
They don't have wifi, unless you mean the z-wave protocol, which is a separate wireless protocol.
You can add a zwave controller that translates ip commands from a home server to zwave commands, but it can be wired or wifi (or usb).
You can also directly program zwave "scenes", which work directly between the z wave devices without needing a controller. Like a direct remote control.
It's proprietary wireless, usually people just say zwave.
My point was that it's not like most iot devices that try to find an open internet connection to phone home. They're 100% as isolated as you want them to be.
Check out Inovelli[0]. They are the darling child of the Home Automation subreddit. I have a few of the last generation switches and they do perform well and have a great feature set for the price.
As far as your "push toggle" concept, you could buy a Z-wave smart switch plus add-on switches and have this today. There's no requirement to actually connect to a Z-wave network, although smart switches are much more expensive than standard switches.
Right. Apologies, I meant to offer this as a small feature that comes with many of these "smart home" additions that actually is kind of nice. Just with the major caveat that the "smart home" aspect is... not really needed or wanted.
And with a huge confession that this isn't that big of a deal.
> It reminds me of those 3D user interfaces in the movie Minority Report. Very cool-looking and futuristic. Who wouldn't want that? The answer is anyone whose arms would be tired after a couple minutes of flailing around like that.
I'm sorry but mechanical light switches are certainly not as cheap as ZigBee alternatives. You need to have copper wires running from the switch to the light. If you want to move the button or add another button it takes hours of work. Working with electricity is even dangerous.
This has always been my general stance. It sounds cool but the juice doesn't seem worth the squeeze. It solves a problem that really isn't that big of a problem, IMO. It's almost a real life CINCO commercial.
One of the main problems as I see it is that dumb devices have a long life and when they fail, they tend to hard fail. It either works, or it doesn’t.
In contrast, smart devices have many soft failures. They work today and they don’t tomorrow. They fail for various reasons that are not under their direct control, and because of that they have a much larger mental overhead for me.
For instance, I can shut the garage door with a traditional switch and be pretty certain it will close. Yes, they do stop working, but it is a hard, catastrophic event that makes me take notice and act.
By contrast when a smart device fails, it can often fail silently and temporarily so I find myself constantly second guessing the thing I just did. “Did it work? Is the sensor showing me the correct state? Did it open on it’s own? I better go outside and check it.”
I find myself thinking about said thing way more during the day than I used to. And this is fatiguing. Yes, it has some added benefit to automate certain tasks, but any benefit is wasted by the overall mental overhead it is adding back on me. Definitely not worth it except for trivial things from my experience.
It would be nice for the building code to start dictating some interop standards for this stuff to overcome the natural tendency of the business to make this a winner-take-all market at the expense of the consumer. At first blush, it feels like it would be a little out of their wheelhouse, but it's not entirely unusual since it responds to consumer interests. For example, Energy Storage Systems (batteries) are a new-ish thing added to the code around 2014.
Codes abs standards have working groups composed of industry, consumers and inspectors. New groups are created all the time to address new technology. Unfortunately the initial cost to set up is typically too high for most individuals leaving progress in building codes/standards up to corporations.
Building codes are primarily for health and safety, secondarily to prevent getting screwed by contractors. Making rules on batteries makes sense but smart home interoperability less so.
Even if we assume that companies are willing to support the systems they sold for an undetermined amount of time (something that will _never_ happen), this still won't solve the fact that the hardware itself enabling the "smart" features is bound to turn obsolete way before the appliance exhausts its lifetime.
For instance, I've seen chest freezers running for >30y without breaking; do you expect today's IT technology to remain viable for the next 30 years? What about new protocols that might appear that the hardware may not support, or new highly desirable features that are simply unfeasible to implement on the limited µcontrollers that power most IoT appliances?
You risk ending up in a situation like with LPT printers - when USB became dominant, lots of old printers ended up in landfills due to how unreliable LPT-USB adapters were. The printers themselves were often absolutely fine, but they had become too annoying and keeping an old machine around to use them was much more complicated than just buying a spanking new network printer.
Given the current dire environmental situation I wouldn't push for technologies that might cause perfectly functioning appliances to become obsolete due to a non-necessary secondary features.
The only way out of this is for "IoT modules" to be removable, standardized and upgradeable, but thinking anyone would do that is delusional.
> but they had become too annoying and keeping an old machine around to use them was much more complicated than just buying a spanking new network printer.
To be fair the problem of unreliable USB to LPT adapters was caused by this. It was more economical to just replace printers so there wasn't enough demand for a quality converter.
If replacing legacy smart home devices is costly then it would provide a market for reliable adapter devices.
My mum kept trying to foist a wireless telephone on me, because then I could use the telephone in bed, or whatever. So this goes on for awhile, but the rechargeable batteries never seem to recharge properly, so the phone goes dead a short while into the conversation. I finally decided "screw all this", and got a cheap corded phone for a few quid. It Just Works (TM). We get the odd power cut, which means that my parents have to fish an old phone out of the drawer so that they can still use the telephone.
Our internet is intermittent, too.
People keep banging on about the environment, privacy concerns about big companies like Facebook are in the public eye, and there's plenty of stories of these smart devices stopping working. Yet people keep buying these overcomplicated, expensive devices with multiple failure points. I can't believe people can be that stupid not to see the obvious drawbacks.
L'il update: when I was a schoolboy doing my 'O' levels, one subject I studied was Technical Drawing. One day, the teacher showed us a pulley. It was used in industrial machinery. It was housed in a little cage which swivelled open. There was a "nipple" at the top to allow for lubrication. I marvelled about how much attention to detail someone had put into a humble pulley. Later on, I came to a strange realisation: there are actually two forms of engineering: industrial, and consumer. Industrial engineering is for stuff that actually has to work, and work properly. Consumer engineering: deliberately designed crap. It's the kind of thinking which leads one to welding batteries onto the phone itself. In some respects, I feel it's the opposite of engineering. Then I think back to how there was some guy, decades ago, that sat down and designed a pulley that was robust and could actually be maintained.
The answer will be a SaaS model to encourage ongoing development and maintenance. Get ready to pay $1.99/mo for up to 6 smart light switches or $6.99/mo for unlimited, for the rest of your life.
Correct me if I'm wrong here, but isn't the failure of smart homes attributed to the SaaS model itself. Nest devices had to be killed because Google wouldn't support the subscription anymore. What it needs is licensing model and not a subscription. Get a smart home installed on your home like you'd install Windows on a PC. We are basically wanting to install a server in every home to cut off recurring costs on vendors and even dependency on a single vendor for that matter.
The bottleneck here could be the network interface once you are away from your home. That is where standards struggle.
Looking at the sibling I currently see right below hits the nail on the head as in vendor lock in is bad for both SaaS and perpetual licensing models.
I think none of the models currently work well for the consumer. If your SaaS provider shuts down its service you have to replace all the hardware too because of vendor lock in.
If your perpetual license vendor stops making this produ t line or changes its proprietary software incompatible say 3 years down the line (which from a software development standpoint is reasonable actually) you can't buy another light switch for the new room you added by subdividing the basement "play room" that would work with your original installation. Software incompatible. New software or other vendors' software won't work on your 'ancient' hardware.
I want a smart home that is isolated. Given that i know all the ways a computer system can fail, I want my smart home to function even when Amazon/Google/Microsoft cloud goes down. Also , I dont want my smart devices to send or receive ANY data from external servers. Of course that's incompatible with current business models. So I prefer my home to stay dumb.
In principle, the answer is getting together and standardizing on generic protocols.
Unfortunately, there are multiple reasons why this can't happen in the current environment.
The business reason is that all the big tech giants are now deeply accustomed to lockin-as-a-model, and you can't have lockin if you have a standard that everyone can use equally. So, literally everyone in the space sees it as their incentive to let other players conform to their standards, but not to conform to anyone else's standards. This does not result in a standard, de facto or de jure.
(I made the same objection to the "metaverse" concept a couple of weeks ago. It can't work as some people are envisioning; the business proposition at this point is complete nonsense. Every participating company will be happy for other companies to voluntarily walk into their walled garden, but no company is going to walk into somebody else's walled garden.)
The technical reason is that any attempt to standardize at this point is more likely to produce a mess like bluetooth than any sort of coherent standard. Plus you get an echoing of the business model problem where every participant will see the standard as a weapon to be fired at their competitor rather than something to conform to for mutual advantage.
Not helping any of this is the problem that the problems that they are solving just aren't that compelling. This limits the amount of value they can extract from me by virtue of the fact I'm not paying for more than I'm getting from the 'smart home', which doesn't really leave them with much money to play with. It also doesn't help that the camel's nose is already under the tent in terms of not simply enabling the user but trying to control the user, with things like letting this authority twiddle with my air conditioner and that authority twiddle with my hot water heater, and in 2021, I have less than no confidence that buying into a smart home won't continue this trend.
Personally, my assessment is that the tech industry of the 2010s and the 2020s is structurally and temperamentally incapable of producing a high-quality smart home. The best they can do is create a locked-in walled garden that works great if you buy into it from top to bottom (and pay the subscription fees!), but gets abandoned in three to five years and then becomes totally nonfunctional as they pursue the next thing.
There are some standards already and a lot of the big companies use them already. Amazon's Alexa uses Zigbee to communicate with 3rd party devices and can act as a defacto hub. Zigbee is an open protocol with no licensing fees (or at least cheaper than Z-wave, it's slight more reliable, but more expensive, cousin protocol).
Many devices are now opting for control over HTTP which is fairly reliable, even if your outside internet connection goes down. It's also easy to setup and discover. There is a larger security concern for these types of devices, but common sense will go a long way towards keeping you protected until more rigorous standards can be agreed upon.
Home automation does not, and in my opinion, should not require and outgoing internet connection. My network is accessible from outside because I want it to be and I've accepted the potential risks (though tunneling into my own network is always an option should I want to go that route).
Bingo. One should aim for the smallest amount of complexity needed for the job. To turn lights on & off, a simple switch is all you need. If it breaks, you can go to Home Depot for another. If you're disabled, there are special devices you can buy.
Same for thermostats. If you're going away for two weeks in winter, turn the heat down to 55. What's so hard about that?
I've lost count of how many times I opened the windows and then went out without turning off the heating causing it to run at full blast trying to fight the winter outside.
A window sensor + smart thermostat + Home Assistant solved that problem and the system has been reliably running for 2 years now. I can now open & close the windows at will knowing that the heating will adapt around that and not waste energy. Note that there is no UI, no voice assistants, not even screens involved. It just works in the background.
Do I agree with the current gimmicky consumer-grade implementation of a "smart home" and controlling everything with your phone/voice assistant? No, that's stupid. But it doesn't mean the idea of a smart home should be dismissed - there are ways to actually make it useful and pleasant if you are willing to spend some time and/or money.
When I go away for two weeks in the winter, I'd also like my lights to turn on and off to simulate my home being occupied for security purposes. A simple switch isn't going to cut it.
I use simple smart switches for that. I can use my iPad to switch the lights on and off manually or they can be automated. It does require some “smart home” components but the switches are simple and cheap. i always hated trying to get the mechanical timer switches setup correctly.
I don’t go in for full integration of smart home features into appliances and such. Using smart switches gives me enough “smartness” without to much risk from a system in flux.
I think the Lutron Caseta system is an example of how to do this in a better way, though it doesn't fully solve the obsolescence problem you mention. It minimizes the size and cost of the potentially obsolete part.
The switches and dimmers themselves degrade gracefully--you'll always have a button to turn things up or down or on or off if the connection fails. And the connectivity is provided by a simple and responsive dedicated radio system, coupled to a bridge. You might need to replace the bridge in the future (to add support for a new protocol or whatever) but the dimmers should last (nearly) as long as the passive switches they replaced.
> adding smart features to previously dumb devices is nice at first, but what about in five years when the software will start to rot and stop working, and probably become a serious privacy and security liability?
There is easy solution - instead of IP-based IoT devices, separate appliance network based on industrial or specialized bus (e.g. Zigbee) and then gateway connected to IP-based network. With this setup all your appliances, switches and lights are pretty well isolated and with trivial software, and you just need to handle software updates and security in gateway.
There is another problem and that is interoperability and long-term availability of these products. I do not want to buy products from a closed system of one vendor, but prefer some industrial standard, so there is hope compatible products would be here even after 10 or 20 years.
> People accept replacing ageing phones and computers because their connectivity features are a large part of the reason they exist.
Though with phones, this is a relatively new concept. Phones used to last for decades. The phones in my parent's house were unchanged and still working after 30 years.
I don't have a landline any more, but I still have a beige rotary dial desk phone that my parents got no later than 70s. It still works, if you can find a place to jack in. It's just a quirky decoration in my living room now.
The article doesn't seem to consider security devices especially cameras and intrusion detection systems as smart home devices which have seen the largest adoption worldwide.
Households which have never heard about Alexa, Apple Home/Google Nest have IP Security cameras with robust motion detection/alert system and in many parts of the world they're credible deterrence as house owner having a gun is not a thing.
That doesn't mean security cameras don't have issues, most are already part of bot net, will never get a security update, hard-linked to the app(Even RTSP is unavailable) and of course privacy issues concerning always on devices in our home sold by companies which don't have reputation for respecting privacy.
But aforementioned issues can be addressed with hardened security cameras along with accessible self-hosted software which are now delegated to only those who wish to get their hands dirty.
What I want out of a "smart" switch is that it exposes an API over a network connection for what it can do (turn on, off and maybe dim the lights) along with a Unique Identifier of some kind. CPU and network stack all running on a Harvard Architecture system with no possibility of upgrade.
The "smartness" should come from a device (be it a smartphone, hub-like device like an echo or my router) that maintains the mapping between device addresses, api and friendly names. That can be upgraded, replaced, exposed to the outside internet.
> fixed lights are supposed to last for decades before needing to be replaced.
Only recently. Light bulbs used to blow rather too often and changing them was a routine thing. LED has changed that thank goodness.
Smart lights have been installed at my workplace. They change their setting over 20 or so seconds ‘to reduce stress’. When you press a button and nothing happens immediately it immediately irritates, and most users press it again (cancelling the change).
Who comes up with these crappy systems and the amateur psychology behind them?
This is why I prefer standalone smart furniture and gadgets. All stuff that doesn’t need to be built in the house, but can be updated and swapped at need. Also, every time I ask myself really hard if I need this thing or just want it.
What a house should have in any case, is a good network and Ethernet plugs in every room. This makes the addition of new gadgets much easier I found. I must admit I’ve not gone the WiFi mesh route, but it looks interesting as well. Guess I’m just old school on that.
I'm surprised Ethernet over Power doesn't seem to get more attention. It's basically Ethernet cabling without running any new wires. It was the only thing that fixed wifi availability throughout our townhouse.
I'd be interested to know how a wifi mesh performs in comparison, but I can't see how it could be nearly as good, without being much more conspicuous.
> and probably become a serious privacy and security liability
Given the quality of IoT devices, they pretty much become a serious privacy and security liability the moment they are first turned on.
Agree on everything else though. There's a fundamental mismatch between parts of a house where the expected lifetime is measured in human generations vs. tech gadgets where the lifetime is months to a couple years before becoming throwaway.
This right here. I completely agree, and I think there will need to be standards that can endure long before we see smarthomes take off in any real sense.
And I hate rolling devices. I bought a high-end smartphone, and a high-end laptop quite a while ago. I haven't really had to upgrade those, and that's really nice. Just a taste of what people want in their home appliances.
My washer and dryer are at least 30 years old. Parts are still available and any idiot with a screw driver can fix them because they are so simple. Family have tried to gift me their top-of-the-line slightly used models and I flatly refuse. Less is more for the vast majority of appliances.
Agreed. People look at homes and things like light switches and blinds that have stayed the same for decades. These things seem boring, but the fact that they haven't changed is evidence of great design.
Smart homes to me have always felt like tech looking for a problem to solve.
No company has successfully marketed an actual smart home system. Google Home and Amazon Alexa are, in reality, personal data collection systems that utilize convenience to get consumers to use them. The goal is not to help consumers, but to collect data.
And from an economical perspective, this makes sense. There's great money in gathering massive amounts of person data about people. Who would have imagined 15 years ago that we'd let Amazon and Google know about what time we get up and go to bed, what time we use our microwave, what sports games we watch, when we call our mothers.
I want something simpler, local-network-only. Not a 'smart' home but an obedient home. Ideally something like an Event Bus with a basic protocol for events and some safety controls to prevent chaos. But there's no economic incentive to build or sell such a thing.
> No company has successfully marketed an actual smart home system.
No company has successfully marketed an _affordable_ or middle-class smart home system. These systems do exist and have for years, and they largely work exactly as you described. The big three companies for such solutions are Crestron[0], Savant[1], and Control4[2]. It's been awhile since I've checked, but if I recall correctly they run on average anywhere from $5-$25k (or more depending on the size of your home and scope of automation). Much of the cost comes from all of the bespoke devices they build to automate existing products e.g. a proximity sensor that goes in your car to trigger presence automations, as well as the cost of the (very granular) white glove setup you get from the installer to prevent the possibility of the chaos you described.
Indeed, the post’s title and source are very misleading, pretty close to pure clickbait. The fully integrated smart home has been completely realized and functioning ever since Bill Gates built his ‘Xanadu’ home back in the late 90’s/early 2000’s.
The fact that at least 95% of that setup has filtered down to be affordable to the average multi millionaire showcases the incredible progress made in the last 2 decades. And this dynamic was evident at least a decade ago for anyone who bothered looking (e.g. me circa 2011 on a slow evening).
It's kind of a PITA to setup and maintain, but for technically inclined folks, I've found HomeAssistant to be a great tool for gluing together a lot of local-only devices into a semi-coherent interface.
I bought a house last year and the first thing I did was remove the Ring video doorbell. But if it hadn't been a company with an Amazon or Google type reputation I'd have at least tried using it for a while.
I would think the odds are much higher of a smaller company either a) selling your data to everyone that they can get money from, or b) just not having the tech skills to keep secure.
Note that I am not criticizing you not wanting to use it at all. Just curious that if it was another name, you would have been ok with it.
I think the point was that with Google or Amazon, you know what their real motive is. With another product, they might actually just be selling a security camera.
Maybe you're right that the odds that another company could be trusted are low enough not to be a distinction compared to companies I know I don't trust.
That said, I mind signing up for throwaway services less because it's easy to use a fake name/email, which is becoming less an option for the major services, who try to enforce real identities and revoke accounts without consequences.
I don't know about that. Amazon and Google have a decent track record of being bad about data privacy, and the size of a company by itself says nothing about that company's expertise.
I checked a while back, and there are at least a couple of smart video doorbells that can be made to work with a local IPTV system. I didn't end up doing it[0], but it does seem possible.
[0]- I wanted to monitor for deliveries, but COVID kind of trivialized that for a year.
In our apartment, we have the following smart features:
* Multiway light switches [0], which allow switching lights on and off from several different locations.
* A boiler, which heats water up automatically should it become too cold.
This covers lighting, heating, and supplies water at arbitrary temperatures between around 15 and 60 degrees celsius on demand, at several locations. The features do not require internet connectivity, are vendor-independent and have zero privacy issues. In the last 30 years since their installation, they have not once failed.
I am at a complete loss why this should not be smart enough. The technical overhead some people introduce into their lives just to avoid pressing a button to turn of the light is mindblowing to me.
> I am at a complete loss why this should not be smart enough. The technical overhead some people introduce into their lives just to avoid pressing a button to turn of the light is mindblowing to me.
My leg is currently broken and I'm on crutches. At this point I'm really glad that I installed voice-controlled lighting in my house. Good reminder that things that seem trivial are very much not to people with disabilities.
Impairments to mobility are a good topic for this discussion. The question is though, are the smart tech sufficiently better than a simple remote? --Best wishes in the healing process.
Not to say a dedicated remote isn't useful, but phones are really convenient.
You generally have your phone with you, so you don't need to keep track of a separate device or devices. Unlike a TV remote, which is used near your TV, a house remote needs to be used anywhere. Everyone in your household can have their own remote (phone), so there's no passing the device around.
The types of controls are varied, and a color touch screen is generally the best option. Light switches, dimmer lights, colored lights, fan controls, AC controls, etc. While I can imagine creating a remote that handles all these different devices, it's going to be a challenge to create hardware that services all these different devices intuitively.
> With kids, it has turned out to be an energy saver.
I wonder, though. Most folks now have LED bulbs, so being diligent about turning off lights isn't as much of a savings as when I grew up and everything was incandescent. Also, if you train kids that the lights go on and off on their own, are they more likely to just never turn anything off?
My teenage kids have friends over frequently, and while my kids have been reasonably well-trained to shut things off, close doors after they go through, etc., their friends are generally pretty bad stewards of my electricity. (And it's not malice; they just don't think about it.)
I like the idea of automating efficiency into the system, but there's always a human element. And that seems to require repetition.
> I wonder, though. Most folks now have LED bulbs, so being diligent about turning off lights isn't as much of a savings as when I grew up and everything was incandescent.
It's very convenient to have the light come on automatically in stairwells and lavatories. But you are right about it taking much longer to pay back. My calculation is that compared to having the LED bulb on continuously it pays back in about two years.
I have several infra red motion detector bulbs in stair wells and in the pantry as well as 'radar' bulbs in several other places (these detect movement by the disturbance in a low level high frequency radio signal).
> (these detect movement by the disturbance in a low level high frequency radio signal).
I'd never heard of this before. One of my former managers is really into Home Assistant and has done a lot with his setup, but I think all of his motion detection was IR. How's your 'radar' setup? Do you get a lot of false positives/negatives? How's the distance on its detection?
It is described as a radar sensor because it uses a high frequency oscillator and monitors the change in the signal due to things moving in the electromagnetic field. Imagine the effect of waving your hand near the aerial of a radio receiver when the signal is weak.
A 4 watt LED lightbulb left on constantly for a year would consume 52 kWh of energy, costing ~£7.50 at UK prices, releasing ~9.5 kg of CO2, the equivalent of driving 23.5 miles in the average US passenger vehicle.
There are good reasons for turning off lights in the house (getting a good night's sleep is one!) but saving energy is no longer one of them.
There are a lot of light bulbs, though. The LED switch has had meaningful impact on US residential energy usage [0]. IEA reports 38B in LED sales in 2015-2020 [1]. Averaging $10 (high) per bulb, that's 3.8 billion bulbs (including non-residential).
Leaving a light on doesn't matter much. Leaving billions on can clearly have an impact.
The billions of bulbs have the same proportional impact on the national energy budget as the few in your home have on yours, that is practically nothing.
Our new house came with a Nest thermostat. It has some perks. Probably the main one is that we can turn the thermostat down to just warm enough to keep the pipes from freezing when we will be away from the home for a while, and then remotely turn it back up so that the house reaches a comfortable temperature just as we're arriving back home.
However, I don't think that that's enough to counteract the many, many UX annoyances we've encountered with it. When it craps out or Google decides to brick it, I intend to replace it with a regular old programmable thermostat.
I admit to being mystified by the value proposition of Nest thermostats. I have a simple programmable thermostat that I can put on a set schedule (and change if I'm going out of town and want to set everything for energy savings) and the incremental value of having something more dynamic just seems miniscule.
If you only have one thermostat getting a "smart" one doesn't really add much unless you live alone and want to take advantage of the presence features (it knows when you're home) in order to save some money.
However, if you have more than one thermostat (e.g. separate downstairs and upstairs air conditioning/heating) having a "smart" thermostat is awesome! I can change the temperature on both from my phone without having to go upstairs.
We also installed the additional sensors in various rooms in the house so I can control which zone gets used in determining the whole-house temperature (there's often several degrees variation between rooms in my house on hot summer days). The thermostats in my home were installed in terrible locations that don't reflect the temperature in places where people actually are so the remote sensors are fantastic.
I installed a Nest so I could add a remote temperature sensor in my bedroom. I configured the nest to heat based on the bedroom temperature at night and the living room (where the thermostat was) during the day.
Before the Nest, the bedroom would get way too hot at night because it was better insulated than the living room, making it difficult to sleep. Afterwards this was no longer a problem. Saved a bunch of energy too.
Other "smart" tech: remote controlled outlets, later swapped for hue bulbs, let me flip all four of my living room lamps on from the entry way OR the hallway. Before, there was no switch at the entry way leaving me fumbling through the dark for the wall switch at the other side of the room to turn the ceiling light on. Then I'd have to turn the floor/table lamps on/off individually.
I'm sympathetic to your overall point, but I'm impressed that you've had zero boiler failures in 30 years. I'm pretty sure that's better than the average. (Mine is about 10 years old and has needed one fairly minor part replacement to fix a failure in that time. The figure that seems to be commonly quoted for boiler expected lifespan in the UK is 15 years.)
I don't have too much "smart home", but I am really tempted by lighting. Being able to dim and change the color temperature of the light depending on the time of day is compelling. Dimming is easy conceptually, but it would mean rewiring large portions of my home. I don't know how I'd manage that with color temperature.
I like 2700K. Wife likes 5000K. Having bulbs that can go back and forth has saved me thousands of dollars by preventing a divorce.
Seriously, though: Controlling multiple bulbs at once is very convenient. Sometimes I sleep downstairs, sometimes upstairs. The lights I want on/off differ in both cases. So a simple routine "Sleep downstairs" and "sleep upstairs" that controls which lights should be on/off is more convenient than getting up and doing them individually. It was mere fun when I programmed it, but once I had it, it was awesome.
Also, having some lights come on at sunset was a real improvement. Gone are the days when I come home during winter to a totally dark house. Granted - you don't need Internet to achieve this.
AC compressors should not operate below 60F outside. Sometimes it's still warm when I sleep but will dip below 60 near dawn. Because of this I can't turn on my window AC all night. I have a smart switch that can handle the wattage of my AC. My next project: Turn it on when I go to sleep, and have it turn off when the outside temperature is below 60F - and turn it back on when it goes above 60F a few hours later.
Philips makes a range of "Warm Glow" LED dumb bulbs that change from ~2700k to 2200k as they dim.
Not nearly the color range (or control ability) you can achieve with something like Hue bulbs, but I've been happy with them for bed side lamps and the like.
You can also put Hue bulbs in and use something like HomeAssistant with the F.lux plugin to control the brightness + color temperature automatically throughout the day. No wall switch-based dimming, but dimmable with a phone.
I'm a big fan of the Hue lights. I've done the IKEA ones, and they are great, but the phone-connected options offer scheduling and automation, if you're into that kind of thing.
I have some led strips installed with a simple remote that allows temp and dim control...I love it and don't think auto changing these with the day would be that big of an improvement than pushing the button a few times until its at where I like it.
For some reason, last year I tried to rederive the circuit diagram for 3 way switches. I failed. It's actually a very clever idea, and higher numbers are fun to think about independently.
The switch way numbering is based on the number of terminals.
A 2-way switch has two terminals and gives a closed (on) or open circuit.
A 3-way switch has three terminals and gives you one terminal tied to either, but not both of the other two.
A 4-way switch has four termimals and gives you two terminals which are each tied to the other two terminals and you can pick; these also give you a big headache when one switch of many fails and one circuit path is open, but many paths still work.
Or in other words if you're used to seeing switches from the electronics side of the world:
2 way: SPST
3 way: SPDT
4 way: DPDT
I once lived in a house with a 5(!) gang box with three of the switches wired to the upstairs hall light incorrectly and another switch upstairs. It was as big a mess as you'd imagine.
There are electronic three-way switches that have either a single button or an up/down rocker that always reacts the same way (up for on, down for off).
These are available in "smart"/connected form or regular electronics-only form. Example here:
There's two way, three way, and four way, and four way can support arbitrary numbers of "ways" - because a path is always found on one flip of the switch and not on the other.
No, GP is responding to the 'multiway switching' link in the top-level comment.
This is about where, in the UK and I'm sure most developed jurisdictions, stairs for example must have a both on and off control at both ends - regardless of the state of the other switch.
If they were simply in parallel as you describe, you could find yourself in the situation of going upstairs to turn it 'off up there', (because it'd need to be 'off everywhere' to be off) then realising it was already 'off downstairs' and falling back down.
If they were simply in series, you could find yourself unable to turn the light on without first going downstairs at speed in the dark, because it's 'off there' no matter what you do upstairs.
The key thing is that it's a double-pole, double-throw switch. You can put any number of them in series, and the circuit essentially counts the parity of the switches, so flipping any of them toggles the state of the lights.
That's what a 3-way or 4-way (should be called X-way) switch setup is - three or four or more switches that no matter what the state of the light is, flipping ANY switch will change the state.
> a path is always found on one flip of the switch
I thought you were taking the comment you replied to to be seeking a simple parallel arrangement of SPST switches, so that turning any one on would turn the light on.
So I was explaining that No, point is you want 'off' on any switch (regardless of the state of the others) too. Seems you understood that though. :)
yes... A way to look at 2 way switches is to look at the connections of usual circuit setups to Boolean logic. Taking two on-off switches,
- parallel is or
- series is and
We can see that a 2 way switch is xor (or xnor depending on the setup).
So now, for 3 way it feels like you have to beyond 2 state (on-off) switches... i don't have a proof of impossibility that 2 state switches cannot do 3 way switching, but in practice they indeed involve more states (multi-throw etc.)
The closest you can get with '2 state' (by which you mean SPST - single pole single throw; SPDT is the standard way to achieve this but is usually still '2 state', it's just that 'off' (if you like) has a terminal too) switches is to add a 'don't care' state, and accept that first deliberate ('do care') switch wins.
To do this you have a live rail and a switched live rail, and two SPST each end: one switch SL hot or not; the other to break its connection to the light. Open the latter for 'off', close both for 'on', close break & open SL for 'don't care'. (Then in the 'don't care' state, SL is hot iff switched on the other end, and if so is also closed through to the light.)
Not legal, not (that I can imagine) desirable!
Proving it is actually simple if you continue your Boolean line of thought: you can't build XOR without an inverter; an SPST switch is only a sort of 'half inverter', a mux between '0' and its input. Since it can never give you '1' from '0' input, you need a schematic for your XOR where each constituent truth table in isolation has '0' out for its '0;0' in line. Which can't be done, because clearly then you could never get '0;0=>1' overall.
These circuits don't run on digital logic, but are instead based on physically connecting and disconnecting an electrical connection between the live and neutral wires. This means you have to work out how to use the two switches to produce a connected circuit when only one but not both switch is in the up position.
Digital logic works on lower voltage than mains voltage, and you'd need to maintain far more wires to hook up the connection (since the actual gates themselves need both power supply and ground drain wires which are separate for the inputs). Possibly also a relay to actually drive the higher voltage/amperage for the light bulb.
I've really liked my RainMachine sprinkler controller. It only requires internet connectivity to download weather reports; all other functions are available via LAN. Being able to turn the sprinklers on via the phone is incredibly useful for verifying repairs, finding leaks/clogs and aiming spray patterns. And it reduces water usage by deferring watering if enough rainfall has occurred.
Why do you need internet connectivity for any of that?
My sprinkler system has a sponge hooked up to the roof. If the sensor detects the sponge has too much moisture it doesn't turn the sprinklers on, because that means the soil is still saturated from recent rain.
A forecast driven watering algorithm can look for rain in the immediate future, rather than the current state of the ground. It can also change the duration of watering based on temperature and relative humidity. RainMachine does that while only consuming weather data. You also have the option of turning that off and just using wireless control of the sprinklers, which as the GP says is extremely handy.
How long is watering suppressed by the sponge? Does the controller "remember" that the sponge triggered a few days ago when the watering day comes around?
It has a permanent sensor on there, so the sensor doesn't remove the signal to not turn the automatic sprinklers on until you press the manual bypass button, or until the sensor no longer detects moisture above the set threshold.
When the controller hits the scheduled time it checks with the sensor unit, if the sensor says there's too much moisture it doesn't turn them on.
I wonder if that actually works that well. Especially with black roofing tiles which get quite hot. Compared to the hopefully shaded by the plants themselves soil. If your plants are sometimes shaded by trees depending on time of day the sponge will also probably fail you because your roof is more likely not to be shaded at all in many places for fear of overhanging tree limbs or roots being a danger to the house over the long term. Though it would be good for saving on cooling costs/energy use but then the sponge stays too wet vs. the soil.
I'd go for a soil moisture probe. That checks whether the plants really need the water, not whether a sponge with a "different reality" than the plants is dry.
Weather report data is cool though. Though you can't trust the weather report only, you can nicely combine the data from a soil probe (a "multi zone" system with a probe per zone would be great, probably easy to program with a Pi) with expected rain but an override via the soil probe if it actually didn't rain.
Since the OP said sprinkler I'm sort of thinking lawn though which I think should just never be watered at all ever. Waste of water just to look nice. And yes our town actually prohibits watering of lawns in summer. And for plants drip irrigation is better than sprinklers but hey, if you have a sprinkler system already go for it.
Caveat: I don't live in a fire hazard zone so take this with a grain of salt.
Wouldn't it be better not to plant any grass if it's that bad where you are? For soil erosion planting something else sounds more adequate.
I do see a problem with septic fields. Can't plant anything else that I know of on those but grass. If you do know of something let me know and I'll plant it on mine instantly lol.
Why would you not plant a garden? I could maybe see not wanting to plant root vegetables, but peas, strawberries, cabbage, etc surely present no issues whatsoever. (And TBH, root vegetables should be just as fine if the septic tank is doing it’s job.)
It's not entirely clear (some people/studies say it's fine, others say it's not) whether certain pathogens can end up in your vegetables this way and/or which vegetables would be fine and which wouldn't. I mean, some people do compost their own #2 and put it on their veggies and while it's been OK so far for many, I personally stay on the safer side of these things until desperate times call for such measures.
Anything that has a root system that is able to reach the septic field's drain lines is potentially able to clog the holes. E.g. I tried raspberries in a planter on top of it for a while but even though I tried, I couldn't contain the roots and/or raspberries fall down and the seedlings grow in the grass. I ended up removing the whole thing just in case as septic drain field repairs are either expensive or a lot of work.
Evaporation apparently plays some part in the effectiveness of the drain field (it doesn't just all drain out into the ground). Though I wonder, why plants taking up the water and evaporating it via their leaves wouldn't do the same thing, potentially even better through a larger surface area, so I guess it's mainly the first two points.
Well, anecdotally, I ate from a septic-fed garden for more than a decade, to no ill effect, from potatoes and carrots, to peas and broccoli. I wouldn’t hesitate to do it again. (But, yes, one does need to consider roots clogging the pipes; annuals perhaps safer than perennials, which surely drive roots deeper.j
Clover is a popular alternative. Some homes are xeriscaped (no lawn). But converting an existing lawn is expensive, even with subsidies. Selecting native grasses and watering less is an affordable compromise.
> I wonder if that actually works that well. Especially with black roofing tiles which get quite hot.
FWIW the sponge isn't literally on the roof. It's on a handle that's connected to the edge of the roof and holds it probably 4 or so inches horizontally away from the roof.
Sprinkler systems have for years included a rain gauge that trips a signal if there is water in the tube and turns off the lawn watering. The smartphone controls would be convenient but only for very specific situations and not 99% of the time you're using the sprinkler system.
I was too cheap to buy a flash system and made one with a tp-link smart plug and a solenoid. It can be activated by the mostly-ok tp-link app, but is also automated in Home Assistant. It turns on for 10 minutes in summer.
By wiring the solenoid correctly, it fails to an off state. It’s basic as all hell and works really well.
Well first of all I think there's a large portion of home automation users that are tech people and just enjoy the process more than the results.
There are some generally good use cases though.
My most useful automation feature currently is that I can turn the air conditioner on and off in my attic office without having to walk up two flights of stairs or even be home. Sure I could leave the ac on a preset temp, but that would be a huge power waste.
Then there's home security features, which are really all just a subset of home automation. The only low tech solution to this problem is basically to move, but that isn't exactly practical.
> The technical overhead some people introduce into their lives just to avoid pressing a button to turn of the light is mindblowing to me.
I use a "smart" socket to switch off most electronic devices in my home for the night (and I tend to hide cables so it's not easy to do manually). And because it's automatic I don't have to remember switching it off and then on in the morning.
Yeah, when I think about smart homes, I think about the possibilities of materials and structure and moisture management and clean-air management and noise abatement, not gimmicks. American houses are surprisingly poorly designed and built given what humanity knows about the discipline.
The motivation is probably due to Star Trek. Who wouldn't want swishy Star Trek doors? Ironically they were opened by hand from behind the scenes, at least in the first shows.
Also, 'Computer, locate my husband.' 'Your husband is in the attic'. etc.
I think the most miraculous invention of the Star Trek universe is a code of ethics that people actually followed. They didn't have the endless power hungry narcissists that are endemic to every human society. It's the least probable aspect of the series even including warp drives and teleporters. Maybe they all died in World War 3?
Yes, although I think the absence of those power hungry narcissists (who no doubt lurk back at Star Fleet HQ) would be more down to having a relatively isolated and small group of people living and working in close proximity. Dark triad individuals can be identified early and worked around or placed where they cannot harm. An explicit code of ethics would be an Achilles heel because any such a code can be hacked or subverted.
For general consumers, I would agree. For geeks, I love my smart home with Home Assistant (thanks devs!). I was reluctant at first and thought I'd just dip my toe in to see what it could do. Now I'm regretting not diving right in.
I love using a browser to manage Schlage keycodes instead of tediously punching numbers on each keypad. I love that my kitchen lights all come on at once instead of having to flip 5 switches all over the room. I love that I can turn on the outside lights when we're returning home late and even unlock the door as we're pulling into the driveway, so the little one can make it to the bathroom on time. I love that my outdoor lights react to both time and outdoor light conditions.
I recently bought and installed an IotaWatt, an open h/w and s/w home power monitoring system. I haven't integrated it into HA yet but I guess I only care about having a dashboard widget.
Everything is local and self-hosted. I don't have to worry about cloud rot, suspended accounts, or Internet outages.
PS, if anyone knows how to control a Mitsuibishi hyperheat heatpump through HA, I'd love to know more. It's the big missing link in the system right now.
Check out https://github.com/SwiCago/HeatPump (and derived projects) for your heat pump. I was just looking at this yesterday - unfortunately my units don't have the CN105/CN92 connector so I'm out of luck but if your unit is newer you should be fine!
I have yet to buy a "smart" device that I haven't been able to control through HA. The Broadlink RF/IR bridges even makes anything with a remote automatable.
Seriously, If it wasn't for Home Assistant, the "smartness" of smart devices would be next to useless.
My smart home strategy is pretty simple: nothing that requires internet connection to work. Only local devices. WiFi where possible over anything low power (BT, Zigbee, ZWave), wired over wireless wherever possible (e.g 1-wire). Policy for anything I include is that it must fail like an escalator not an elevator - a dumb switch for example, not a light that can’t be switched on. This means smart lightswitches, not smart lights.
It’s a lot of work to configure and maintain but tools like home assistant, proxmox and docker at least makes it somewhat manageable. It’s still a pain though, and it feels like a special kind of masochism to tinker with.
I think it’s entirely possible that an average user who isn’t a programmer or willing to pull a neutral wire here and there could have a decent “smart home” set up by themselves some time within the next three or four decades…
I'm seriously considering getting rid of most of my smart home devices not because they suck but because the software to control it sucks. The Samsung Smart Things app used to be so responsive, I could turn devices on and off without much delay. Now when I open up the app I have to restart it wait for it to connect, then wait for it to find every device, then 20 seconds later I can push a button that does something. That's unacceptable.
When I started my home automation, I wanted as little frustration as possible. I knew that you just can't rely on companies, so I ensured everything is controlled in-house.
Use ZigBee. ZigBee devices create their own mesh network, so Wi-Fi can be down and things still work. ZigBee does use 2.4 GHz, so you may need to move your router to use a different frequency.
Use a Hubitat Elevation or other dedicated server with ZigBee and a UPS power backup. Again, this means the automation logic is independent of the network. Wi-Fi and/or Internet can go down and home automation continues to work. The Hubitat has community driven package management. Once set up, it's a breeze to perform updates.
If you still want to use Alexa, Google and Siri, use bridge software to the services. I use Home Bridge to give access to my Hubitat to HomeKit. Home Bridge in docker means upgrading is letting Watchtower do its thing. Home Bridge rebuilds it's plugins every start, so restarting the docker container brings everything up-to-date.
The most unreliable part has been Google. I have Nest thermostats, and Google cancels the Home Bridge login token now and then. HomeKit integration is excellent, but I wish Siri wasn't so daft at times recognizing speech.
Several of my friends that used to be on the SmartThings bandwagon have migrated to self-hosted Home Assistant and are much happier. I haven't made the jump myself yet, but plan to soon.
While I understand not everyone has time to tinker with it, to me Home-Assistant is really the only way to do smart homes in a way that brings value because of its flexibility with a real programmable interface.
For example, I was able to code an action to start my espresso machine (which has a 20 minutes preheating time) by hooking it at the "first look at phone" event in the day. To me this is obviously something I would never agree to share with a company selling ads, but since it never leaves the premises it's pretty cool.
Automating the coffee machine is amazing - it was my first foray into Home Automation. It saves time, it works great and can also be done before you get home at the weekends.
The just-audible click at 6am followed by the sound of the boiler heating is very satisfying, even 5+ years after setting it up.
If you’re willing to take the time to set it up, openhab is actually pretty great.
I’m using it here to manage a bunch of stuff, with absolute reliability. All of our stuff here is fail-stupid, so if any of the automation breaks, there failure mode is the least damaging one possible, and there always a manual override.
We use it for:
- PV, wind, generator, future hydro monitoring, power regulation, routing (water heaters come one when the batteries are nearly full, air rads come on when batteries are full and water is at 85.
- Water management. Sensors in tanks, and river, tell us relevant levels. Pumps run when they need to to keep things topped up, but also based on power availability.
- Agriculture. Got moisture and light sensors in each bed, valve on irrigation triggered by soil and sun conditions. Weather station at the house provides additional parameters. Going to add nutrients to the system in the future.
- Lighting. Right now it’s just doing dimmers/LRGB for the home theatre - still can’t quite get it syncing up LRGB for ambient lighting based on display but working on it. Also, currently rigging up fairy lamps in the trees along the road to trigger based on presence.
I don’t think many people make a decision to “smartify” their home at any one point in time. The reality is that it creeps in over time. If you are shopping for a mid to high end anything today - lighting to a dishwasher - some smart features are likely going to be standard on many models. People get these things home, probably try and like some of the smart features. But inevitably over time you discover how poorly integrated they all are with one another, you become frustrated when the smart features start chronically disconnecting, and suddenly it’s all more trouble than it’s worth.
The only way out of this that I can see is the passage of time and steady iteration by device manufacturers, and real investment by a few platforms (eg Apple with HomeKit) to give us a decent front end and quasi home iPaaS. I worry they don’t have the incentive to keep investing.
I couldn't agree more. There is no magic feature that is going to make me want a smart home. The whole concept sounds like a nightmare on so many levels.
A lot of off-the-shelf consumer smart home products do suffer from the drawbacks you've listed, but they do not generally apply to the smart home concept. My home is automated using components that use the KNX standard, and although the initial investment is higher than a conventional electrical installation, I'd argue maintenance is actually simpler and cheaper – as pretty much all of it is in the software.
Additionally, there's quite a few benefits:
- It's all wired, so it's been rock solid;
- It's vendor agnostic;
- It's entirely local, so there's no subscriptions or even an internet connection dependency;
- There's a ton of gateways available that allow me to interface with non-KNX products;
Overall, I'm pretty happy with it. It is just a matter of convenience however, it's hardly a must-have.
I'm building a house at the moment and will go the KNX route. It really does seem great, it's just pretty much unknown outside of Germany for residential use.
The way communication works is each device has a number of inputs and outputs that have defined data types (there are one hundred or so in total). When programming devices you just tell them which channel (group address) the inputs and outputs (group objects) should be linked to. Inputs can be linked to any number of channels, but outputs can only be linked to 1. Some devices have basic logic units if you want to do more complicated things. The group address is 16 bits, so there's lots of room for flexibility.
For example, if you have a switch and the on/off output is linked to channel A, when you switch it on, it would send a 'switch on' message on that channel. If you have a light actuator (in simple terms, a programmable relay) that listens on channel A, when it receives that message it would turn the light on. If you want to add another switch at the other end of the hallway, you only need to program that switch to send the on/off output to channel A - the other devices don't need to be reprogrammed.
Devices are certified (which is one reason why they are quite expensive compared to consumer smart home devices), but that means they are guaranteed to work together regardless of the manufacturer. The protocol itself hasn't really changed over time (there are a few minor new features which are not backwards compatible) so if you have a installation from the early 00s, you can buy new hardware today and know it'll work without issues.
The main thing I want is to change the colour temperature and brightness of lights throughout the house based on the time of day. I live in Northern Europe, and find lighting has a noticeable effect on my mood in winter. I'll also have presence and motion sensors throughout the house, so won't need to touch a light switch.
So basically first world problems. And for sure you could try to do something similar with Philips Hue bulbs (I have in the apartment I live in now), and 80% of the time that works, but I'd like to solve the other 20% too.
I'll also have a heat pump for hot water, heating and cooling and solar PV, so I will tie those together to try to optimise my power usage to use as much solar as possible. I'd love to be able to do that with appliances too (e.g. load up the washing machine in the morning, and have it turn on when there is excess solar), but it seems that's a few years away yet. Now I can monitor power usage and know when the washing is done. And if nobody goes in the room for an hour (from motion sensor) I can send a push notification to remind someone to get the washing.
if it's full offline with no subscriptions or internet connectivity - I'm up for some.
If it needs any kind of connectivity or it comes as a service - no way in hell I'm exposing my family to the risk of not being able to flush a toiled one day because of a tweet or god know what.
1) People don't understand and don't want to have to understand how WiFi, servers, clients, protocols, GUIDs, etc work so they can up a standalone system. Manufacturers create systems requiring internet connectivity so they can abstract away some of the complexity from the end users. Unfortunately, that means when stuff fails, it's totally opaque, as described in TFA. I mean, shit, most people couldn't wire a lamp correctly or change a tire [citation needed, but you get my point].
2) Rent seeking and data mining. Why let the customer pay for the lightbulb once and use it forever when you can make them pay for a monthly subscription and sell the usage data? And then you can force an upgrade when you deprecate the service before the light bulb burns out. Win/win!
Remember: when somebody says "internet of things" what you should hear is "billions on unpatched, unmaintained Linux devices sold by companies whose core competencies aren't securing computers racing each other to the bottom".
to be entirely honest I see the IOT part as a blessing and a curse.
I can see what potential benefits IOT devices could bring but the risks of all these outdated devices in my home scare the hell out of me. I know it sounds like paranoid talk but my whole life is digital these days.
Okay, but neither of those problems are actually an issue with smart homes as a concept, they're mostly issues with the way profit motive drives behavior in a capitalist system.
The first point is confusing because the majority of tools people use today they can't repair themselves. That's what happens as technology advances, you need specialists to understand each system.
If you look at things like earthships that are built to manage a lot more factors than a typical home, it becomes pretty clear how useful automating some of those systems and allowing them to share data could be.
If you had good open standards for the tech and were actually trying to make things that improved people's lives or saved energy rather than mine data or recurring revenue, you could make some cool shit.
Really the complaint should be about the way these products are developed and marketed, which is hardly unique to this sector.
The problem with this is that it seems like you are now on the hook for long term maintenance of some hacky system you threw together.
I think if you view this is a hobby and a chance to play with electronics then that is probably a great use case. But if your goal is to minimize the amount of time you are spending on this stuff, I'm very skeptical that whatever system you are throwing together on your own is really a good way to do that.
There is a reason that so many of the systems out there suck. Building good ones is really hard, takes a ton of research, R&D, and collaboration between many experts. I fail to see how removing all of that and throwing something together yourself really creates a reliable system
> If it needs any kind of connectivity or it comes as a service - no way in hell I'm exposing my family to the risk of not being able to flush a toilet one day because of a tweet or god know what.
That sounds like the plot to a Black Mirror episode
Yeah, I consider "smart home" to be one of those complexity feith driven developments. Switches are simple, understandable, minimally-abstracted interfaces that almost never fail. Why would I want to add a several computers and some software to that? That just brings a whole bunch of new failure modes to the table for a miniscule gain in convenience.
It's funny how first item on your list of drawbacks (higher maintenance) was already there with first-wave of home automation, back in last century [0]. Current iteration not only fails to solve that problem, it also brings 3 additional problems.
Yeah the 'better' it gets the more layers of code, abstractions, and networked system are required to work in perfect harmony for my home to not start creating problems in my life.
>There is no magic feature that is going to make me want a smart home.
Imagine the fun when it all goes obsolete in five years, a controller board dies with no replacement available, all for very little advantage.
Maybe this is Peak Technology. Sticking a computer and communications network on things that gain little from the addition. The one exception I might make is low-cost surveillance cameras, but of course it took the industry about ten seconds to add subscriptions and spying-on-you to the business model.
It's bad enough that LED lights exploded in variety and change early and often. I had one of ten, 20-year lifespan ceiling flood lamps go dark, and of course the replacements don't exist because these particular lamps were only made during a 1-2 year window. To add insult to injury the fault is not in the actual light emitting element but (of course) in the driver.
> the fault is not in the actual light emitting element but (of course) in the driver....
Being over-driven.
Big Clive loves to tear down LED bulbs to examine the driver boards. Often shows you ways to drop the current to sustainable levels and get back your 20-year lifespan LED bulbs (caveat, at lower power levels).
What's really sad is that there's no need for any smart home concept to result in a loss of privacy. However people are cheap, and its difficult to impossible for somebody selling a self-hosting smart home appliance that you just plug in to your wall to outcompete somebody who is practically giving away the hardware to make the money on your data or on subscriptions over time.
The author is right the, most current home automation products suck.
The stuff from big brand names is poorly supported, pricey and designed to suck your money away from you.
The cheaper Chinese stuff requires the use of a cloud server located somewhere I don't know and it comes with god knows how many security holes. Plus, the build quality is ropey at best, I clearly remember that time a desk lamp controller in my children's room literally melted.
There are a few devices on the market that don't rely on cloud servers. I have a few plug controllers that you can control via WebSockets or a REST API. I've been dabbling with them for a while and they seem pretty decent. It's just a pity there isn't any off the shelf home automation software that I could use to control them.
I think this is a combination of various weaknesses acting in discord (rather than in concert) to create pure chaos.
1) Making the devices Internet-Must devices. Not only does this mean some dumb failure modes when your connection goes down -- and it will -- but temperature adjustments might be made based on far-away weather stations, processed through layers of averages and aggregators, instead of drawing from local sensors.
2) Not thinking about security (conflicts massively with #1). Security isn't as daunting if you are not hooked up to the Internet. It can still be an issue, but opening up your attack surface to "the world" only makes things worse.
3) Proprietary, and therefore incompatible standards. If I want X, but I already have Y, I hope to hell that YCorp has an X-like offering in their works.
4) Being more about data collection (see also Samsung's SmartTVs exfiltrating data, or Amazon Walk ... or maybe now Apple is dipping their toes into that with their CSAM "won't somebody think of the children?" business)
5) The general disposablity of today's electronics. This also plays into standards because it'll just burn out so why be backwards-compatible? Just make a new standard. And you're back to #3.
My wife and I were constantly warring over whether our living/dining/kitchen window blinds should be open or close, and there were too many and the faux wood was too heavy to manipulate so much. To make things easy, we got Lutron motorized cellulars shades so we could just ask Siri to open or close whatever. Totally worth it.
Since we had the shades, we also did our lights in that room, so now we can quickly put down all the shades and turn on all the lights.
That being said, I can’t see doing it for any other room in our house. The bother factor just isn’t there like it was with our main open living space. Well, maybe the bedroom, we don’t open the shades very much even though I’d like to.
I never wanted a smart home. I don't even want "smart" appliances. I want energy-efficient, reliable appliances and homes, with behavior I can predict, that don't spy on me, or rely on remote APIs, and have high usability factors. I don't mind remotes, and voice activation might be nice if I was disabled, but for the most part I'm happiest when my home and I don't get in each other's way.
I think that the smart home is failing/will fail because people have a limit on how much they are willing to be tracked. Smart devices from TVs, to cameras to speakers to vacuums are reporting details of our lives that most people just don't want to give up.
It used to be the case that people feared that the government could watch them through their TV or listen in via their phone. Now, those things seem like they are product features.
Anecdotally: I don't have any smart home stuff, but literally every other member of my family does (parents, siblings, inlaws, etc). I've had numerous conversations in which they ask me why we accept an Echo or a Nest or something as a Christmas present, but my privacy concerns are always greeted with a blank stare. They just don't see it as an issue.
To them, the (perceived?) issue is how hard everything seems to setup and maintain. I haven't tried connecting any smart stuff lately, so maybe it's really easy for all I know, but all that stuff seems really advanced to them and they seem kind of afraid of it.
I doubt it. Most people literally don't care if Amazon is recording all their conversations. Even more people don't care as long as nobody at Amazon is actually listening.
Smart homes aren't super popular because of integration issues imo.
I feel like the biggest problem I've had with smart home technology is that there is no out of the box standard for controlling or having access to all of your devices. Almost every device seems to require interacting with it with the vendor's own app or other utility, and so I've needed to install a multitude of at least 3 apps just to control the lights. So not only are vendors trying to compete with their smart devices, but additionally with the apps they've created to even use them, probably to see how much more money they can squeeze out of selling your personal information.
Thankfully, there's Home Assistant[1], which is an amazing project that aims to bridge connectivity and control between all of these devices into a single centralized environment. Pair that up with Tasmota[2]-compatible devices and you can have a smart home that can be completely disconnected from the cloud and in your control.
I am a bit disappointed that no one has mentioned a big problem with all these machines: security. Most of these items were developed by companies with only limited experience in security. They offer a tempting attack surface for attackers. Once they are compromised they can be used to attack other devices in your network. Most of them run proprietary software, so it is not possible to verify its code. They require that the network is correctly configured which is not at all easy. There have already been a lot of incidents where surveillance cameras were exposed on the internet.
Even if they do have security experience (i.e Google), I would expect that for many vendors security is WAY, WAY down the list. And security doesn't end when they ship the device, it also must be present throughout all updates, even when the device is no longer supported by the vendor.
To this day I am baffled by the smart lighbulbs. So you take a LED light, strap a bulky AC/DC convertor to it, then also glue in a WiFi module, and also some microcontroller for managing all that stuff. Then you cram all that into E27 fitting and have problems with heat dissipation if your LED is actually about as bright as an old 100W lamp was: the capacitors give out after a couple of months of work so you have to buy a new lightbulb (and those things ain't cheap because of all of that additional stuff it has inside it).
Obviously, a sane way to do things would be to have a single microcontroller/WiFi module mounted near/inside any of traditional manual light switches, have a single AC/DC convertor near your circuit breaker box, and have it power the whole lighting wiring (from which the microcontroller would draw too) with 9V DC, and now the LEDs inside the lightbulbs will have enough space to cool properly even with size-reduced fittings.
But where is money in doing things the sensible way, right?
>But where is money in doing things the sensible way, right?
the non-cynical explanation is that the lightbulb socket is the form factor that is most easily used by the consumer. Some sort of setup involving changing the lightswitch or the bulb socket might be better, but would also require a bunch of work and risk (live wires!) to install.
Yes, cheap crap is crap. But the problem with even non-cheap 1500 lumen bulbs (which I have 3 in my ceiling chandelier, and it's barely enough to make my room comfortably bright) is that they don't last too long because of overheating: each drains about 20W and apparently it's very hard to dissipate that much heat through the plastic fitting.
Having a controller in the bulb also makes it easier to handle things like colour temperature and hue, more difficult to manage from a wall switch.
Agree wifi is clumsy, things like Zigbee work much better for these applications.
Commercial spaces and some houses in the UK already have 12V DC lighting and sockets for spotlights using the GU5.3 socket. It lets you have smaller and cheaper light-bulbs.
They're not great to have as a renter, though; many of the AC/DC converters were designed for Halogen light-bulbs, and have a minimum wattage requirement, so the more power-efficient LED light-bulbs don't work. Plus, since the fittings are less common than E27, the choice of light-bulbs is much lower.
If solar power and home electric storage batteries become more common, we might start having DC run directly into houses, to avoid the DC/AC conversion from the battery, then AC/DC back to computers/light-bulbs/TVs. But that seems years away!
P.S. smart light-bulbs need to have individual micro-controllers if you want to have colour control. I rarely use the funky colours like green/blue, but I do love being able to go from white light in the morning, to a warmer/orangish light at night.
But I have a van that I converted to an RV and it has a 12V fridge, 12V fans, 12V pump (for the water), 12V chargers for my phones and of course 12V lighting.
I think a good deal of electronics in a home could be run on 12V DC.
My local hardware store has a fairly large range of smart light switches. I had no idea they existed until now.
But the cost and difficulty of installation must drastically reduce their appeal. Sure they're not hard to install, but it's probably best not to encourage DIY electrical work.
I would actually like someone knowledgable to chime in if possible: what is the actual risk here on a standard 110 switch? I’ve changed a bunch of switches (and even zapped myself a bit) without ever having the slightest of issues.
I'd class swapping outlets, switches, and light fixtures, as among the kinds of home-repair stuff you have to know & do unless you're rich enough to never personally do work on your house. Like replacing a faucet, a leaky branch water shut-off valve, or the tank-innards of a toilet.
I can only add that I've been doing my own electrical work for over a decade now and have been successful.
Haven't been zapped either (I'm extremely careful about that, not so much when I was young).
To be sure though I have stopped at the point where a new circuit/breaker would need to be added to the panel, new wiring run. Although I do understand in theory how it is done, how to select the right gauge wire for the circuit, staying within the current limits of the panel....
It does here also in newer homes. But just getting it into the switch and then shoving everything back in the box is a challenge. Smart switches are much fatter than normal switches, and by boxes at least were wired with hard copper that was difficult to shape. I had to take out the switch for my cabinets to make everything work.
Even a controller for a ceiling fan is a challenge. Sometimes, if you are lucky/persistent you can replace the box itself with a deeper one to make wiring easier.
I want to get a deeper box in eventually, but I have the feeling I should hire an electrician to do that, then the dry wall will have to be redone, and...maybe it isn't worth it now. Besides, I can get some good battery operated cabinet lighting to replace the ones I sacrificed the switch on.
if you're using 9vdc to power all your light bulbs from a central location, you're going to need a lot thicker cables since the amperage you're using is going to increase tenfold (or ~x20 for anywhere using 230v). and switching DC is going to be similarly more problematic since you won't have a 0-crossing 100/120 times a second to break any arcs, wearing out your switches
I have a bunch of philips hue lights, a nest thermostat and some blink cameras hooked up to both alexa and google hub and it all just works and has for years. I used to have problems when I was buying a bunch of cheap china mix and match products.
Fun story time - One time I woke up in the middle of the night. Not knowing what time it is, I shouted into the darkness "Hey Google, what time is it?" To my surprise, the machine reply in a excessively cheerful tone, "Welcome to what fruit are you" and proceed to give me a personality quiz. I wasn't able to get the time - but I found out that I'm a watermelon.
So I think it is a mistake to call smart home a failure.
My house is a big, fancy container for people. Where it is nice, is where I can easily reconfigure it. Think moving furniture around, wall coverings and so on. Easy. Where it hardest to get right is where you can't easily re-configure and re-decorate. Bathrooms that are fixed by plumbing. Kitchens fixed by immobile cabinetry built around plumbing. Inflexible wiring. Smart isn't the problem. Flexible is.
Reading the article I realised I actually have plenty of "smart" home products. The catch is that they are not really smart but I like what they do.
Small lamps which I can switch on and off directly from my phone without having to go through the room are nice. Being able to make the light warmer when I want is nice too.
I have a smart thermostat which I motly use to warm my place the day before I get there when I am out for a long time. That's pretty enjoyable too.
My scale graphs my weight automatically when I remember to use its "smart" feature. When I was paying attention to it, it was handy.
I also like the fact that it's all piloted from my phone which I mostly always have and apart from the thermostat none of them require being online.
At this point, the smartness is mostly just remote control and sometimes a feedback loop with a sensor. That's not very smart but I don't really need more.
Sounds like discretionary adoption along the lines of a typical hype curve based on what you describe;- Initial optimism about a hyper integrated, seamless home full of intelligent devices that sync up and act as a cohesive whole. This gives way to eventually settling on reality where a few disparate clever devices deliver high value while being relatively low complexity and working mostly independently.
I've struggled to find even that much value in smart products - we flirted with a Hue system but at literally 10x the cost of ordinary led lights we didn't get beyond a few bulbs before realizing it was a money pit with at best marginal benefits. We have a few nice nanoleaf feature pieces that run themselves on a simple schedule. I can't see much else making it's way into the place and creating additional headaches.
I felt like a cynical old git shaking my head about new the work coffee machine having a touch screen and needing wifi to work, but seeing how often the same machine was 'down' waiting for a tech and the absurdity of colleagues who didn't quite know how to work around such an outage... This kind of stuff may be creating value for investors somewhere but in the long run there's little chance it's improving the lives of those it's being marketed to.
the entire architecture is awful. The smart home has all these smart devices calling out to cloud APIs for even the most basic functionalities that could be performed locally.
The "Cloud" is used to turn on a light in the house. I'm not sure if they want to collect information but smart devices that operate locally have far less failure cases.
You mean, I shouldn't be distributing my 'light switched on' and 'light switched' off events over a highly available kafka cluster hosted on another continent? What if I want to replay the entire history of my switch flicking?
Are there any companies or projects working on a completely local “smart
home” hub? I suppose most companies are care more about sucking up our data and providing a “service” with a monthly bill than about providing useful hardware. It feels like there’s a huge gap between the best possible smart home product and what’s actually available.
Home Assistant is the leader here. One of the largest projects on GitHub with a “local-first” philosophy. They have a cloud service for exposing your instance but you don’t have to use it.
I’ve been using it for years and it’s matured quite a bit recently. Combine with Zwave of ZigBee devices and the only thing you need internet for is voice assistant if you want it
(Totally not required)
I set up one by our kitchen that is mounted near the refrigerator. The kids use it to see whats on the calendar, what chores they have in the reminder app, and what is for dinner (using a summary sheet in numbers that pulls from our meal plan).
The reminder app is synced across all our devices, so we (parents) can add things when we are out and about. Also, the shortcuts app can automate adding repeated events.
I can imagine a future where there are little displays near light switches that communicate such information in different rooms. We will likely do this with old iOS devices under the same appleID.
I guess that I have a smart home albeit a cheap version. I have used sonoff smart switches to give me remote control of my flat's electric radiators. This has given me extra control over their schedules than available using their in-built interfaces. Plus I can turn them on remotely if I am unexpectedly coming home so the flat is warm when I get back.
I live in France and the radiators have a control pin that accepts various 230V AC waveforms (!) to set different functions but luckily simply on/off can be used to toggle between the radiators' high and low temperatures (that I set via each radiator's internal interface). It's crude but effective. I use an app to control the remote switches; I'm not totally happy about relying on a cloud based app but I can always bypass the remote control locally if there's an issue and in summer the radiators' are simply switched off so the risks are minimal.
The biggest hindrance is showing value in upgrading to the "smart" devices. A light switch connected to wifi and a phone is just not needed in 99% of the cases. Operating it with a phone is just not incentive enough to replace it with something that is 10x more expensive than the original. Same thing with other "smart" items too, at least in my case. Something like a garage door opener is different because its relatively expensive to replace than fitting an adaptor to automate, and there could be more people needing to operate the garage door remotely. I don't see an explosive market there. The only easy market is the "scare" market which is what Ring exploited and pretty much rules the camera doorbell.
By far my best experience with any sort of smart home device was for my apartment in Boston. My overhead light did not have a switch, and the previous homeowner used a pull cord. I replaced it with a Phillips smart LED that connected wirelessly to a physical switch that I pasted onto the wall. No electrician, and now my light has a switch. $30 well spent.
I never did get around to connecting it to an app or anything like that.
Oh, I also had a WiFi enabled outlet! That was a gift from my sister in law. I used it on an air conditioner -- if I remembered, I could turn on the AC an hour before I got home.
I also have a WiFi enabled AC in my new home and can set each room's temperature individually by app.
These things are useful and convenient but I'm not really going out of my way to set them up.
Yes, “why bother?” works for most. Those of us who do, or don’t, want “smart home” don’t need to justify it.
For those who do . . .
Should be simple installation matching existing patterns. Switch & outlet covers shouldn’t need to change, as there’s just enough failure to fit to make updates unworkable (haven’t seen any smart switches fit a 2-3 switch cover plate). Blend traditional operation with smart (smart wireless switch next to normal wired switch annoyingly interfere). Don’t compel grouping of lights (wife’s light must not be coupled with mine in automatons). Ensure sensible symbiosis (“on at sunset, off at 9pm” shouldn’t leave light on when sunset is 9:02pm). Stay on the right network (I have two).
The problems are petty and easily fixed - yet nobody wants to solve them.
After what happened to Revolv customers when Google bought them I'm never going to spent more on a smart device than I would on a good lunch. Sure, I'll put some five and ten dollar smart switches on lamps but this whole home integration crap is for the birds.
I've often thought that it would be good to have a app that could tell me that my doors and windows were shut and locked for when I suddenly can't remember if I actually did it. However the security implications of having such a system would worry me, particularly if it became popular. If hacked, it could tell thieves which houses had been accidentally left unlocked in an area, which could be worse than accidentally leaving a door unlocked all day. Also it might even make the company running the system liable for any thefts caused by hacks to their system and so it might be hard for a company to run such a system.
My issue with smart devices is that they aren't smart. They are just internet controlled devices that can you can control via an app. To me, smart devices are objects that have added value by doing a task more intelligently. Think like a thermostat that can learn your lifestyle and adjusts the temperature accordingly. Or like a microwave that sets cooking time based on what was put in it. These are smart devices, and honestly, I don't think there are actually many out there now.
I've said it multiple times, and I'll say it again. Smart home is a supplement. Your home should work in a way that when smart stuff fails, home won't fail. Take a light switch as an example. You have to be able to get up and manually switch lights on/off when an app or a smartswitch won't work. If smart heating won't work (sensors or smart regulators fail), heating should work anyway, it just wouldn't be as smart.
I had a Nest thermostat before they were acquired. I liked it because the key feature I bought it for, the ability to automatically adjust the temperature, worked pretty well.
I think that's the key for me personally: will this actually make things easier for me? I don't want a ton of notifications I have to respond to.
Our older Logitech Harmony hub was useful for the same reason: it reduced multiple remotes to a single app that worked anywhere in the house via wifi.
I've had a "smart home" for about 20 years now, and it's been great! Definitely not a failing concept.
But I wouldn't buy any of the modern "smart home/IoT" devices mostly because they're security nightmares, although tons of them have other serious issues as well.
The way that most modern "smart home" equipment manufacturers design their devices is the failing concept, not smart homes in general.
User story: It is 5AM and you are fumbling into the kitchen in the morning to make coffee and you cant see anything. You reach out to the wall and flip the light switch into the "on" position. Memories of smartphones and voice assistants are still vague concepts on your mental horizon.
I feel like zero engineers in the smart home space have read the Don Norman book.
User story: It's 3AM. Your home confirmed with you yesterday that you want to wake up early for a flight whose confirmation it found in your email. The lights are gradually brightened and blue-shifted through till 4AM. While that's happening, at 3:50AM your coffee maker is preheated. An audio alarm gets you up at 4AM, and motion-activated baseboard lights guide you to the kitchen as you stumble around in your groggy state. A glow emanates from the cupboard door hiding your favorite mug. After your coffee is done, as you're rushing out the door you realize you can't find your keys -- panicking for a moment you ask your home where it last saw you carrying them, and it tells you they're on your bedside table. You head out the door and catch your flight.
Everything mentioned above is pretty straightforward to rig together with off-the-shelf hardware and software, even if a little time consuming.
> Everything mentioned above is pretty straightforward to rig together with off-the-shelf hardware and software, even if a little time consuming.
Sorry, but it's not. There's a reason why you won't find any useful smart coffee machine on the market. And your house knowing where your keys are is either so specific a scenario that it's useless or contents of fantasy science fiction.
The one thing on your list that could be made today is the lighting going on at the right time, and migrating from yellow to blue. You will probably need to assemble your own lamps, because even high-end stuff don't have enough quality for that use-case, and proprietary software will get on your way on every step. But it's possible.
> There's a reason why you won't find any useful smart coffee machine on the market.
Sure! It's expensive to shove a bunch of smart technology into every single device, especially given that everyone wants something a little different so your interface would need to be feature-rich to support those use cases. Note that we don't really care about having a smart coffee machine in the outlined scenario; we need a coffee machine that can be preheated, which doesn't have to be any more complicated than a switch on the power supply on the right dumb machine.
> And your house knowing where your keys are is either so specific a scenario that it's useless or contents of fantasy science fiction.
With off-the-shelf software it's a mix between the two. The hard bit wouldn't be the location of simple objects like keys (throw a few sensors around the house, customize the final layer on a pre-trained model, and pick your favorite approximate inverse-3d mapper to get a rough location in space of the object); it would be the location of complicated objects (house, where's my Lang's Algebra?), cases where you don't have enough information but would hope that the house might figure it out (e.g., boxes of things), and the humanistic output (rather than saying they're on the bedside table, you might have to resort to a screen that can show locations of things on a heatmap).
> You will probably need to assemble your own lamps
That was a communication error on my part. "Off-the-shelf" included things like off-the-shelf microcontrollers, wiring, ....
> and proprietary software will get on your way on every step
And is best avoided. I'd be tempted to hook into Google Calendar or something for upcoming trips since they've already implemented models to find that kind of thing, at least starting out, but otherwise I see no reason you'd want to use proprietary software in a project like that.
Sure it does. There's a setup cost, and it's fully possible that it's not worth messing with something like that unless you enjoy tinkering (and the way I see the market moving it seems likely that smart homes will stay that way for awhile). The user story definitely exists though.
> What if you feel like oversleeping and now your coffee is cold?
In the outlined scenario, your snooze function would want to also turn off the lights, the motion-activated baseboard lights going to your coffee maker would work fine, and your coffee maker would still be preheated and have wasted a bit of power if you didn't have some kind of a timed cutoff or similar.
The first smart switch I installed was to preheat my coffee maker in the morning. Motion activated lights heading to the kitchen before dawn came soon after.
I have a semi-smart home in that I’ve got a few smart plugs connected with Alexa and it’s useful to turn on/off a bunch of lights at once - or have them come on in the evening while away. Other than that I’m not really sure what more I’d want. Maybe some kind of instant-disco lighting so I can rock out to Dire Straits?
A smart thermostat is useful. I would also like smart curtains (I made a prototype that I hooked up to Google Home ok - "Hey Google open the curtains" works, but the actual physical bit is tricky).
A smart lock would be very useful too, e.g. for granting temporary timed access to cleaners etc. But there's no way I trust any of the smart locks on the market with my home security.
I have some smart scales too. They are pretty great.
It isn't the concept that is the problem. It's the lack of standards and long term support. Non-smart curtains, scales, lights, locks, etc. essentially work forever, whereas I'd give smart versions 10 years tops. And I don't really want to lock my house into 5 different vendors' proprietary systems.
> But there's no way I trust any of the smart locks on the market with my home security.
The security provided by most door locks is completely illusory. You can break into most homes very easily. Most people have unsecured windows on the ground floor.
Most locks are there as a way to signal that this door shouldn't be opened. It's mostly a psychological detterent. As long as your insurance is fine with your smart lock, it's probably not much worth than a dumb lock.
The exception is of course the upper floors of apartment complex in large city in which case the five points locking system on your door is there to signal that it will me more time effective to break into your neighbor place.
> It's the lack of standards and long term support
Completely agree with that. All companies want to build a locked plateform. That's hindering the market growth a lot.
> The security provided by most door locks is completely illusory. You can break into most homes very easily. Most people have unsecured windows on the ground floor.
Maybe in America but that's definitely not true in the UK. Modern windows, doors and locks are really quite secure. It's definitely not completely illusory security.
If you watch LPL's videos many smart locks are designed with very little consideration for physical security. Exploits are often as easy as drilling a hole in a zinc casting and pressing a reset button, or even just removing screws.
That's way easier than drilling or snapping a lock with drill/snap protection or trying to lever open a composite door with multipoint locking, or even breaking through a toughened glass window, all of which my house has (and it's not at all an outlier).
Yeah I feel the plugs are a good enough thing for now. If they are duff in ten years? So what - it was a £7 plug and I’ll get some more with a hopefully open standard by then. I’d never go all-in the way things currently are.
Businesses can adopt "smart" technologies because the decision makers are spending someone else's money, either on the initial outlay or on future maintenance / obsolescence. At work, the army of maintainers are your colleagues. At home, the maintainer is you.
Insteon is pretty good (Wall switches, lamp dimmers and appliance modules). No central hub required. They form relationships with one another without a middle man. There's also a USB module if you like programming your own stuff.
I've had great success with smart switches/dimmers/outlets/locks, mostly because they allow you to set more complex schedules and timers.
I rarely use the remote on/off capability, but just having a device that works on a schedule is really nice.
Examples:
- Our late-1980's house has a lot of accent lights scattered about. We never used them before because we had to flip a switch on each one. Now they come on automatically at dusk and turn off after we go to bed. It's nice.
- Our TV speakers don't have a sleep mode. We use a smart outlet to power them down automatically at the times when we usually don't use them.
- Our 2-year-old used to turn his very bright closet light on and stay up late in his room. We set a smart switch so that turns off after a couple of minutes. Now he sleeps much better.
- Our kitchen lights were way too bright and people kept leaving them on, plus the two sides of the room were on different switches and one was brighter than the other. Now they come on at an even 20%/40% brightness unless double-clicked, and automatically shut off after a few hours if people leave them on.
- Our garage door shuts itself if we forget to shut it.
- It was too inconvenient to keep the busiest exterior door locked. Now the deadbolt locks automatically after a few minutes, and we can give out visitor codes the babysitter and for people who come in to work on the house etc.
We put all this stuff on a separate WiFi network (to alleviate some of the possible security issues) to the extent it needs wifi. We have a mishmash of different things from different manufacturers, and no central hub. But things have been working really well for a couple of years now.
I don't own, but when I do I will go out of my way to try and buy unconnected devices. There is no physical device in my home that I want "as a service" in any way.
tldr: unreliable hardware + bugs ... results in endless hassles
On the other hand, you have the ultra-high-end of smarthome tech with systems like Crestron. It seems ok but it's very expensive (million dollar mansions/estates budgets) and the custom programming is proprietary so the homeowners can't modify it themselves.
Until this gets solved somehow smart features will always remain a marketing gimmick. There's no way around it.
People accept replacing ageing phones and computers because their connectivity features are a large part of the reason they exist. This isn't true for a light switch, for instance, which is mostly seen as a way to turn lights off.