The smart home is flailing as a concept
fastcompany.com
fastcompany.com
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.
The smart home should be an open source hub that integrates appliances, switches and lighting etc.
But the business model requires vendor lock-in closed systems.
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.
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.
The bottleneck here could be the network interface once you are away from your home. That is where standards struggle.
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.
Vendor lock in bad.
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.
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).
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.
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.
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.
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.
And with a huge confession that this isn't that big of a deal.
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.
Hold and release to manual dim.
That said, they do fit my bill really well and I'm almost certainly going to switch to them eventually.
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.
Edit: Regardless of what I meant, I think you proved I should look more into these. Thanks!
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.
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.
This way you're not stuck with a useless home in case your controller lost power, is down for maintenance or whatever.
"Press the button for 3 seconds, but not five."
"Do the pairing near the receiver."
And you have to swap batteries.
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.
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.
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.
For what it's worth this is possible now with their Smart Button
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.
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.
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.
For example, Software Engineering :-)
"Gorilla arm"[1]
[1]: https://en.wikipedia.org/wiki/Touchscreen#%22Gorilla_arm%22
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.
Photocells / IR motion sensors turn lights on or off as ambient light levels fall or presence of a human is detected.
Audio switches turn on or off with a sound ("The Clapper", etc.)
Timers can be used to turn circuits on or off on a set schedule. This may not be ideal, but can work for general cases.
There are remotes which work based on a handheld control and either IR or RF frequencies.
All of these are backwards-compatible with old technology, are tried and proven, and will continue to function for years.
"Smart Home" implementations remain 99.99996% crap.
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.)
YEEEESSSS
Minority Report is my go-to example for sexy, total impractical.
Not quite “IOT” but the bridge between that low power network and the internet is a natural add on.
Smart homes to me have always felt like tech looking for a problem to solve.
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?
The timer is something you can leave on all the time. If you don't come home before dark some day, you still want the lights to go on.
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'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.
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.
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'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.
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.
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.
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.
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.
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.
I'm not sure this holds up any more. It seems to me most mid-range or lower modern appliances fail within a decade.
That said, I only have one "smart" appliance (a Miele dishwasher), and its "smart" features were useless last time I checked.
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?
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.
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.
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.
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.
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.
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 was just going off the larger number of password dumps there are out there. Most are from small companies.
[0]- I wanted to monitor for deliveries, but COVID kind of trivialized that for a year.
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.
[0]https://www.crestron.com/Products/Market-Solutions/Crestron-... [1]https://www.savant.com/ [2]https://www.control4.com/solutions/whole-home/
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).
Just a small issue: huge installation cost (because wires, but also software license)
* 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.
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.
- 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.)
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.
https://www.electrical101.com/4way-switch-wiring-using-nm-ca...
> 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. :)
Completely arbitrary on a two way (wait, they're called three way?) switches.
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.
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.
These are available in "smart"/connected form or regular electronics-only form. Example here:
https://www.amazon.com/Lutron-Incandescent-Single-Pole-Multi...
I agree with you and much prefer them to the traditional switches.
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.
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.
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'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.
Depends on the situation. My town wants us to water less frequently, but keep our grass from drying out to prevent a fire hazard and soil erosion.
I do wish my front lawn was smaller to be sure. I'd be happier with more garden beds or garage/shed/workshop space.
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.
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.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.
By wiring the solenoid correctly, it fails to an off state. It’s basic as all hell and works really well.
Also, 'Computer, locate my husband.' 'Your husband is in the attic'. etc.
all the times the crew got trapped because the doors were made inoperable by [plot device]?
Of course they did. Most of the Admirals were like that.
Well there’s the problem.
- connect wires behind light switch so bulb is always-on
- screw in smart bulb
- stick smart button on top of the hole of the old switch
I think you can do the same with Hue.
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.
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.
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.
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.
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.
True, but phones present a much higher security risk than remote controls.
Only because the manufacturers insist on networking the devices instead of using Bluetooth.
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.
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.
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 have a light that automatically comes on in the bathroom when someone enters, shuts off a few minutes after they leave.
With kids, it has turned out to be an energy saver.
No hub required.
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.
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).
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?
Something like this https://i.ebayimg.com/images/g/nvAAAOSwEK5fmkrH/s-l1600.jpg
No I'm not, but those work too.
This is what I bought: https://www.ebay.com/itm/153285261299
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.
Until you're drunk AF and trying to take a piss (or whatever it is you're doing) at 1am and then the "smart" light blinds you.
Here's my solution: https://www.ebay.com/itm/265164805260
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.
Leaving a light on doesn't matter much. Leaving billions on can clearly have an impact.
See David MacKay's Sustainable Energy Without The Hot Air, chapter 19. https://www.withouthotair.com/c19/page_114.shtml
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.
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.
Or, you can use a Broadlink IR blaster to clone the remote.
If your heat pump talks to MEL cloud (Mitsubishi's cloud thingy) then this should work: https://www.home-assistant.io/integrations/melcloud/
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.
https://www.home-assistant.io/blog/2016/01/19/perfect-home-a...
An app per device class/vendor is beyond useless. A centralized, single controller that mostly has no interface at all is the solution.
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…
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.
The just-audible click at 6am followed by the sound of the boiler heating is very satisfying, even 5+ years after setting it up.
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.
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.
- higher maintenance
- loss of privacy
- loss of ownership of things in my home
- reliance on subscriptions
Every extra feature is a regression.
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.
That sounds like the plot to a Black Mirror episode
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".
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.
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.
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
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.
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.
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.
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.
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).
https://www.youtube.com/user/bigclivedotcom
More specifically, example: https://youtu.be/klaJqofCsu4
> Non-magic users: collect crystals, call their pet a familiar, draw pentagrams
> Magic users:the most magical things I keep in my house are rocks, and I keep a hammer next to them in case they act up
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.
Home Assistant: https://www.home-assistant.io
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.
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.
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.
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.
Smart homes aren't super popular because of integration issues imo.
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.
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?
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.
We've been using about 50 Hue bulbs for years, without a single one going bust. But I'm sure you're able to buy cheap crap that is... well, crap.
2) Replacing one bulb every few years is much less time/money than a wiring project IMO.
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.
https://www.screwfix.com/p/tcp-1-gang-1-way-smart-light-swit...
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....
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.
I'm picturing my future home though, festooned with "cigarette lighter" outlets throughout the house.
But you won't replace all the 15A/30A lines on your home with 12V lines. it's not viable.
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.
So I think it is a mistake to call smart home a failure.
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.
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.
https://www.bbc.com/news/technology-54554408
https://www.bbc.com/future/article/20200511-how-smart-home-d...
https://www.nytimes.com/2018/06/23/technology/smart-home-dev...
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.
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 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.
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.
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.
And by control I mean including turning them on / waking.
I think the whole line is discontinued now, but looks like you can still find them easily.
https://www.youtube.com/watch?v=jyrTgtPTz3M
Many of the presented ideas are eubiquitous today, and in a luxury modern house, almost all of the presented features exist.
I wonder if any vision for future living exists today that will turn out to be as prophetic as the Westinghouse Total Electric Home.
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.
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.
It's really the only thing that's good in the smart home world rn. It makes all sorts of otherwise crappy IOT hardware work decently together.
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.
Everything mentioned above is pretty straightforward to rig together with off-the-shelf hardware and software, even if a little time consuming.
What if you feel like oversleeping and now your coffee is cold?
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.
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.
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.
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.
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.
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).
https://www.cnet.com/home/smart-home/tenants-worry-smart-hom...
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.
https://news.ycombinator.com/item?id=25099615
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.