There’s not really anything that can happen to an older car, aside from a major accident or rust, that can “brick” it. Every part is replaceable and reasonably available for the common ones.
I’m probably going to need to own a car again in the not too distant future and more than likely will just try to find a ‘70s VW to get up and running. Though…I do still technically own a ‘74 VW Bus my dad was using for a gate the last time I saw it around 20 years ago, wonder if he still has that thing?
I don't think the market always efficiently sorts out every dumb technology fad, but I have a moderate level of confidence that people will stop buying cloud-dependent physical products after being burned enough times by stuff like this. Or, maybe they just realize it's not all that useful in the first place.
Alternatively, the products could get much better, and doubters like me will have to stop complaining and jump on board. We'll see about that.
But, I don't think a law would improve the situation. More likely it would just backfire in some unexpected way, as such well-intentioned laws often do.
> I don't think the market always efficiently sorts out every dumb technology fad, but I have a moderate level of confidence that people will stop buying cloud-dependent physical products after being burned enough times by stuff like this. Or, maybe they just realize it's not all that useful in the first place.
Individual people probably will, but the market will keep supplying new people who haven't yet been burnt and are willing to fall for the same old scam.
This is pretty common, I bought a really nice looking scale that was machined aluminum and had no screen from Community Thrift. It was $5 but only when I got home did I realize it was some app-first company that had closed. So it was a brick.
I've also seen things like Lighthouse (?) that was a really nice IP security cam with the same fate. IMO the coolest thing would be if these companies would open-source their code when they went out of business.
But in all likelihood that is probably a legal issue. Companies don't just go out of business, they sell IP to something and ultimately some other company probably ends up owning it all and they probably don't want to share anything.
If the governments of the world can force companies as large as Apple to move to USB-C to reduce e-waste, it doesn’t seem unreasonable that they could demand open sourcing of software as part of a company’s wind-down.
Easy: legislate an implied obligation for ongoing support of past sales for a reasonable period of time and as part of wind down that obligation is a liability that must be discharged with posted credit of some sort (or the purchaser would have to take this liability on) or via open sourcing the software.
Fair point. That's called a warranty, and they should have been legally required to support their product until the end of warranty period. Posting a notice on their website in June and shutting down the servers in September is a jerk move.
My point still stands though. They can't force them to open up their trade secrets.
If the government can make a law that says someone can't sell a particular product (Like say, children's toys with razor blades in them) why can't the government pass a law that says that if a company sells a particular catagory of product then the company must open the product up if they discontinue the product or go out of business?
All laws are social contracts. Of course they could force them to open up their trade secrets. It wouldn't even to be something extreme, Congress could just pass a law saying "discontinuing a commercial product requires publishing everything needed to maintain that product", and that would be that.
> You can't force someone to open up their intellectual property. The company might have wound down but the IP still belongs to someone.
You can't, but governments can. There are laws that protect IP; those laws could be modified to say that, if you decide to stop supporting a product that relies on your IP, then you must provide the means for individual users to support it themselves. That doesn't even have to mean giving up your IP; it could just mean designing from the beginning with the idea that a company won't last forever—but that'll never happen unless it's forced by law.
Source code escrow agreements are reasonably common. When a startup receives an investment from BigCo, it comes with the condition that BigCo gets a perpetual license to the source code / IP if the startup fails.
It would be straightforward to mandate such agreements be included with all IoT device purchases.
Ideally, they'd also be triggered by acquisitions or modifications to privacy policies. I can dream, right?
I’m not sure I understand… the sensor detects the clothes are dry, sends the owner a message, and the owner has to walk downstairs and turn it off? Or message them the appliance is overheating due to blocked lint trap?
Why not just have it turn the machine off without intervention? This is what my last X dryers have done and generally done a decent job at it (for dryness, not for lint trap blockage). The key has been not overloading the appliance with too many heavy items and keeping the lint screen clear, both of which should just be habit for anybody who does laundry.
This is presumably for people that don't have such a choice. I'd have a decent dryer if I didn't rent, but I do and such things are typically included, and consequently are the cheapest money can buy.
To do this I'd need to integrate the dryer sensor with a smart plug.
It’s hard to turn a dryer off with software. Most are gas and even for electrics, smart plugs usually don’t support 240V. So you’d need a different kind of arm or hand for each kind of dryer.
Maybe not in US. In AU, for example, they do. Also I'm surprised to hear that. I would have expected them to be universal and work anywhere between 110 and 240V.
In the US the 240v outlet is a physically different shape and size, and there's usually only a few in each home, where powerful appliances are expected to be fitted.
You could probably fit an adapter, but I imagine smart plugs in the US are built with the assumption they will never have to handle such a high load.
In the US a 240v plug is a physically very different plug designed for running high wattage domestic appliances like a dryer, oven, or electric range. Most homes only have one of these installed in the location where that appliance is expected to be. They are typically rated at 30 amps and are actually two phases of 120v. It is certainly possible to make a smart outlet that can handle two phases and 30A, but it would be expensive, and have limited market.
A smart plug must extract a small amount of power from the line to run itself--and that means a power supply that very well might care. There's little reason to make the power supply universal as the plug is part of the unit and the plug specifies the voltage. You very much do not want plugs meant for different voltages to be interoperable! (20+ years ago I had multiple run-ins with this. China uses 240V and one of the two plug types is the same as US 120V. My FIL would try to plug US stuff in without using the transformer and of course it would promptly self-destruct. By now most things one might be traveling with are universal voltage but back then few were.)
Not only are US 120V plugs different than 240V plugs, but the 120V/20A plugs are a superset of standard 120V/15A plugs--a 15A plug will fit a 20A receptacle but not the other way around. I do not believe code permits plugs for devices beyond this range--the really heavy stuff is always hardwired. I believe this is to prevent morons from trying to power their house from a generator with a suicide plug--forcing people to use a proper transfer switch that can't backfeed. I suspect this will change in the future due to how dirty small ICE engines are--as it stands things like electric mowers aren't very good because they can't get enough power to be good. Likewise, pressure washers--the high power versions are always ICE. Where I grew up you needed at least 4hp in a lawn mower--and you can't draw 4hp from a normal wall plug.
This is a very US centric view. In the countries I have been I only encountered electric dryers.
Personally I have a heap pump dryers that does a good job at turning off.
Utility natural gas does tend to be regional. Where it's available, it's most often much less expensive to use natural gas for heating than electric resistance heating, although you usually have to pay more for the appliances. If electricity is inexpensive, gas is less desirable. Most gas appliances can be adapted to use propane, and propane deliveries are available in a wider area, but you probably want several propane appliances to justify the plumbing and the storage tank.
Electric heat pumps can be more cost effective than gas though, but there's details.
It's a sensor that goes inside the dryer, so it wouldn't be able to turn it off for you unless also connected to a smart plug, but this is also problematic as another comment pointed out. Many are stuck with typical dryers that are nothing more than a heat setting and a timer. If your machine already does this for you, then the product obviously isn't for you.
Sharing service parts suggests they aren't exactly the same, ie it has changed as times have changed. Also sharing service parts doesn't preclude them adding a moisture sensor and shut off.
Apparently you buy it because you 1) can't hear the humidity sensor alarm built into the dryer already that already turns off the dryer, and/or 2) are a useless gadget cuck who can't go without having it on your smart phone.
This type of device seems perfect for HomeKit... but some company decided to DIY the tech so they could find a way to eventually monetize. What an unfortunate waste.
As a power user, of course I prefer to connect to devices directly rather than through a sketchy cloud service.
But we've also got to acknowledge that the networked security cameras of 15 years ago regularly ended up making their videos public to anyone with a web browser - or becoming part of a DDOS botnet. So I can see why manufacturers aren't in a hurry to ask users to forward ports from their home firewalls any more.
There should be a different lifetime attached to these cloud-based smart home devices. It's an implicit lifetime separate than the hardware: how long can the company last? And what's their frequency of plug-pulling? If it's not an established company, I would be wary buying anything from them. Even established companies tend to pull the plugs on cloud-based hardware, so caution also applies to them.
My current guideline for purchasing:
- How long will the company last?
- How likely is it for the company to sunset a product?
- Are there fallbacks in place in case of company closure or server shutdown? This should be supplied by the company, and not cobbled together by users after losing server connection.
The biggest thing, I think, is that it needs to be clearly advertised on the packet (in a “batteries not included” style) that the device depends on XX service provided by YY company.
I’m as mad as everyone else about unnecessary cloud integration and companies bricking useful hardware but that description seems a REAL stretch here.
> In other words, SmartDry will become a tiny brick inside your dryer unless you're willing to procure a little ESP32 development board, load some code onto it, plug it in near your dryer, and set up your own alerts in your Home Assistant server.
Now I’ll admit there are cases that still count as “bricked” even if the device is recoverable with a Dremel and a JTAG debugger, but generally if you can fix it without specialist tools or software it’s not bricked.
In this case it sounds like the device still works fine and you just need to provide a server for it to talk to.
The "hub" is a 110 V smart plug that can shut off gas dryers, and the smart phone app doesn't work with other servers, so there is significant loss of functionality.
The actual sensor isn't bricked at all. The functionality most people bought it for is going away.
Why isn't there catch-all service for hosting lambda-like functionality for regular, like there are a billion offerings for cloud storage?
For example, I'm 99% sure Azure blobs are the underpinnings of Onedrive, and I'm pretty sure using just blobs, you could implement most of the functionality with just blobs, including stuff like push notifications.
They could offer a service where people could connect their smart devices to their Onedrive accounts, and it would set up a few service endpoints for their devices, and all hosting for these online features would be taken care of by the cloud provider. The load per user would be so low, it would generously fit into any 'free tier' these providers offer.
The infra is there, people are already used to paying for cloud storage, why isn't this a thing already
This is why anything remotely “smart” that I own and use can be self managed. Relying on the “cloud” and longevity of companies is asking for years. I hate that crap.
I'm just fascinated by how this term "cloud-bricked" is probably understood by everyone today, not techie people, everyone or rapidly approaching everyone, probably understands the essential properties and concept like any other basic vocabulary word, simply because this stuff has permeated all levels of everyday life like electricity has before that.
The term, and this idea about how ordinary a concept like that is now, made me think just now that modern English up to about 1990 (widespread adoption of the internet) may eventually become a new Old English.
There will be so many terms that have such different meanings and usage before and after a certain time, that they're almost different languages. And we're kind of in the middle of the transition in real time right now.
This is one of the many reasons why I NEVER buy any type of HW that requires a cloud account to even function. There is no sales pitch or value proposition so good that it can overcome this downside of Cloud/IoT in practice.
I was looking at a weather station recently that used some pretty innovative sensor hardware but then it only works with a cloud account. Literally if you want the data on your phone, you have connect the station to your internet connection, let it upload the data to the cloud, and then have your smartphone download it again. A thousand-mile trip for a 10 foot distance. F___ that noise!
When you live in an area with spotty internet service, this is utterly Epic Fail.
So it was an automatic no-go. No sale on that fine weather station. Hope you go broke!!
Instead I'm building a system from scratch with professional grade components that use 4-20 mA wire links that I'm connecting to an RPi. The cost for the components is definitely higher but they are the same as what the NWS uses. And I can chose whether to upload or share any of the data. And I will have access to MY DATA for as long as everything powers up and operates.
F___ companies that require cloud access to your own data
They got the use of it for however long, for whatever price. Was it worth it? Maybe.
It's not a shocking surprise that a device has a limited lifetime. People who buy $1000 smartphones, just to buy another one two years later, are complaining about this?
Just because something ends, doesn't mean it wasn't good while it lasted.
65 comments
[ 2.0 ms ] story [ 125 ms ] threadthat seems like a catchy term
There’s not really anything that can happen to an older car, aside from a major accident or rust, that can “brick” it. Every part is replaceable and reasonably available for the common ones.
I’m probably going to need to own a car again in the not too distant future and more than likely will just try to find a ‘70s VW to get up and running. Though…I do still technically own a ‘74 VW Bus my dad was using for a gate the last time I saw it around 20 years ago, wonder if he still has that thing?
HN is never short on comments advocating for specific policies that would sum up to that effect, so yes.
But I'm totally with you on fighting for the necessary regulation! In the meantime, I need to do the needful to protect myself.
Alternatively, the products could get much better, and doubters like me will have to stop complaining and jump on board. We'll see about that.
But, I don't think a law would improve the situation. More likely it would just backfire in some unexpected way, as such well-intentioned laws often do.
Individual people probably will, but the market will keep supplying new people who haven't yet been burnt and are willing to fall for the same old scam.
But in all likelihood that is probably a legal issue. Companies don't just go out of business, they sell IP to something and ultimately some other company probably ends up owning it all and they probably don't want to share anything.
My point still stands though. They can't force them to open up their trade secrets.
If the government can make a law that says someone can't sell a particular product (Like say, children's toys with razor blades in them) why can't the government pass a law that says that if a company sells a particular catagory of product then the company must open the product up if they discontinue the product or go out of business?
You can't, but governments can. There are laws that protect IP; those laws could be modified to say that, if you decide to stop supporting a product that relies on your IP, then you must provide the means for individual users to support it themselves. That doesn't even have to mean giving up your IP; it could just mean designing from the beginning with the idea that a company won't last forever—but that'll never happen unless it's forced by law.
It would be straightforward to mandate such agreements be included with all IoT device purchases.
Ideally, they'd also be triggered by acquisitions or modifications to privacy policies. I can dream, right?
Why not just have it turn the machine off without intervention? This is what my last X dryers have done and generally done a decent job at it (for dryness, not for lint trap blockage). The key has been not overloading the appliance with too many heavy items and keeping the lint screen clear, both of which should just be habit for anybody who does laundry.
To do this I'd need to integrate the dryer sensor with a smart plug.
https://www.homedepot.com/p/Hotpoint-6-2-cu-ft-White-Electri...
https://www.homedepot.com/p/GE-7-2-cu-ft-White-Gas-Vented-Dr...
I thought they were required by law these days.
https://www.binglee.com.au/products/inalto-ivde45w-4-5kg-ven... This is a functional equivalent of what I've had in multiple rental properties in Australia. It has "Hot", "Hotter", and an egg timer.
Maybe not in US. In AU, for example, they do. Also I'm surprised to hear that. I would have expected them to be universal and work anywhere between 110 and 240V.
You could probably fit an adapter, but I imagine smart plugs in the US are built with the assumption they will never have to handle such a high load.
Not only are US 120V plugs different than 240V plugs, but the 120V/20A plugs are a superset of standard 120V/15A plugs--a 15A plug will fit a 20A receptacle but not the other way around. I do not believe code permits plugs for devices beyond this range--the really heavy stuff is always hardwired. I believe this is to prevent morons from trying to power their house from a generator with a suicide plug--forcing people to use a proper transfer switch that can't backfeed. I suspect this will change in the future due to how dirty small ICE engines are--as it stands things like electric mowers aren't very good because they can't get enough power to be good. Likewise, pressure washers--the high power versions are always ICE. Where I grew up you needed at least 4hp in a lawn mower--and you can't draw 4hp from a normal wall plug.
Electric heat pumps can be more cost effective than gas though, but there's details.
Based on my (obviously limited) set of data, I'd certainly argue against "most" dryers being gas-powered.
And as noted in another comment, even the cheapest dryers in the US have sensors (and have for decades).
https://patents.google.com/patent/US3402478A/en
https://patents.google.com/patent/US3758959A/en
As a power user, of course I prefer to connect to devices directly rather than through a sketchy cloud service.
But we've also got to acknowledge that the networked security cameras of 15 years ago regularly ended up making their videos public to anyone with a web browser - or becoming part of a DDOS botnet. So I can see why manufacturers aren't in a hurry to ask users to forward ports from their home firewalls any more.
https://community.home-assistant.io/t/clothes-dryer-automati...
It’s been ages since I had used a clothes line, so I don’t remember if that has ever been an issue.
My current guideline for purchasing:
- How long will the company last? - How likely is it for the company to sunset a product? - Are there fallbacks in place in case of company closure or server shutdown? This should be supplied by the company, and not cobbled together by users after losing server connection.
> In other words, SmartDry will become a tiny brick inside your dryer unless you're willing to procure a little ESP32 development board, load some code onto it, plug it in near your dryer, and set up your own alerts in your Home Assistant server.
Now I’ll admit there are cases that still count as “bricked” even if the device is recoverable with a Dremel and a JTAG debugger, but generally if you can fix it without specialist tools or software it’s not bricked.
In this case it sounds like the device still works fine and you just need to provide a server for it to talk to.
The actual sensor isn't bricked at all. The functionality most people bought it for is going away.
For example, I'm 99% sure Azure blobs are the underpinnings of Onedrive, and I'm pretty sure using just blobs, you could implement most of the functionality with just blobs, including stuff like push notifications.
They could offer a service where people could connect their smart devices to their Onedrive accounts, and it would set up a few service endpoints for their devices, and all hosting for these online features would be taken care of by the cloud provider. The load per user would be so low, it would generously fit into any 'free tier' these providers offer.
The infra is there, people are already used to paying for cloud storage, why isn't this a thing already
The term, and this idea about how ordinary a concept like that is now, made me think just now that modern English up to about 1990 (widespread adoption of the internet) may eventually become a new Old English.
There will be so many terms that have such different meanings and usage before and after a certain time, that they're almost different languages. And we're kind of in the middle of the transition in real time right now.
I was looking at a weather station recently that used some pretty innovative sensor hardware but then it only works with a cloud account. Literally if you want the data on your phone, you have connect the station to your internet connection, let it upload the data to the cloud, and then have your smartphone download it again. A thousand-mile trip for a 10 foot distance. F___ that noise!
When you live in an area with spotty internet service, this is utterly Epic Fail.
So it was an automatic no-go. No sale on that fine weather station. Hope you go broke!!
Instead I'm building a system from scratch with professional grade components that use 4-20 mA wire links that I'm connecting to an RPi. The cost for the components is definitely higher but they are the same as what the NWS uses. And I can chose whether to upload or share any of the data. And I will have access to MY DATA for as long as everything powers up and operates.
F___ companies that require cloud access to your own data
F___ engineers who design such abortions
It's not a shocking surprise that a device has a limited lifetime. People who buy $1000 smartphones, just to buy another one two years later, are complaining about this?
Just because something ends, doesn't mean it wasn't good while it lasted.