Companies that shut down connected systems should be required to release all device and server source code and hardware electrical schematics. I don't care if it'll be useless to the average consumer, I don't care if it'll hurt their "sensitive intellectual property" (frantically cobbled dog shit) portfolios, but it'll make them think before they roll things out they don't actually intend to support literally forever...
Honestly, the technological exit strategy should be integrated into every development plan. There are loads of options... Spin out, sell the technology, open source. Simply dropping support should almost never be an option. I daresay there are situations it makes sense, but I can't think of them.
Put the code / schematics into escrow, and require that the company or its successors must release the contents of the escrow in the event that the service ceases to be commercially viable. If another company wants to take over the assets and operate the service, they can, but they are obligated to release the contents of the escrow in the event they cease to operate the service.
Honestly, escrow is usually just one small part of the puzzle compared to the many SaaS/hosting services chained together to make the thing work. I’m sure most iOT startups that fail are no different.
In general, throwing some code over the wall for some commercially unsuccessful (or just something small time the creator got tired of) project is pretty much useless 99.99%+ of the time. As you say, for starters, it's probably only a piece of a fully-documented product.
How do you achieve that with company that is going bankrupt? Big corps, will spin up a branch responsible for that product and cut it off in case of the problems.
That must become illegal at some point. I spin up a branch, take a billion in off the plan appartments deposits then declare that branch bankrupt while my agency branch rakes it in.
It's a legal grey area, generally. It's not uncommon for companies to utilise special-purpose vehicles/strategy bankruptcies to insulate themselves from the financial fallout of their actions, though.
You can tax the products at the time of sale for expected externalities. Then you can still let people run reclamation/returns themselves and claim against tax and be responsible for overruns.
An example is the drinking container taxes in Norway, where single use containers are subjected to a graded tax depending on return rates. If the containers are part of approved returns schemes etc. and achieve 95%+ return rates you pay nothing. Below that the worse the return rates the higher the tax per unit (even if returned in other recycling). This gives an incentive to avoid containers that are not part of the recognised returns schemes. [Some returns schemes, like the bottle return scheme further applies a specific deposit amount per unit, so the amount added, or a portion of it, is recoverable for customers]
While ensuring payment for returns processing is important, getting the cost of externalities pushed to the point of sale is also important, because it punishes you with higher prices for products that are hard to get decent return rates for.
And that's my suspicion about why open source often seems two steps behind: when it's in the open, you have to do it right and can't hide the bodies. Doing things right is hugely slower at first. It's only in the long run when doing it right pays off. Steady incremental progress outruns the FCDS junk when that inevitably reaps the whirlwind of the technical debt incurred during the initial breakneck development.
Sometimes the FCDS is a consultancy deal from the outset and you cannot maintain the steaming bowl of canine leavings in front of you at all. Ever wonder why every now and again a serviceable, but clunky and unmaintained shovelware-like app (parking, alarm system, transport ticketing, that kind of thing) gets replaced with one that's basically the same but a different icon and a handful of trimmed features? New consultants just got done frantically cobbling it from scratch. This is clear in the app landscape, but hardware is often a similar consultancy thing.
Wait, wait, wait... Behind? If you'd asked me, I'd have said that's why open source always seems to be a step ahead! You're under no obligation to do things 'right', and worse is better. But free software means that you start from scratch much less often. You can take what already works and use it again and again in new products, and so can everyone else. Reusing the core of Linux in a billion small devices seems like a good idea to me.
Open source is usually inferior at first, because it's often one guy, or a small group of people doing it after work and at weekends, not a big company piling gobs of money into a full-time team. You usually only get that initial rush in open source when it's a big company driving the project in the first place (LLVM, say).
Linux is one of the great examples where FOSS has built up momentum and outrun several commercial competitors. Ditto for git (which actually was a bit "FC" in the first place). GCC and LLVM have between them stomped a few commercial compilers.
If you want examples of open source lagging the commercial equivalent: FreeCAD Vs SolidWorks, KiCAD (which is beginning to edge out some competitors: Eagle just went down) vs Altium, GIMP Vs Photoshop, KDEnlive Vs Premiere, Octave Vs MATLAB.
Not to say they're not amazing projects, or that they don't have their own share of technical debt, but they'll all have to play the long game to reach the overtake (and they're all decades into the race, so it's a slog).
Not being into home automation myself, I can't really comment on open source in this space, but the commercial churn puts the FOSS approach in a good place for the software side. Hardware, not so much.
Freedom as in Beer is starting to mean a lot more to people now. That could be considered a major feature at this point. I started telling people about my jellyfin setup and suddenly "normie" friends that had no obsession with archiving and self deployment like I do are asking for help setting up their NAS and finding cool things to deploy in https://github.com/awesome-selfhosted/awesome-selfhosted
This has always been the end game for me of open source, beyond all the other benefits: offer tools so good for free that when someone tries to charge people for it, they'll be met with a resounding "why would I pay you for that when I can get this for free?"
Not to mention open source software basically reigns supreme in Venezuela, India, Philippines, other places without high capital in global currencies.
> If you want examples of open source lagging the commercial equivalent...
In all these examples, the missing part is not the technical prowess, but the optimized math which makes these software useful, which is a serious secret sauce.
Evaluating numerical calculations correctly under all circumstances is a kind of dark art.
Develop better algorithms, and you'll be in the business of disrupting them. See, OpenFOAM.
I disagree. Having tried most of the products on that list and their competitors, the problem isn't the math. They generally get that right enough for most applications.
The big problem is the user experience. FreeCAD has single-handedly soured me on 3D modeling in general because it is awful to work with. There are a dozen incompatible ways to do anything, and they all have weird undocumented limitations. And rather than getting basic 3D modeling working properly, they add a whole architecture workbench.
We saw this with Blender a few years ago. The thing which made it go from "toy product" to "viable commercial alternative" was a huge UI overhaul in 2.7x. Something similar happened in KiCad 6. Both Blender and KiCad have attracted significant commercial backing since then and are slowly starting to become serious alternatives to their proprietary competitors.
100% agree, but how do we actually make that happen? I've joined RealThunder's (major FreeCAD dev's) patreon, but other than that I'm just resigned to a long wait.
The two I’ve seen is 1. a company needs it and doesn’t care so they just make a huge investment in the tooling that they open source or 2. someone with a lot of free time/independently wealthy just puts their head down and goes at it.
Get someone who cares to throw their weight behind it. You need a way to get developers to a) care about it, and b) do a lot of non-fun work. A Patreon isn't going to work because then the devs just end up doing more of the things they enjoy.
It seems like getting a commercial company behind it is pretty much the only way to do it. You just can't tell an independent developer that the UI of their pet project sucks and they need to spend six months rewriting it from scratch. Heck, getting them to agree that the UI sucks is pretty much impossible already! Stuff like this is a lot easier when your primary incentive is "being paid a salary" rather than "having fun with my pet project".
> not a big company piling gobs of money into a full-time team
Think this is a bit of a straw man, to be honest. I've done turns at AWS, Microsoft, Dropbox, and now a quant fund. Most of the big-impact projects at all of these places have been bottom-up initiatives, not top-down -- they've been things started by one or a couple people, wedged into spare time, because the existing ecosystem had gaps. Even things that turn into full-fledged, public-facing services usually start quite small. A former teammate at AWS was part of the team who launched Mac on EC2 -- he was the third engineer on the team, and he started on the project as a part-time engineer on loan from another team. In my experience, this is how things are usually built -- even for things that go on to be part of the re:Invent keynote.
Maybe for some software, but rarely for anything involving hardware unless you're dummying it up on an Arduino. The hardware platforms I've worked would handily incinerate my entire household net worth in a year or so and relentlessly consume full time attention.
Even a part time engineer on loan is probably more time than some dude's weekend.
But i think part of it is how fast things can be scaled up. Corporations can go from the tiny to large scale fast of things look good. It takes a lot of time for open source to build momentum.
Frantically-cobbled: from April 3 2005, announced April 6 and becoming self-hosting on April 7 by Linus Torvalds to replace BitKeeper after they dumped Linux (the public announcement of BitKeeper scrapping the free version was April 5, but the writing was on the wall for a month or two before that).
Evidence of the rapid initial development can still be seen in the famously...quirky...CLI syntax.
> You usually only get that initial rush in open source when it's a big company driving the project in the first place (LLVM, say).
LLVM began as a research platform at UIUC. In terms of public record, Apple only began putting money into it around 2005 (primarily by hiring Lattner). It might have received other corporate funding before and during that time that isn’t well documented, but I don’t think it’s accurate to say that it was driven by a big company from the get-go.
Yes, and if it wasn't for that, it would still be research platform like many others that predated it in compiler development, like Amsterdam Compiler Toolkit for one such example.
No disagreement there. My point was just that LLVM is not an "initial rush" case, to use the parent's language. It's a success story for the "small group of people" model.
Did the rush come before or after the huge (I imagine millions per year for a team of Apple devs working on what was to become a core technology pillar) infusion of cash and warm bodies? It was a fairly quiet and incremental research project and then Apple zoomed it to the stars (and Clang wasn't always even guaranteed to be open sourced: that came later, two years after the Clang project started in 2005).
Of course, this is actually one of the more unique cases, because it was both well-funded and open-sourced in the end, and, though developed rapidly, is actually a decent project, technically. As opposed to the kind of behind-closed-door too-embarrassed-to-publish shitware that's probably powering things like Hives.
All the examples where FOSS is ahead, is in tools by and for programmers. They're acceptable, maybe with an advantage due to cost or lack of cloud subscription stuff, but otherwise not great for the rest.
Most everything I like about Linux seems to be driven by the red hat types more than the general community.
Home automation software is pretty trivial, HW is what matters, and until very recently the HW was tied to the proprietary software. Plus the software can't rely on you being the product like commercial stuff, so you have to pay for easy remote access on home assistant.
The other issue is even though the functionality is simple, two important features are good UI, and not randomly breaking on updates, which FOSS isn't great at.
I use FreeCAD, KDEnlive, and others all the time, because the commercial stuff just isn't practical, if you're not making 40k a year or having a company that is tech focused enough to have a license.
I'm glad it exists, I'll continue to contribute in GitHub if I see an easy to fix bug..., I'll probably keep using my DIY automation system because of a few specific features, But I'd really rather FOSS do stuff a tiny bit more like commercial does.
Only for the software. Most commercial companies in the home automation space are slinging hardware and/or hoovering data. Even Apple is fundamentally selling iPhones here.
Open source hardware is a thing but it's genuinely hard and it's a place where the open source iterative building on a rising base idiom doesn't bring such long term dividends because real manufacturing costs real money, and lots of it, and even the best open source electronic designs eventually go out of date when parts EOL.
Hardware vendors are reluctant to provide access to run open firmware on their devices because not only then it's a hop, skip and a jump to a clone, inevitably cheaper since the design phase costs a lot, but also they'll lose the device lock-in (e.g. Apple) or app-based data siphon (everyone else).
Well yes. But that isn't an option for most people.
I suppose it goes to the heart of what you want open source/free software to be.
If you're happy for it to be a niche thing only for 'geeks' that pov is legitimate. If you want the wider world using it then such answers aren't reasonable.
I don't want it to be anything. Free software is what it is. People work on what they want to work on. Some people make software for others to use. Most, I think, make software for themselves to use. Then you have people making commercial software that they can sell support contracts for. And there are people making free software so they can write academic papers.
No, your framing is incorrect. Free software is, by its very nature, open for anyone to use or contribute to. If you don't know how to program, then you can learn. You can pay someone to do it, or convince someone to do it some other way.
This is how all human relations operate. It is not specific to free software.
And as with my job example, the labour market is free for anyone to join.
But stopping at that and taking a laissez-faire attitude is a position with implications.
Just like failing to switch the junction and allowing the train to run over X people.
You can't say "I didn't do anything" you didn't, and X people are now dead.
The labour market is not a free market, but a highly regulated one in most places. There is a strong argument that government intervention is much more a cause of than a solution to unemployment. That's a separate political debate of course, but it isn't as simple as "laissez-faire attitude bad". Every position you take has implications of one kind or another. It's a trade-off, and not an easy answer.
>Just like failing to switch the junction and allowing the train to run over X people. You can't say "I didn't do anything" you didn't, and X people are now dead.
Our law has an established distinction between acts and omissions. You are responsible for your acts. If you act in a way that harms another, you are (prima facie) responsible. If you omit to do an act, and that act may have saved another person from harm, you are (prima facie) not responsible.
So actually yes, if you fail to switch a junction and a train runs over 1000 people you are (prima facie) not responsible. Of course if your job is to supervise the junction, it would be. Or if you took responsibility for doing so, or if you make a representation that you would do it and others reasonably relied upon it.
Lawyers and judges didn't just pull the act/omission distinction out of thin air. It's the way that the law has evolved because it actually handles situations that arise in practice in a just and workable way. All that is to say, I disagree that you are necessarily responsible for someone's death if they are tied to a railway track and they're run over by a train, and you stood by and didn't save them. You may be in some circumstances, but in reality nobody is going to actually just stand there for no reason. They will omit to do so for some other reason: they might be paralysed by fear, they might have some other conflicting responsibility, etc. The responsible party is the guy that tied them to the train tracks in the first place. Let's focus our attention on that.
The reason this came up is that you seem to think that someone is doing something wrong if they don't use their programming talents to add features that non-programmers have expressed a desire to have added to free software. Maybe that isn't actually your position and I have misunderstood it. But if it is, I disagree strongly. If someone is capable of adding CMYK support to GIMP but doesn't do so despite the clamouring of the masses, I do not think they're doing anything wrong. There are always other things they could be doing with their time, whether it's lying in a hammock listening to Pink Floyd, or sitting on a beach reading a novel, or sitting on a couch eating fried food and watching daytime TV, or, of course, contributing in some other way to some sort of free software project that they're more interested in and which is more useful to them. I don't think any person has a responsibility to write free software just because some other random people would benefit from him or her doing so.
That's why I don't like the framing of (paraphrasing) "if you want it to just be for geeks, that's fine, but that's what you're saying". It's not that I want free software to just be for geeks. I don't have any preference over who should or shouldn't use free software. It's not my place to say who should or shouldn't use it, because that's up to them. But I don't think that anyone has any kind of obligation, be it moral or legal or otherwise, to do so. I think people that choose to write software they won't enjoy themselves but which will benefit others are probably doing a good thing. It's like planting a tree that will take so long to grow that you will never enjoy its shade, to absolutely butcher that famous phrase. But I think ... I was about to say 'there are plenty of legitimate reasons not to do so'. But it's not even that. I don'...
This is a extremely reasonable attitude to have, as a FOSS maintainer.
For someone who doesn't write C at a "GIMP maintainer" standard (maybe I'm not even a programmer and just want to make pictures) it will be far more cost effective to chuck a few hundred dollars at Adobe than spend person-years learning and working enough to get that kind of feature in, or funding the development. And then while doing that learn C++ to a standard to get your topological naming rework into FreeCAD via OpenCASCADE. And so on. Amortised over the everyone, it's a tiny cost relative to what Adobe costs each person, but that's the open source curse: you can't easily scrape up 10,000 $10 donations to fund development work worth millions in the end (to other people).
I don't really have a solution other then grind onwards.
A large number of C programmers can write code at that level. But over the many, many years that this supposed problem has existed, not one has stepped up to do it. Many other problems have been solved with open source software, many of them much harder than this. So maybe the fact it hasn't been done indicates it's not really that important.
The problem is the intersection between C programmers competent and tenacious enough to drive it though, and those who really care about CMYK image editing. Maybe both large populations, but the overlap? And also the population of people who make that their number one priority and not, say, working on LibreOffice.
The overlap is large enough for GIMP and Krita and Inkscape and dozens of other free software tools to exist. There's overlap between competent C programmers and accountants (GNUCash, ledger, hledger), and between competent C programmers and lawyers, and between competent C programmers and typesetters, and between competent C programmers and desktop publishers, and between competent C programmers and the target audience for every other open source tool.
It seems easier to assume that it just isn't a big deal. I don't really understand the issue myself, but it seems like the simplest explanation. It would be strange for this to be the one random area where 'competent C or C++ programmer' and 'domain expert' doesn't have an overlap.
I mean maybe there's a pattern here with some other areas that lack good free software? I don't know. People complain about other 'creative' software lacking good free software tools but I think that's usually due to a lack of familiarity rather than a lack of quality in the software itself. Creatives seem to be more bothered by UI they're not used to than others are.
Open Source often takes the shortest available route from A to B, to an MVP which solves a problem for a technical user who doesn't mind duct tape and configuration.
The fit and finish required for a non-technical user is frequently neglected, because that's not a itch that an OS contributor needs to scratch. Design is especially neglected.
And then there's the long term. OS solutions are not immune to bit rot, forked projects, people getting bored and another solution gaining traction. It can start requiring a higher and higher technical bar (e.g. compiling from source, or applying patches to library dependencies) to keep something old running, when the new stuff requires functionality only available in more recent hardware.
I suspect your perspective is based on using well-established open source projects that have a long history and are popular enough to have reached critical mass.
The reality is that most open source projects are playing catch-up, especially in the hardware space. Linux still has many issues with newer GPU drivers (binary blobs instead of actually OSS ones), and even a semi-popular project like OpenWRT doesn’t have WiFi 6E support, even though devices have been out for a while.
I think I almost shit myself laughing at this post and the one it replies to.
FCDS: from first-ever internet mention (according to the googles) to internet culture icon/acronym in under 60 minutes. Bravo, HN, you still have what it takes.
Or any kind of product requiring a server, there's a lot of games heavily dependent on online features and when the online platform closes you lose of lot of features or even sometimes the best part of the game. You can still play Mario Kart Wii online today because someone saw the issue coming and reverse engineered the server code but it's not a viable solution
> "sensitive intellectual property" (frantically cobbled dog shit) portfolios
That phrase always makes me think of crazy person that thinks you're coveting his shopping cart full of precious things. Or maybe as my dad alluded to, classified information usually means stuff that makes us look like lame and petty buffoons.
I'm not saying they should not be required, as I have not thought about this deeply, but in whichever way I imagine this playing out, it seems likely to lead to situations in which people are using some device with a lot of security vulnerabilities that were made much easier to find and exploit once the code was opened up.
An unmaintained device was always going to be a security problem; at least with open code and a way to install it there's a way to patch problems after the vendor stops providing fixes.
I'm on the same page, but was wondering if there would be any concern about liability on the part of the company that released the code (mandatorily, as suggested).
I agree, but suspect most such systems are using licences from other closed-source projects, IP from 3rd party agencies they commissioned, and stuff that may well have cross-licencing and other crap going on. I doubt a company like BG would do more than just glue together other people's stuff.
That's kinda what I'm getting at though. If you regulate that everything is going to be escrowed with a contingency (by which I mean guarantee) that it'll end up being forcibly released under GPL/whatever, you suddenly need to use components that are compatible. Short term this would be an e-waste bill, long term it would help strengthen the FOSS community and by extension the quality, speed of development, and general performance of the whole industry.
Smart homes are a corporate fever dream, adding extra layers of complexity, technical debt, servicing costs, for zero benefit. Wall switches and dials are already user friendly, the only exception being for people with disabilities.
Thankfully the Hive thermostat is Zigbee and works offline (I have one setup).
It's not mentioned in the article so I'm unsure on its longevity but when the time comes the majority of users however will not be able to do this and might make them worthless to those people, more ewaste.
My recent experience with trying to get a working led dimmer combination has almost pushed me to switching to a phillips hue. So not all dials are user friendly. I did notice though, that almost all high-end light bulbs are smart bulbs. Getting name-brand good quality dimmable leds is difficult (except for phillips warm-glow, but the compatibility horror of those bulbs is what started me on this journey)
Philips Hue has so far been an exception to mediocre smart home products. We had Hue at home for ages now, it always works, no lights got deprecated or any of that nonsense, and even though they have a cloud component, it works fine without.
The only downside for me is that I don’t use the smart part. So for me they are just good lights. My wife does use the app/widgets, etc.
All the Hue stuff is actually zigbee under the hood and is totally compatible with any zigbee Hub not just the Phillips one. If you want to do a deep dive one weekend and go completely local control go check out home assistant
It mostly works. I’ve had it hiccuping on me, when the switches and lights got unresponsive for a few minutes, for no obvious reason. And a number of times software got out of sync and told me the lights are off when they are actually on, etc.
But it’s definitely better than all the competitors I’ve seen.
I have some 11 year old Philips LED bulbs that were early on the LED development and crazy expensive; maybe $60 each. These aren’t Hue, but plain recessed type screw-in bulbs. These have worked consistently great with dimmer switches with no strobing, humming or low voltage issues. I can’t find good bulbs like that anymore. Quality and reliability went down as cost went down and competition went up.
I have Hue bulbs in one room, but lag in control of wall switch communicating to adjust lights is terrible compared to direct dimmer control.
I've been using Hue for a decade or so now, I can't fault it.
I don't use any of the connected stuff and just manage it with the app and local API.
I decided to take the hub out of my network a couple of weeks ago and use the Home Assistant SkyConnect to manage by bulbs, switches, light strips etc, it's been a less than stellar experience. The hue hub works very well, it's signal strength is excellent, I had it sitting inside a closed rack and never had an issue, the SkyConnect loses signal to a bulb 2 feet away if you breath on it.
Would highly, highly recommend hue to anyone. That said, prices have gone up and bulb quality dropped since Signify took over (many of my bulbs are metal and glass) the newer ones are plastic and plastic for more money, and though none of the original bulbs have failed to date, one of the newer ones randomly cuts out.
The reason I want to avoid Hue is not so much because of network security, but because of simplicity. I've seen a whole lot of packet parsing and RF stuff. I want none of that to be required for my lamp to turn on when I flip a switch.
However, it seems like dimmable leds might almost be doing packet parsing over very weird PWM-modulated AC signals in order for the dimmer and LED to cooperate. I like the simplicity of using the power-line from switch to LED, but I hate the ad-hoc nature of the accidental communication, so perhaps it makes sense to upgrade to ZigBee, accepting the complication of RF and network addressing in return for having well specified protocols.
One thing I'd say is that, I've switched to using ZigBee binding so I don't rely on the hub, and even if my automation is down, the buttons bound to the lights in each room can still be used to control them. I lose access to extra features like colour changing etc in that scenario but honestly, after about a decade of experience, I'd suggest not bothering with colour bulbs except for novelty or exceptional cases where you want to add colour accents; 99% of the time in 99% of my home, my family and I use the bulbs in a shade of white.
Additionally, Hue added a feature some time ago to configure the "state" the bulbs comes back on when power is cut, so it's quite feasible to continue to use a physical wall switch for normal use and ZigBee other times, with the caveat that by cutting the power to the bulbs you're messing with your ZigBee mesh, so YMMV.
Ultimately, aside from my children who enjoy the option to colour the lights in their rooms, I could get away with smart switches instead of smart bulbs, and I've just bought some Hue-compatible ones for the living room, landing and similar where the chandeliers have large numbers (5+) of bulbs that would be expensive to replace with Hue and where very little advantage would be taken of the features other than dimming.
Dimmable LEDs is basically impossible to do well without a smart bulb because of how they work. Hue is among the best, but even that is making me tempted to replace everything with ancient incandescents and oil lamps.
Smart devices are the closest thing to driving me to be a Luddite because nothing is remotely as annoying as having to reboot and update your lightbulb.
People who buy into smarthomes are corporate fever dreams. It allows the companies to focus on lock-in for profits at the expense of everything else (ie stuff you mentioned)
> However, it's quite an old product and if you went for a Hive system from 2016
These things are part of the house, like an appliance. 7 years is nothing in terms of a building, it's ridiculous to disable products in such a short time.
Exactly. This is something that should be protected in consumer law. There should be a minimum (10 years? more?) period in which companies are required to support smart home devices and keep the cloud services that support them operational.
There is an e-waste issue here as well as a consumer rights one. The old analog boiler control/timer box and thermostat in my house look they were installed in the 1980s. A bit yellowed and beat up but they still work just fine!
If you did it right you could make it a marketing sell point - list the number of guaranteed days left in the UI and then start telling you how much you “beat” the system once that’s past.
In the UK this is protected by consumer law under the terms of the The Consumer Rights Act 2015. Anyone who bought their devices direct from British Gas should write to them saying that they want repair, replacement with equivalent devices that will work, or a full refund.
There is no time limit, only "reasonable", which for something that is fixed into your house, 20 years seems reasonable to me. You could settle for prorating the 20 year life, so if you've had it 5 years then they will refund 75% or something.
British Gas is easily big enough to absorb the losses, and has a lot of bad press at the moment because of the profit it has been making. It would set a good example for future cases.
Ah but to play devil's advocate, consumer rights only kicks in when you "take ownership" of something. I am sure that in the hive contract somewhere it says that you do not own it and just have a limited-use license so you are not technically a "consumer" of their service but some sort of licensed "service installation" or some other legal bullshit or sleight of hand that means they have zero obligations.
Best Buy gave up on a bunch of their IoT switches, etc, and sent me Best Buy gift cards for the full purchase price.
Amusingly, because they’re HomeKit compatible, they continue to work because Apple required HomeKit to work without some weird manufacturer cloud system.
> There should be a minimum (10 years? more?) period in which companies are required to support smart home devices and keep the cloud services that support them operational
10 years (or whatever) from time the product comes to market or from the time they stop selling it?
If it from the time they stop selling it that would probably lead to a lot of things not being sold for long. Manufacturers will want to get the clock started as soon as possible to minimize the amount of time they end up having to back port security fixes to libraries that upstream has stopped supporting.
This could result in manufacturers discontinuing things a year to two after introducing them, replacing them with newer products that do most of what the old did but probably also incorporate newer features.
That would be annoying for consumers. Say I buy some internet connected home control and security system, but don't buy all the accessories it works with. Then 5 years later I want to add say a camera. But the cameras designed for my system were discontinued 4 years earlier, and the cameras the company now sells are 3 generations ahead of that and require features or interfaces in the base system that aren't in mine.
On the other hand if it is from the time the product comes to market that will also probably discourage manufactures from keeping a product available for a long time. Consumers might be reluctant to buy something that has been for sale several years knowing that it only has a few years of mandatory support left.
Probably what you need are two limits, one requiring a certain number of years from the time the product is introduced, and the other requiring a certain smaller number of years after the product is discontinued.
> "10 years (or whatever) from time the product comes to market or from the time they stop selling it?"
Consumer law is typically based on the date of purchase. So, 10 years from the date of purchase, for example.
Although, as pointed out by another commenter, UK consumers effectively have these rights already, as they have the right to return (to the retailer) a product that is no longer functioning after a reasonable amount of time for repair, replacement, or refund.
Don't see why being fixed in place in a house means that a £60 control chip should be required to last longer than a £1000 smartphone, especially since the latter doesn't offer you a discounted replacement when the OEM kills most of your functionality with incompatible OS updates locking you out of appstores and then the non-removable battery swells up and kills it.
do the same with phones then. unlock them when they stop being supported so you can run your own software.
id accept the same for hive if there was an open ecosystem to move it to.
point is, if companies want to retain control of something they actually need to maintain it. if they dont want to maintain it, they need to give up control. they cant have it both ways.
No, which is why I wouldn't have every single switch and dimmer in my house controlled by individual computer with custom Internet connected firmware.
We're talking about a single inexpensive thermostat that's being replaced at half price after 5-7 years, I'm definitely not joking when I'm suggesting that this might actually have less impact on consumer wallets or ewaste pile than very expensive consumer devices whose OEMs' business model revolves around being replaced biannually
If you do things reasonably and only have a smart thermostat, 7 years isn't awful, but I'd still spend a little effort seeing if I can switch instead of buying v2. It'd really bite if you went all-in on smart home stuff.
The price is irrelevant. What matters is how long that kind of device can be reasonably expected to last. Dumb wall sockets cost about £5 and last >20 years. Smart sockets should last a comparable time, or come with a warning.
Because cellphones are traditionally something you replace every 3-4 years it's expected that they only last that long, and you are fully aware of that when you are buying it.
I'll note that people have been conditioned to think that phones need replacing every 3--4 years. Used to be that a wired-in landline would last a quarter-century or more, and even wireless phonesets could be expected to last a decade.
With smartphones, the principle drivers of technical obsolescence are ultimately communications standards (e.g., phase-out of 3G communications), chip speeds, and the ever-increasing resource requirements of connected services (apps or websites).
Tools for dealing with phone & SMS spam and fraud would be another principle concern.
But really, there's no reason otherwise that a handset from a decade or two ago couldn't work, and there are many who look on classic Nokia handsets with fondness and/or mourning.
If you're manufacturing something that's going to go in a house...
...and that means it should be expected to last 20 years...
...then maybe just don't use the cheapest part you can, and instead actually engineer it to last 20 years?
You're talking about it like the manufacturer has no choice but to make it as cheap as possible and hey, if it fails, it fails—but that's not true at all. If there were a requirement for something to be more durable, then the manufacturer would be both able and obligated to follow it. Unfortunately, I don't believe there is such a requirement as things stand.
Customers need to stop buying "connected" smart home devices, but more importantly, legislation needs to prevent this bullshit;
I still remember the story of Sonos intentionally bricking up-traded speakers that were perfectly functional so they couldn't be sold on, and some other company (I now forget who) bricking their remote with a firmware update because they no longer wanted to support it.
Force anyone who wants to release connected products like this to put the code in Escrow and if they go belly up without a plan, it goes public.
Communities across all disciplines have reverse engineered defunct products since forever, imagine what we can do with the source.
It's insanity. I get really irate about this sort of thing because it's so anti-consumer and environmental impacts be damned, that's a problem for little people.
When are engineers going to understand that the bullshit corporations buy the politicians that write the laws? They don't need to buy all of them, just slightly more than half. They're fine with some rebel types who submit noble-sounding bills for the EFF or whoever to rally around that never go anywhere.
> When are engineers going to understand that the bullshit corporations buy the politicians that write the laws?
Never. Because as engineers we're optimists, if I just resigned to this fact, there'd be no hope left. It's not that we don't understand, it's that we don't accept.
Many of us also aren't capitalists, so we can't believe that the "free market" will resolve it either.
We build things, somewhat in isolation to the "human problem", some of us (I speak for myself) because technology, machines, are better? easier? more consistent? predictable? reliable? quantifiable? than humans.
We can solve technical problems, we can't solve humans being completly greedy, selfish pricks; those corporations buying those politicians are just humans.
Engineering is _easy_, compared to people and people problems.
I think real engineers are actually pessimists. Software developers are optimists; that's most of the difference between software developers and engineers.
The optimist believes the glass is half full. The pessimist believes the glass is half empty. The engineer says the glass is twice as large as it need be.
I feel like, in some circles, the adjective "smart" has gained a highly negative connotation when referring to anything other than a human, as it necessarily implies some sort of resistance to control and ownership.
"Smart" is just about synonymous with "connected" these days, and any product with the originating company in its feedback loop represents a risk that the product will start working against the user in some way either immediately or after some honeymoon period.
A smart TV means it can run apps basically. There is really nothing smart about that...
It would be smart if it would mute itself during commercials, for example.
All these 'Smart' things should work for me instead of for the vendor really.
Concerning the connected stuff: I ordered a new robot vacuum the other day and checked if it could integrate with Home Assistant and fell into the 'local push/poll' category of integrations [0].
I wonder if the tainting of "smart" over the last few years ever made Kwikset regret naming their entirely mechanical user rekeyable lock system "Smart Key"? Those came out in 2008.
For those who haven't seen those locks, they look like normal door locks, with a normal key, and are used just like normal locks (insert key, turn clockwise to lock, counterclockwise to unlock), except that there is a small hole next to the keyway.
If you put a working key into one, turn it 90° clockwise, insert a simple tool Kwikset provides into that small hole far enough to press a button at the bottom of the hole, and remove that tool the lock will release the key. You can then, being careful to not rotate the cylinder, pull out the key and put a new different key in.
The lock then rekeys itself to that new key, and you can then turn the cylinder counterclockwise 180° to verify this. Then turn the cylinder back 90° clockwise and remove the key and you are done.
This is quite convenient. I recently replaced my locks with these. Before I had one brand of knob and deadbolt on the front door, a different brand on of level and deadbolt on the back door, and a third brand of knob on the garage door. Each of these locks was on a different key, so that was 5 keys in total. And I'd long ago lost the garage key and the back door lever key.
I bought a Kwikset Smart Key comb knob and deadbolt for the front door, a combo level and deadbolt for the backdoor, and a knob for the garage. So that was 3 different keys. Then I picked the key that had the most variation in the depth of cuts, and went around to the other locks with the rekey tool and rekeyed them all to that key. I also picked up a couple of Kwikset padlocks that use this system and rekeyed them to that key. Later I added a deadbolt to the garage and put it on that key too.
I also had a sliding door lock. There doesn't seem to be a Kwikset Smart Key lock in the right form factor for that, but I noticed that the sliding door lock (a PrimeLine E2000) uses Kwickset KW1 keys, which is the same key blank that Smart Key uses. I was able to take that apart and rekey it to match the rest of the locks.
My key ring now only needs my car key and one Kwikset key. It's a lot less jingly now. If I'm not going to be using my car I just carry a loose Kwikset key in my pocket.
This was inevitable, it's always been a secondary thing for British Gas.
I worked on the predecessor to Hive back in about '08-'10, the engineering consultancy I worked for designed a version of the "hub" and a number of their security and energy monitoring sensors.
It was the first thing I worked on that made it out into consumers hands (v2 of the hub). I remember one day buying a copy of Wired magazine, and it had an article about them. Seeing a product I had designed in that magazine was a really memorable moment.
Back then they were not part of British Gas but a Cambridge UK startup called AlertMe. A really good bunch of people with a vision for the smart home. The plan was clearly to start with the security products, then energy monitoring was a secondary market. However after getting a big contract with British Gas to supply monitoring to all their customers, they pivoted almost completely to that - and ultimately they were acquired by them.
As I said, they were a really interesting team, their CTO was a former early Acorn employee who had lead RiscOS.
Deceptive writing. It's just the outdated (and presumably insecure) Hive Nano 1, that is being deactivated early, and for a good reason probably. The author should explain the necessity to switch off insecure products and motivate people to stay up to date in terms of IT security. Instead, the text makes it sound like some villain corporation is trying to screw their customers for no good reason at all.
So, for a smart home where some parts are actually, well "hidden" in the building you find if feasible to just upgrade?
I mean, how much could a house cost, right?
Stuff like this should be forbidden. Non-experts simply cannot fathom the consequences of their decisions.
Nobody would expect parts of their house need complete replacement after just 7 years (if they were in good condition in the first place and there was no catastrophe)
This is not even talking about e-waste and the environmental issues that arise from more high tech garbage
Are you just guessing that it's insecure and for some reason that can't be fixed with a software update? Your link doesn't mention that as far as I can see.
Nano 1 was on sale as recently as 2016. It's only 7 years old - there is literally no reason why they can't keep it working. And if they say they can't they are literally incompetent and their tech department needs to be replaced.
I’m super late to the “smart home party” but the few smart things I do own are all HomeKit certified, as a requirement for that is that the device needs to work as long as it has local connection to a HomeKit hub (like an iPhone). Internet not required, let alone a connection to a vendors servers.
Before I go all in, I’ll need to verify how well this works in practice but at least it should prevent a scenario like the one linked here…
Withought getting in the weeds this is generally the right approach. The most important thing is that devices should work without an internet connection.
I use Home Assistant as my "hub" so that's another requirement but that's more personal choice.
Unless they’re really established (and even then) you’re about right.
Turn the Internet off after installing any smart device and see what happens - that’ll be how it works when they shut down the servers. If it don’t work at least somewhat, send it back.
If you’re just looking for some basic remote on/off functionality for your heating and hot water, with built in timers, not dependent on any cloud service or app, may I wholeheartedly recommend Shelly relay products. Simple web interface and easy to physically wire up. MQTT if you need to wire them up to HA or Graphite ( power meter varieties) or whatever.
I went all in on the Hive system (heating, cameras, etc) and have only negative things to say about the entire company. Even for their heating systems, do not use them. Don't take my word for it. Go and look at the app reviews. The app is buggy, barely functional and their support is incompetent. Frequently, I'll turn the heating on and the app will (after about 10 seconds) show it's worked. I'll turn off my phone and wonder why the house is still cold. Open the app, and yeah, it didn't actually turn on.
The UX is very clunky too. They've tried to make this "app/phone" like dashboard, where you can drag and drop devices into pages. But it's completely broken (on Android, but looking at the reviews, I assume iPhone too). When scrolling/swiping the page just get's stuck half way through.
Every update makes the lag/delays and errors more frequent.
On the other hand - I've had a Hive thermostat for the last 5 years and it works flawlessly. Every time I start it from the app it works. Use it with a combi boiler here in UK and it works absolutely fine. I only have that one device from them so I can't comment on anything other than heating.
This is why I self-host. I don’t like the idea of buying something and having no control over how I use it, where my data goes, etc.
I strongly urge anyone with a tech interest to get a Raspberry Pi (now that they’re widely available again) and install HomeAssistant. Many of your devices will work through the many plugins available. If you can get your head around Python coroutines (… I can’t but I bodged it) you can implement your own.
Some devices will still need to send data to a central server to work, but you can often integrate their APIs until such a time that the service is discontinued. Then search out devices that are known to work with HomeAssistant.
> Many of your devices will work through the many plugins available
Alas, the Hive stuff doesn't because BG moved their auth to AWS Cognito and the Python module hasn't yet[1] managed to work out how to keep auth going[2]
[1] There is a workaround which involves getting a device token and using that instead but also I think I may have cracked the "refresh token" flow, just need to push a PR.
[2] You get 1h expiry on the access tokens with mandatory 2FA. Hilarity ensues.
I recently added a smart thermostat but picked a provider that works completely offline and has a good Home Assistant plugin (Wiser https://www.draytoncontrols.co.uk/products/Smart-Heating/Wis...). It was recommended to me by people in the trade that Wiser is the "cheap" model, but I'd much rather go for the budget option if it means I can go offline and it still works. Would never touch Nest, Hive, Tado etc, I want this to last longer than any of these companies are willing to provide support.
Anyone know of a dumber system that has thermostatic TRVs + per room scheduling? Ideally with an offline screen to administer?
Perhaps someone on here can help me. I'm one of the poor souls who has this device. I ordered the replacement hub late, it arrived while I was on holiday and I only just tried to install it. Unfortunately the migration instructions seem to require that the old device is on and functioning, which it isn't because it is past the 1st of August. I can't find instructions on how to connect the hive 1 thermostat to the new hive hub. If it turns out this is impossible and I need to rewire my boiler I will be very sad indeed.
Edit: a quick WhatsApp chat with a British Gas advisor sorted it. For anyone else reading this, the v1 thermostat is compatible with the latest hub
I really, really don't understand why these simple-ass microcontrollers are running proprietary software instead of something open source like ESPHome or Tasmota. There is no added value in the proprietary software running on an MCU with a sensor, it is all in the service that run on top of them.
It infuriates me that all these perfectly fine devices now are junk, because some clown decided that they needed to develop their own "platform".
156 comments
[ 0.23 ms ] story [ 202 ms ] threadAn example is the drinking container taxes in Norway, where single use containers are subjected to a graded tax depending on return rates. If the containers are part of approved returns schemes etc. and achieve 95%+ return rates you pay nothing. Below that the worse the return rates the higher the tax per unit (even if returned in other recycling). This gives an incentive to avoid containers that are not part of the recognised returns schemes. [Some returns schemes, like the bottle return scheme further applies a specific deposit amount per unit, so the amount added, or a portion of it, is recoverable for customers]
While ensuring payment for returns processing is important, getting the cost of externalities pushed to the point of sale is also important, because it punishes you with higher prices for products that are hard to get decent return rates for.
Judges have significant freedom to stop that kind of thing, it's not a default.
And that's my suspicion about why open source often seems two steps behind: when it's in the open, you have to do it right and can't hide the bodies. Doing things right is hugely slower at first. It's only in the long run when doing it right pays off. Steady incremental progress outruns the FCDS junk when that inevitably reaps the whirlwind of the technical debt incurred during the initial breakneck development.
Sometimes the FCDS is a consultancy deal from the outset and you cannot maintain the steaming bowl of canine leavings in front of you at all. Ever wonder why every now and again a serviceable, but clunky and unmaintained shovelware-like app (parking, alarm system, transport ticketing, that kind of thing) gets replaced with one that's basically the same but a different icon and a handful of trimmed features? New consultants just got done frantically cobbling it from scratch. This is clear in the app landscape, but hardware is often a similar consultancy thing.
Linux is one of the great examples where FOSS has built up momentum and outrun several commercial competitors. Ditto for git (which actually was a bit "FC" in the first place). GCC and LLVM have between them stomped a few commercial compilers.
If you want examples of open source lagging the commercial equivalent: FreeCAD Vs SolidWorks, KiCAD (which is beginning to edge out some competitors: Eagle just went down) vs Altium, GIMP Vs Photoshop, KDEnlive Vs Premiere, Octave Vs MATLAB.
Not to say they're not amazing projects, or that they don't have their own share of technical debt, but they'll all have to play the long game to reach the overtake (and they're all decades into the race, so it's a slog).
Not being into home automation myself, I can't really comment on open source in this space, but the commercial churn puts the FOSS approach in a good place for the software side. Hardware, not so much.
This has always been the end game for me of open source, beyond all the other benefits: offer tools so good for free that when someone tries to charge people for it, they'll be met with a resounding "why would I pay you for that when I can get this for free?"
Not to mention open source software basically reigns supreme in Venezuela, India, Philippines, other places without high capital in global currencies.
In all these examples, the missing part is not the technical prowess, but the optimized math which makes these software useful, which is a serious secret sauce.
Evaluating numerical calculations correctly under all circumstances is a kind of dark art.
Develop better algorithms, and you'll be in the business of disrupting them. See, OpenFOAM.
Disclosure: I did my Ph.D. on this subject.
The big problem is the user experience. FreeCAD has single-handedly soured me on 3D modeling in general because it is awful to work with. There are a dozen incompatible ways to do anything, and they all have weird undocumented limitations. And rather than getting basic 3D modeling working properly, they add a whole architecture workbench.
We saw this with Blender a few years ago. The thing which made it go from "toy product" to "viable commercial alternative" was a huge UI overhaul in 2.7x. Something similar happened in KiCad 6. Both Blender and KiCad have attracted significant commercial backing since then and are slowly starting to become serious alternatives to their proprietary competitors.
It seems like getting a commercial company behind it is pretty much the only way to do it. You just can't tell an independent developer that the UI of their pet project sucks and they need to spend six months rewriting it from scratch. Heck, getting them to agree that the UI sucks is pretty much impossible already! Stuff like this is a lot easier when your primary incentive is "being paid a salary" rather than "having fun with my pet project".
Think this is a bit of a straw man, to be honest. I've done turns at AWS, Microsoft, Dropbox, and now a quant fund. Most of the big-impact projects at all of these places have been bottom-up initiatives, not top-down -- they've been things started by one or a couple people, wedged into spare time, because the existing ecosystem had gaps. Even things that turn into full-fledged, public-facing services usually start quite small. A former teammate at AWS was part of the team who launched Mac on EC2 -- he was the third engineer on the team, and he started on the project as a part-time engineer on loan from another team. In my experience, this is how things are usually built -- even for things that go on to be part of the re:Invent keynote.
But i think part of it is how fast things can be scaled up. Corporations can go from the tiny to large scale fast of things look good. It takes a lot of time for open source to build momentum.
Evidence of the rapid initial development can still be seen in the famously...quirky...CLI syntax.
LLVM began as a research platform at UIUC. In terms of public record, Apple only began putting money into it around 2005 (primarily by hiring Lattner). It might have received other corporate funding before and during that time that isn’t well documented, but I don’t think it’s accurate to say that it was driven by a big company from the get-go.
Of course, this is actually one of the more unique cases, because it was both well-funded and open-sourced in the end, and, though developed rapidly, is actually a decent project, technically. As opposed to the kind of behind-closed-door too-embarrassed-to-publish shitware that's probably powering things like Hives.
Most everything I like about Linux seems to be driven by the red hat types more than the general community.
Home automation software is pretty trivial, HW is what matters, and until very recently the HW was tied to the proprietary software. Plus the software can't rely on you being the product like commercial stuff, so you have to pay for easy remote access on home assistant.
The other issue is even though the functionality is simple, two important features are good UI, and not randomly breaking on updates, which FOSS isn't great at.
I use FreeCAD, KDEnlive, and others all the time, because the commercial stuff just isn't practical, if you're not making 40k a year or having a company that is tech focused enough to have a license.
I'm glad it exists, I'll continue to contribute in GitHub if I see an easy to fix bug..., I'll probably keep using my DIY automation system because of a few specific features, But I'd really rather FOSS do stuff a tiny bit more like commercial does.
Open source hardware is a thing but it's genuinely hard and it's a place where the open source iterative building on a rising base idiom doesn't bring such long term dividends because real manufacturing costs real money, and lots of it, and even the best open source electronic designs eventually go out of date when parts EOL.
Hardware vendors are reluctant to provide access to run open firmware on their devices because not only then it's a hop, skip and a jump to a clone, inevitably cheaper since the design phase costs a lot, but also they'll lose the device lock-in (e.g. Apple) or app-based data siphon (everyone else).
I suppose it goes to the heart of what you want open source/free software to be.
If you're happy for it to be a niche thing only for 'geeks' that pov is legitimate. If you want the wider world using it then such answers aren't reasonable.
Your comment implicitly indicates a certain world view.
Like saying, you don't mind people not working, ignoring the fact there aren't any jobs or unemployment benefits.
Ie it's for those that have the ability to write their own code. Or people that have the ability to pay for the specific feature they want.
If you're happy with that, fair enough, but that is the implication.
This is how all human relations operate. It is not specific to free software.
But stopping at that and taking a laissez-faire attitude is a position with implications.
Just like failing to switch the junction and allowing the train to run over X people. You can't say "I didn't do anything" you didn't, and X people are now dead.
>Just like failing to switch the junction and allowing the train to run over X people. You can't say "I didn't do anything" you didn't, and X people are now dead.
Our law has an established distinction between acts and omissions. You are responsible for your acts. If you act in a way that harms another, you are (prima facie) responsible. If you omit to do an act, and that act may have saved another person from harm, you are (prima facie) not responsible.
So actually yes, if you fail to switch a junction and a train runs over 1000 people you are (prima facie) not responsible. Of course if your job is to supervise the junction, it would be. Or if you took responsibility for doing so, or if you make a representation that you would do it and others reasonably relied upon it.
Lawyers and judges didn't just pull the act/omission distinction out of thin air. It's the way that the law has evolved because it actually handles situations that arise in practice in a just and workable way. All that is to say, I disagree that you are necessarily responsible for someone's death if they are tied to a railway track and they're run over by a train, and you stood by and didn't save them. You may be in some circumstances, but in reality nobody is going to actually just stand there for no reason. They will omit to do so for some other reason: they might be paralysed by fear, they might have some other conflicting responsibility, etc. The responsible party is the guy that tied them to the train tracks in the first place. Let's focus our attention on that.
The reason this came up is that you seem to think that someone is doing something wrong if they don't use their programming talents to add features that non-programmers have expressed a desire to have added to free software. Maybe that isn't actually your position and I have misunderstood it. But if it is, I disagree strongly. If someone is capable of adding CMYK support to GIMP but doesn't do so despite the clamouring of the masses, I do not think they're doing anything wrong. There are always other things they could be doing with their time, whether it's lying in a hammock listening to Pink Floyd, or sitting on a beach reading a novel, or sitting on a couch eating fried food and watching daytime TV, or, of course, contributing in some other way to some sort of free software project that they're more interested in and which is more useful to them. I don't think any person has a responsibility to write free software just because some other random people would benefit from him or her doing so.
That's why I don't like the framing of (paraphrasing) "if you want it to just be for geeks, that's fine, but that's what you're saying". It's not that I want free software to just be for geeks. I don't have any preference over who should or shouldn't use free software. It's not my place to say who should or shouldn't use it, because that's up to them. But I don't think that anyone has any kind of obligation, be it moral or legal or otherwise, to do so. I think people that choose to write software they won't enjoy themselves but which will benefit others are probably doing a good thing. It's like planting a tree that will take so long to grow that you will never enjoy its shade, to absolutely butcher that famous phrase. But I think ... I was about to say 'there are plenty of legitimate reasons not to do so'. But it's not even that. I don'...
For someone who doesn't write C at a "GIMP maintainer" standard (maybe I'm not even a programmer and just want to make pictures) it will be far more cost effective to chuck a few hundred dollars at Adobe than spend person-years learning and working enough to get that kind of feature in, or funding the development. And then while doing that learn C++ to a standard to get your topological naming rework into FreeCAD via OpenCASCADE. And so on. Amortised over the everyone, it's a tiny cost relative to what Adobe costs each person, but that's the open source curse: you can't easily scrape up 10,000 $10 donations to fund development work worth millions in the end (to other people).
I don't really have a solution other then grind onwards.
It seems easier to assume that it just isn't a big deal. I don't really understand the issue myself, but it seems like the simplest explanation. It would be strange for this to be the one random area where 'competent C or C++ programmer' and 'domain expert' doesn't have an overlap.
I mean maybe there's a pattern here with some other areas that lack good free software? I don't know. People complain about other 'creative' software lacking good free software tools but I think that's usually due to a lack of familiarity rather than a lack of quality in the software itself. Creatives seem to be more bothered by UI they're not used to than others are.
The fit and finish required for a non-technical user is frequently neglected, because that's not a itch that an OS contributor needs to scratch. Design is especially neglected.
And then there's the long term. OS solutions are not immune to bit rot, forked projects, people getting bored and another solution gaining traction. It can start requiring a higher and higher technical bar (e.g. compiling from source, or applying patches to library dependencies) to keep something old running, when the new stuff requires functionality only available in more recent hardware.
The reality is that most open source projects are playing catch-up, especially in the hardware space. Linux still has many issues with newer GPU drivers (binary blobs instead of actually OSS ones), and even a semi-popular project like OpenWRT doesn’t have WiFi 6E support, even though devices have been out for a while.
FCDS: from first-ever internet mention (according to the googles) to internet culture icon/acronym in under 60 minutes. Bravo, HN, you still have what it takes.
That phrase always makes me think of crazy person that thinks you're coveting his shopping cart full of precious things. Or maybe as my dad alluded to, classified information usually means stuff that makes us look like lame and petty buffoons.
[1] https://en.wikipedia.org/wiki/Source_code_escrow
It's not mentioned in the article so I'm unsure on its longevity but when the time comes the majority of users however will not be able to do this and might make them worthless to those people, more ewaste.
I don't suppose you have any useful "here's how to connect using Zigbee" tutorials? Would be handy to skip the whole Hive API shenanigans...
The same way certain systems work on 80' hardware, they do a job and only a job without internet connection.
The only downside for me is that I don’t use the smart part. So for me they are just good lights. My wife does use the app/widgets, etc.
It mostly works. I’ve had it hiccuping on me, when the switches and lights got unresponsive for a few minutes, for no obvious reason. And a number of times software got out of sync and told me the lights are off when they are actually on, etc.
But it’s definitely better than all the competitors I’ve seen.
I don't use any of the connected stuff and just manage it with the app and local API.
I decided to take the hub out of my network a couple of weeks ago and use the Home Assistant SkyConnect to manage by bulbs, switches, light strips etc, it's been a less than stellar experience. The hue hub works very well, it's signal strength is excellent, I had it sitting inside a closed rack and never had an issue, the SkyConnect loses signal to a bulb 2 feet away if you breath on it.
Would highly, highly recommend hue to anyone. That said, prices have gone up and bulb quality dropped since Signify took over (many of my bulbs are metal and glass) the newer ones are plastic and plastic for more money, and though none of the original bulbs have failed to date, one of the newer ones randomly cuts out.
However, it seems like dimmable leds might almost be doing packet parsing over very weird PWM-modulated AC signals in order for the dimmer and LED to cooperate. I like the simplicity of using the power-line from switch to LED, but I hate the ad-hoc nature of the accidental communication, so perhaps it makes sense to upgrade to ZigBee, accepting the complication of RF and network addressing in return for having well specified protocols.
One thing I'd say is that, I've switched to using ZigBee binding so I don't rely on the hub, and even if my automation is down, the buttons bound to the lights in each room can still be used to control them. I lose access to extra features like colour changing etc in that scenario but honestly, after about a decade of experience, I'd suggest not bothering with colour bulbs except for novelty or exceptional cases where you want to add colour accents; 99% of the time in 99% of my home, my family and I use the bulbs in a shade of white.
Additionally, Hue added a feature some time ago to configure the "state" the bulbs comes back on when power is cut, so it's quite feasible to continue to use a physical wall switch for normal use and ZigBee other times, with the caveat that by cutting the power to the bulbs you're messing with your ZigBee mesh, so YMMV.
Ultimately, aside from my children who enjoy the option to colour the lights in their rooms, I could get away with smart switches instead of smart bulbs, and I've just bought some Hue-compatible ones for the living room, landing and similar where the chandeliers have large numbers (5+) of bulbs that would be expensive to replace with Hue and where very little advantage would be taken of the features other than dimming.
Smart devices are the closest thing to driving me to be a Luddite because nothing is remotely as annoying as having to reboot and update your lightbulb.
These things are part of the house, like an appliance. 7 years is nothing in terms of a building, it's ridiculous to disable products in such a short time.
There is an e-waste issue here as well as a consumer rights one. The old analog boiler control/timer box and thermostat in my house look they were installed in the 1980s. A bit yellowed and beat up but they still work just fine!
There is no time limit, only "reasonable", which for something that is fixed into your house, 20 years seems reasonable to me. You could settle for prorating the 20 year life, so if you've had it 5 years then they will refund 75% or something.
British Gas is easily big enough to absorb the losses, and has a lot of bad press at the moment because of the profit it has been making. It would set a good example for future cases.
https://www.which.co.uk/consumer-rights/regulation/consumer-...
Where's that quoted from? Legislation? How do you think that consumer rights apply to electricity supply and phone contracts, for example?
I have no knowledge of whether consumer rights apply to Hive but I don't understand this "take ownership" principle that you're proposing here.
> KEY INFORMATION > > Taking ownership > The clock starts ticking from the date you take ownership of the goods
No they don’t. Consumer rights law in the UK applies to services too.
Amusingly, because they’re HomeKit compatible, they continue to work because Apple required HomeKit to work without some weird manufacturer cloud system.
And more importantly you can easily buy wire-compatible replacement controllers to this day.
10 years (or whatever) from time the product comes to market or from the time they stop selling it?
If it from the time they stop selling it that would probably lead to a lot of things not being sold for long. Manufacturers will want to get the clock started as soon as possible to minimize the amount of time they end up having to back port security fixes to libraries that upstream has stopped supporting.
This could result in manufacturers discontinuing things a year to two after introducing them, replacing them with newer products that do most of what the old did but probably also incorporate newer features.
That would be annoying for consumers. Say I buy some internet connected home control and security system, but don't buy all the accessories it works with. Then 5 years later I want to add say a camera. But the cameras designed for my system were discontinued 4 years earlier, and the cameras the company now sells are 3 generations ahead of that and require features or interfaces in the base system that aren't in mine.
On the other hand if it is from the time the product comes to market that will also probably discourage manufactures from keeping a product available for a long time. Consumers might be reluctant to buy something that has been for sale several years knowing that it only has a few years of mandatory support left.
Probably what you need are two limits, one requiring a certain number of years from the time the product is introduced, and the other requiring a certain smaller number of years after the product is discontinued.
Consumer law is typically based on the date of purchase. So, 10 years from the date of purchase, for example.
Although, as pointed out by another commenter, UK consumers effectively have these rights already, as they have the right to return (to the retailer) a product that is no longer functioning after a reasonable amount of time for repair, replacement, or refund.
id accept the same for hive if there was an open ecosystem to move it to.
point is, if companies want to retain control of something they actually need to maintain it. if they dont want to maintain it, they need to give up control. they cant have it both ways.
Huh? You're joking, right?
You want to go around and replace all switches, dimmers, etc every other year in a home?
We're talking about a single inexpensive thermostat that's being replaced at half price after 5-7 years, I'm definitely not joking when I'm suggesting that this might actually have less impact on consumer wallets or ewaste pile than very expensive consumer devices whose OEMs' business model revolves around being replaced biannually
smart != internet connected. KNX, Zigbee, Z-Wave falls into this category as well, and those are local to the house, unless there's a gateway.
Anyway... as others said it before, a non-smart switch lasts decades. Smart ones should as well.
This isn't necessary.
Because cellphones are traditionally something you replace every 3-4 years it's expected that they only last that long, and you are fully aware of that when you are buying it.
With smartphones, the principle drivers of technical obsolescence are ultimately communications standards (e.g., phase-out of 3G communications), chip speeds, and the ever-increasing resource requirements of connected services (apps or websites).
Tools for dealing with phone & SMS spam and fraud would be another principle concern.
But really, there's no reason otherwise that a handset from a decade or two ago couldn't work, and there are many who look on classic Nokia handsets with fondness and/or mourning.
...and that means it should be expected to last 20 years...
...then maybe just don't use the cheapest part you can, and instead actually engineer it to last 20 years?
You're talking about it like the manufacturer has no choice but to make it as cheap as possible and hey, if it fails, it fails—but that's not true at all. If there were a requirement for something to be more durable, then the manufacturer would be both able and obligated to follow it. Unfortunately, I don't believe there is such a requirement as things stand.
I still remember the story of Sonos intentionally bricking up-traded speakers that were perfectly functional so they couldn't be sold on, and some other company (I now forget who) bricking their remote with a firmware update because they no longer wanted to support it.
Force anyone who wants to release connected products like this to put the code in Escrow and if they go belly up without a plan, it goes public.
Communities across all disciplines have reverse engineered defunct products since forever, imagine what we can do with the source.
Note that Google turned it off worldwide, all because the USPTA is a bunch of fucking morons.
Many of them on HN too.
When are engineers going to understand that the bullshit corporations buy the politicians that write the laws? They don't need to buy all of them, just slightly more than half. They're fine with some rebel types who submit noble-sounding bills for the EFF or whoever to rally around that never go anywhere.
Never. Because as engineers we're optimists, if I just resigned to this fact, there'd be no hope left. It's not that we don't understand, it's that we don't accept.
Many of us also aren't capitalists, so we can't believe that the "free market" will resolve it either.
We build things, somewhat in isolation to the "human problem", some of us (I speak for myself) because technology, machines, are better? easier? more consistent? predictable? reliable? quantifiable? than humans.
We can solve technical problems, we can't solve humans being completly greedy, selfish pricks; those corporations buying those politicians are just humans.
Engineering is _easy_, compared to people and people problems.
I think real engineers are actually pessimists. Software developers are optimists; that's most of the difference between software developers and engineers.
The optimist believes the glass is half full. The pessimist believes the glass is half empty. The engineer says the glass is twice as large as it need be.
"Smart" is just about synonymous with "connected" these days, and any product with the originating company in its feedback loop represents a risk that the product will start working against the user in some way either immediately or after some honeymoon period.
A smart TV means it can run apps basically. There is really nothing smart about that... It would be smart if it would mute itself during commercials, for example. All these 'Smart' things should work for me instead of for the vendor really.
Concerning the connected stuff: I ordered a new robot vacuum the other day and checked if it could integrate with Home Assistant and fell into the 'local push/poll' category of integrations [0].
[0] https://www.home-assistant.io/blog/2016/02/12/classifying-th...
For those who haven't seen those locks, they look like normal door locks, with a normal key, and are used just like normal locks (insert key, turn clockwise to lock, counterclockwise to unlock), except that there is a small hole next to the keyway.
If you put a working key into one, turn it 90° clockwise, insert a simple tool Kwikset provides into that small hole far enough to press a button at the bottom of the hole, and remove that tool the lock will release the key. You can then, being careful to not rotate the cylinder, pull out the key and put a new different key in.
The lock then rekeys itself to that new key, and you can then turn the cylinder counterclockwise 180° to verify this. Then turn the cylinder back 90° clockwise and remove the key and you are done.
This is quite convenient. I recently replaced my locks with these. Before I had one brand of knob and deadbolt on the front door, a different brand on of level and deadbolt on the back door, and a third brand of knob on the garage door. Each of these locks was on a different key, so that was 5 keys in total. And I'd long ago lost the garage key and the back door lever key.
I bought a Kwikset Smart Key comb knob and deadbolt for the front door, a combo level and deadbolt for the backdoor, and a knob for the garage. So that was 3 different keys. Then I picked the key that had the most variation in the depth of cuts, and went around to the other locks with the rekey tool and rekeyed them all to that key. I also picked up a couple of Kwikset padlocks that use this system and rekeyed them to that key. Later I added a deadbolt to the garage and put it on that key too.
I also had a sliding door lock. There doesn't seem to be a Kwikset Smart Key lock in the right form factor for that, but I noticed that the sliding door lock (a PrimeLine E2000) uses Kwickset KW1 keys, which is the same key blank that Smart Key uses. I was able to take that apart and rekey it to match the rest of the locks.
My key ring now only needs my car key and one Kwikset key. It's a lot less jingly now. If I'm not going to be using my car I just carry a loose Kwikset key in my pocket.
I worked on the predecessor to Hive back in about '08-'10, the engineering consultancy I worked for designed a version of the "hub" and a number of their security and energy monitoring sensors.
It was the first thing I worked on that made it out into consumers hands (v2 of the hub). I remember one day buying a copy of Wired magazine, and it had an article about them. Seeing a product I had designed in that magazine was a really memorable moment.
Back then they were not part of British Gas but a Cambridge UK startup called AlertMe. A really good bunch of people with a vision for the smart home. The plan was clearly to start with the security products, then energy monitoring was a secondary market. However after getting a big contract with British Gas to supply monitoring to all their customers, they pivoted almost completely to that - and ultimately they were acquired by them.
As I said, they were a really interesting team, their CTO was a former early Acorn employee who had lead RiscOS.
He was profiled a few years back, as discussed here: https://news.ycombinator.com/item?id=24226740
Shame big corp came in and destroyed it, I hope everyone was paid hansomly in the takeover at least.
That might be a mixed blessing; horses are expensive![0]
[0] https://en.wikipedia.org/wiki/Hansom_cab
https://en.m.wikipedia.org/wiki/Boxer_(Animal_Farm)
https://community.hivehome.com/s/article/Nano1-shutdown-Step...
I mean, how much could a house cost, right?
Stuff like this should be forbidden. Non-experts simply cannot fathom the consequences of their decisions.
Nobody would expect parts of their house need complete replacement after just 7 years (if they were in good condition in the first place and there was no catastrophe)
This is not even talking about e-waste and the environmental issues that arise from more high tech garbage
Before I go all in, I’ll need to verify how well this works in practice but at least it should prevent a scenario like the one linked here…
I use Home Assistant as my "hub" so that's another requirement but that's more personal choice.
Turn the Internet off after installing any smart device and see what happens - that’ll be how it works when they shut down the servers. If it don’t work at least somewhat, send it back.
The UX is very clunky too. They've tried to make this "app/phone" like dashboard, where you can drag and drop devices into pages. But it's completely broken (on Android, but looking at the reviews, I assume iPhone too). When scrolling/swiping the page just get's stuck half way through.
Every update makes the lag/delays and errors more frequent.
I strongly urge anyone with a tech interest to get a Raspberry Pi (now that they’re widely available again) and install HomeAssistant. Many of your devices will work through the many plugins available. If you can get your head around Python coroutines (… I can’t but I bodged it) you can implement your own.
Some devices will still need to send data to a central server to work, but you can often integrate their APIs until such a time that the service is discontinued. Then search out devices that are known to work with HomeAssistant.
Alas, the Hive stuff doesn't because BG moved their auth to AWS Cognito and the Python module hasn't yet[1] managed to work out how to keep auth going[2]
[1] There is a workaround which involves getting a device token and using that instead but also I think I may have cracked the "refresh token" flow, just need to push a PR.
[2] You get 1h expiry on the access tokens with mandatory 2FA. Hilarity ensues.
It integrates well with Zigbee2MQTT and Home Assistant though, so I can still control it remotely.
Anyone know of a dumber system that has thermostatic TRVs + per room scheduling? Ideally with an offline screen to administer?
Edit: a quick WhatsApp chat with a British Gas advisor sorted it. For anyone else reading this, the v1 thermostat is compatible with the latest hub
It infuriates me that all these perfectly fine devices now are junk, because some clown decided that they needed to develop their own "platform".
Lol