Sonos: users must accept new privacy policy or devices may “cease to function”
zdnet.com
zdnet.com
On that subject: https://twitter.com/internetofshit/status/895881274614665217
Sonos sounds great, has nice features and "just works".
Perhaps not for long, if you do not accept their take-it-or-leave-it terms. Personally, I don't mind not accepting the TOS and simply leaving my software in the state that it currently is in, forever. My device is currently working acceptably, so I don't see why I need an update. But if they're saying it's going to eventually stop working, that's where any sane person should draw the line.
On the other hand, if I had to buy a $1000 device for the privilege of paying them $10/month, my anger level would probably depend on how many years I got utility out of that $1000 device before they shut down.
What part of the privacy policy do you specifically have an issue with?
Netflix and all other content providers collect "usage data" too, as does the devices these may run on such as "Roku" and other players.
I am trying to think of concrete examples of scenarios where actual harm would REALISTICALLY be done to individual consumers because of Sonos privacy policy.
2. Government compels Sonos to provide data on customers' listening histories.
3. Chilling effect, political implications, oppression.
"Alexa? Send details of my subversive plan to the government!"
"OK Google, upload the audio of the last 30mins discussion with my co-conspirators directly to PRISM!"
"Hey Siri? Which law enforcement agency needs to know about this surprise protest we just planned?"
As for the data collection you can't opt out of, it's super basic data like your ip and account registration info. How do you expect to use a cloud service if it doesn't know your ip?
> Sonos may modify or update this Statement when necessary to reflect customer feedback and changes in our Products and Service; so, please review it periodically. When we update this Statement, we will revise the “Last Updated” date at the top of the Statement. If there are material changes to the Statement or in how Sonos uses your personal data, we will notify you either by posting a notice of such changes before they take effect or by sending you a notification. We encourage you to periodically review this Statement to learn more about how Sonos is using and protecting your information. Your continued use of the Service or Products after any modification to this Statement will constitute your acceptance of such modification and updates.
I wouldn't have to just agree to the policy as you linked it today. One agrees to all future versions of the policy.
Perhaps, as with various other unreasonable or one-sided provisions, there should be a presumption under consumer protection laws that any such terms are invalid, that any substantial change in an agreement must have the customer's explicit, active consent, and that if the customer does not consent and this results in something they purchased no longer providing the expected benefit then the usual protections about broken purchases should apply. You'd have to be a bit careful, because sometimes these self-updating terms are used to deal with minor changes in the legal landscape that are formalities with no real impact on anyone, but that seems like the sort of thing a court could sensibly decide in the event of a dispute.
More importantly, I don't think consent is sufficient in this sort of situation. It has to be a real choice, so even if you choose not to consent to the change, you're not then losing out on something you already paid for or otherwise giving up your normal legal consumer rights as a consequence.
In a situation where the continued effective functioning of a purchased item is dependent on some other facility, and that facility and the related functionality can be turned off if you don't accept the manufacturer/vendor moving the goalposts retrospectively, I don't see much ethical or practical difference between using this dependency as a lever to force a customer to accept revisions in some other aspect of the deal and just doing the other thing without their consent anyway.
> Your continued use of the Service or Products after any
> modification to this Statement will constitute your
> acceptance of such modification and updates.
Is that even legal? I thought a contract required equal bargaining power, and explicit acceptance. If you have already purchased a good, and suddenly the terms change, I would think you would, at that point, have decidedly unequal bargaining power -- the company already has your money, and you either have to stop using the product, or agree to the terms. That doesn't seem "equal" at all.You only "own" something so long as you pay taxes (aka rents) for the right to do so, otherwise someone, usually the government, will take it from you. Even if you don't want to play along, you can only be somewhere public for so long until you're loitering and subject to fines (more rents), confiscation (again, more rents), or removal.
The whole system is pretty blatantly anti-competitive [1], detrimental to individuals [2][3], and rotting the system from the inside out [4][5].
1. https://fivethirtyeight.com/features/the-rent-seeking-is-too...
2. http://www.zerohedge.com/news/2017-06-22/us-economy-perverte...
3. https://hackernoon.com/rise-of-the-rent-seeker-how-the-subsc...
4. https://devinhelton.com/2013/04/14/rent-seeking-economy/
5. https://www.bloomberg.com/view/articles/2016-09-13/banks-and...
reviewing the rise of top-down city planning especially in recent modern cities like Brasilia (which systematically removed public space & spontaneous street interactions by design)
https://en.wikipedia.org/wiki/Seeing_Like_a_State https://en.wikipedia.org/wiki/Bras%C3%ADlia
Thanks, it sounds like something I'd love; ordering a copy!
Now let's talk about having control over my speakers.
If I don't mow my lawn or pick up my garbage (labour) then my homeowners' association or local government can initiate legal proceedings which may end in me not "owning" my home.
Look, I'm actually for taxes generally, whatever you want to think about that, as long as they provide equal value to the people. However, our current system is designed so that any local government can arbitrarily decide that they're incorporated and I'm subject to their laws and taxation. 99% of us work towards putting more into the system than we get back and many do so exponentially.
Last time I checked, eminent domain and asset forfeiture are still actual things that happen in 2017.
We may not live in an outright "feudal system" but many of us live in a system closer to one than not.
No? Don't pay your property tax. See how fast you get evicted from that home you "own" by the sheriff.
People who didn't pay the "taxes" were evicted and their possessions given to loyalists; sounds a lot like renting.
The core of the issue really was that Britain defended the American colonies from French invasion during the Seven Year's War and the British taxpayer bore the brunt of the taxation (I can't remember exact figures, but Britons were taxed much, much heavier than Americans, even after the stamp duties and tea taxes). After the war, Parliament came looking for payback, understandably. So Britain wasn't using taxation to control the colonies, it was looking to recoup the costs of defending them.
On a sidenote, this is actually quite interesting in today's context over NATO military spending, where now it is Britain freeloading on American military spending.
This is an entirely legitimate statement.
> ...now it is Britain freeloading on American military spending.
...but this you're really going to have to back up. Can you point to a specific example and explain why it's 'freeloading'?
My argument wasn't sound - I'm glad you provided historically accurate facts that were both enjoyable and informative!
Your sidenote is also interesting - and on another note (albeit silly) - have you heard of the AMC drama TURN: Washington's Spies? I've started watching it and have quite enjoyed it so far.
Too many hardware products are being tethered to opaque "updates" delivered via untrusted computer network, namely, "the Internet". Security issues of the delivery mechanism aside, purchasers of these products are given no options on whether and how they may choose to "update" the manufacturer's software, or replace the manufacturer's software with an open source alternative.
For example, if the manufacturer's software is closed source, the purchaser cannot verify what is in an "update". The manufacturer is under no legal obligation to tell her. As such, she may not be in a position to decide
[ ] YES, I could use this modification or
[ ] NO, this modification serves me no benefit
The modification may serve some benefit to the manufacture, e.g., collecting data, but no benefit to the purchaser.It may be bundled with some other modification that benefits the purchaser but the purchaser cannot choose to accept or reject the modifications individually.
Or the manufacturer may simply state a possibility that breeds fear among certain purchasers: that if purchaser does not accept all updates, then the hardware may "cease to function". Consideration by the purchaser of whether this is actually probable may be superseded by a perceived risk of not being able to use the hardware to its full capabilities.
(For example, a router company may inform the purchaser that only the router company's software will allow certain performance specificatons to be met. If the purchaser installs an open source alternative software, then she will not be able to obtain the same performance. Of course there may be other benefits to the open source alternative besides performance, but the perceived "loss" of not reaching the performance specs advertised by the manufacturer is enough to convince many purchasers to tether the router to the manufacturer's "updates".)
Purchasers own the hardware they purchase. It is the software, including "firmware", pre-installed by manufacturers that gives rise to these shenanigans. The less access and control, including choice of alternatives, that the purchaser has to such software, the better for the manufacturer who wishes to continue to exploit the sale of the hardware well-beyond the purchase date.
When the purchaser of hardware is forced to accept the manufacturer's software, then I tend to agree with the comment. Regardless of whether the purchaser owns the hardware, if the manufacture controls the software, e.g. the manufacturer can "update" it at will or break it at will for anyone who does not "update", then with all due respect I submit that the manufacture ultimately controls the hardware. Thought experiment: Apply the above view to every item of hardware you have ever purchased. Consider whether it will apply to the next item you purchase.
That is the only way
If capitalists are not going to let us abolish copyright, we should abuse it, not pretend it does not exist.
I would much prefer it if all of these services followed a publicly documented protocol and allowed me to configure which endpoint they talked to. I can't see this ever happening without regulation though.
They implicitly agree under duress, because their time is worth far too much to even read 10 pages of fine print for every product they use. I'm not being flippant; it bloody well is.
I'm not sure why courts consider that to hold any weight; to me it goes against the whole notion of informed consent in binding agreements. And to be fair, often the courts don't consider to hold any weight.
You don't need to read 10 pages of fine print for every product, because it's functionally the same fine print for every product.
If you don't like this, don't ever buy smart devices because they /all/ reserve the right to do this kind of stuff.
I avoid smart devices like the plague. If I want my speakers smart, I'll connect them to a suitable control unit. The only thing I want from my speakers is for them to translate electrical waves into pressure waves, the more accurate the better.
Another benefit is not having an audio player tethered in one place.
A third benefit is being able to change the audio settings from any one of a half dozen devices in my house.
https://www.theverge.com/2016/4/4/11362928/google-nest-revol...
That's the future. The dude with the 80's hifi setup knows what's up, in my opinion.
Another benefit is not having an audio player tethered in one place.
A third benefit is being able to change the audio settings from any one of a half dozen devices in my house.
Okay, but that's just wireless tech. Why do they need to be constantly connected to the internet and phoning home?
It seems odd then that you have to actually buy the device before you can even read the terms. I guess the idea is that you can always return it if you don't agree?
My willingness to have a poorer experience is unfortunately not as strong as some FOSS enthusiasts. I'm happy to relinquish some control to Apple so long as they continue to provide me with positive experiences (AirPlay, for instance). Fortunately everyone is piling into the "works with everything else" crowd, so you don't really have to opt for something particularly power-user-oriented anymore. HomeKit doesn't fall too short of an ISY-99i based system, for instance.
I can only imagine if I wasn't excited by FOSS and linux to begin with. There is zero hope of escaping the cloud-trap for someone who doesn't have the knowledge or willingness.
If they brick my devices, I will only be half upset.
Pretty crappy default configuration on Sonos' part (using old default cost values), if you ask me.
Instructions I followed which solved everything: https://bsteiner.info/articles/disabling-sonos-wifi
Sonos's spanning tree implementation is problematic (and anachronistic: last I looked, it was using classic STP not RSTP). In a similar situation I've watched them flap away merrily, spitting constant topology change BPDUs at each other through both interfaces. It's not a pretty sight.
They seem to me to be totally in denial about it, judging by their forums. Lots of blame-the-customer and blame-someone-else's-device, and very little (or no) technical documentation. I think they are a speaker manufacturer first, a streaming audio interface second, and a networking company a very distant third.
I've had one tiny but annoying bug: Occasionally a podcast will show 0:00 to 0:00 on the scrubber and nothing -- not rebooting the whole system and my phone -- will fix it. It can be frustrating to be unable to scrub to the 40 minute mark on a podcast where I've already listened to half of it.
I have a Sonos in every room of my house and I've owned them since the very first generation. Sonos has been extremely good about updating the software. The current software still works on the very first hardware, with all the functionality save for a single feature, room-correcting equalization, that requires the newer DSP. This company is the gold standard of ongoing software support for consumer goods and this article is trying to spin the situation in just the perfect way to make the Internet commentariat explode.
[edit] apparently the models have a line-in socket? In that case, I agree, them turning into regular speakers is better than bricking. However, I still think the article is relevant as it highlights the problem of ownership and terms of sale with IoT devices. (The conditions under which you bought the device can change arbitrarily after the sale)
If you have any google account or android device, they already have literally orders of magnitude more info on your life than Sonos would ever be able to collect. Not to mention Sonos has said it doesn't ever sell the data, AND you can opt out of usage info. Not seeing the problem here.
That said last I saw their software was just as buggy as its always been. Not worth the price IMO.
Answer: Forget the protection afforded by the state. Protect yourself. Blacklist the scum manufacturers, warn your acquaintances.
What other suggestions.
How many gates would it take to have a pure hardware decoder for ATSC?
A microcontroller-based dishwasher timer is much cheaper to design and manufacture, and wears out more slowly under many conditions, than a mechanical timer.
I wouldn't be surprised if a micontroller could wring another 1% efficiency out of your refrigerator vs. an analog or mechanical controller.
A microcontroller inside the speaker could perform some digital signal processing to account for the speaker's frequency response, so that the receiver doesn't have to know any details about the speakers.
Now, I do see little reason to make any of the non-TV appliances you mention reliant upon network connectivity or anything outside themselves apart from power and physical consumables (soap, water, etc.). Ideally, the television itself would be split into a screen with only HDMI in, and a small box or dongle that performs ATSC reception, decoding, and any smart TV functionality. The television itself should be something like a Chromecast.
We could summarize it as "needlessly Internet connected, reporting to some endpoint in the cloud outside of your control, which is now a point of failure".
Indeed my dishwasher is likely more reliable, and my fridge can get more efficient with a microcontroller. But the 2017 version is likely to suddenly stop working is someone in Korea forgets to renew a certificate (or worse, become part of a botnet, or act as an breach in the security of your network like the aquarium in Vegas I read about lately). Most of the times, the "added value" of this new dependency is harvesting your data for profit (which won't get passed down to you as a better price, don't worry).
Now it's, "if you don't kn... oh f*ck it, you're the product!"
Just because you pay for the product, doesn't mean you can be the product too.
And let's be real: what you've written should be written as "privacy in perpetuity or maybe part of your money back years later after a class-action lawsuit, maybe, if it isn't just store credit." That, as much as anything, is why consumers caring is attenuated.
The whole 'if consumers care' thing is kinda bullshit. They simply don't have the time or ability to get into the details. It's even more of a problem when companies train users to just accept changes to agreements, so they can keep using a physical device that sits in their home.
It's that simple. Stop putting up with closed non trivial firmware and these sorts of problems go away.
I love the idea of smart devices, but only as long as the software is Free and Open. I really don't understand people who think situations like this are acceptable.
a bug that's been around forever and they plain refuse to fix it.
Selecting sonos as output device from spotify must be done a few times before it sustains (although that might be a spotify problem).
1. Have guests over.
2. Want to play music.
3. Get out phone.
4. Launch Sonos app.
5. "Hi, we think now would be a great time to update the firmware on your Sonos devices. We could have done this at any time during the last two weeks since you used any of your Sonos devices, but your time is worthless to us, so we'll make you babysit the update for the next 5 minutes. Hope you weren't launching the app because you wanted to play music right now."
At a minimum they should have the decency to launch the app-store review flow during this forced delay so that they'd have some public recognition of the supreme annoyance this process causes.
Also, since we're getting some new data protection laws across the EU next year that are heavily weighted against organisations that process data, someone who is upset by this could probably annoy Sonos by doing things like accepting the policy, getting whatever updates they need to keep their device running, and then immediately giving notice that they withdraw their consent for processing any personal data.
In the long term, I think consumer education may be the best solution to the problem of normal devices depending on fragile remote facilities. It is illegal to sell cigarettes in my country without showing very obvious warnings about how smoking is harmful to your health. Perhaps it's time that any software or electronic devices that rely on someone else's servers or remote hardware to function have to have 1/3 of their packaging (including 1/3 of the total screen area for any web pages or app purchase screens) used for a big statement that this may cause the software or device to stop functioning at any time, or to give an actionable guarantee about how long such support will be provided for as a minimum, and that any advertising for such things has to describe the "purchase" in terms of being a rental rather than buying something.
Typically under consumer protection laws in places like Europe, a product that ceases to work properly unreasonably quickly after it is purchased will trigger certain rights for the purchaser, such as getting the device repaired or replaced so it works again, or getting a full or partial refund as compensation.
Importantly, there is typically no exception to this principle just because the product in question involves software or depends on remote services. In fact, the EU strengthened consumer protections on digital purchases not so long ago, so the direction of the legal currents is very clear.
I suppose you could say there has been some sort of informal understanding that because it's so hard to write software with no bugs or security flaws, it's reasonable for the suppliers of products with software components to supply software updates post-sale to keep things working properly without falling foul of the basic "it no longer works" trap if users don't then apply those updates. But the sorts of shenanigans we've seen recently with software developers bundling unrelated and possibly unwanted changes in with their bug fixes and security patches are stretching good faith here, and at some point I suspect something has got to give.
The same principle applies to products with dependencies on remote services that don't necessarily last as long as the product itself, whether that's a server for a multiplayer game, or a DRM scheme with remote authorisation, or a provider of audio data that makes a speaker system useful.
a) the high price of Sonos equipment
b) the HomePod coming out sometime in the near future
c) other competitors that are starting to be more competitive (eg. "smart speakers", bluetooth speakers, etc)
The principle here is whether the customer is getting what they reasonably believed they were paying for, or whether some part of that benefit is being lost prematurely because of the actions (or inactions) of the device manufacturer/vendor.
To continue with your own example, whether it matters that Spotify changes an endpoint probably depends on how the product was originally presented. If it was advertised as offering access to a variety of specific services and some or all of those services become unavailable earlier than the customer might reasonably have expected, then clearly part of the value that the customer was paying for has been lost.
This is something other manufacturers could learn from. It seems these days most products are launched by marketing teams: fire and forget, the moment the product is out the door, all software development ceases and it never gets updated.
I do hope they reconsider the new privacy policy, though. It's worrying.
You'll have to pry my physical wall switches and copper wires from the cold, dead hands.
I was about to buy 2 Play1's and a Connect to do a stereo install in a room where I don't want to run speaker cable. I had researched wackier home-built solutions and was going to give up and go for the Sonos. Now I'm once again considering wackier home-built stuff like 3 raspberry pi's attached to line-in and two dumb powered speakers.
1) 4 to 1 HDMI splitter with audio out: https://www.aliexpress.com/item/HDSW4-HDMI-Switch-4X1-With-A...
2) Connect all your sources through HDMI (your PC, your TV box, a Chromecast, a raspberry, etc). There are versions with more inputs if you need that. That one has a nice simple remote.
3) Connect the HDMI splitter audio output to a RF audio transmitter: https://www.altoproaudio.com/products/stealth-wireless
4) Use some powered studio monitors with one of the RF receivers each (there are plenty to choose from)
This way if you're streaming audio+video everything syncs up because the RF link adds very little latency (unlike the Sonos that adds 70ms). The chromecast gives you easy streaming of audio and video from all your apps and if you are only streaming audio you can just turn off your TV/projector and the audio still works fine. The speakers can be placed anywhere there's power. It doesn't have the fancy multi-room and 5.1 possibilities of the Sonos nor the good industrial design.
I would be shocked if the commercial offerings were significantly better than most open source stuff (except for iphone interoperability...)
Sonos Just Works — some people (like me) care about that deeply. I have better things to do with my life than fiddle with audio systems. If your hobby is building these kinds of systems, then of course it makes perfect sense to roll your own. But if it isn't, I wouldn't waste my time.
Plug the RPis into decent stereos (not a powered speaker unless you are some kind of rabid animal) and you've got a good setup that can be controlled from anywhere in the house. Multiple stereos can playback in sync and be controlled from the same UI.
Not a solution, but snapcast is a useful part of the puzzle for the actual streaming and multiroom sync.
And they might. Which would be more frustrating to a user - their Sonos turning into regular speakers, or arbitrarily and randomly not functioning as expected?
Imagine the difference between your phone suddenly bricking itself, or random refusing to dial certain numbers.
I consider this motivation for DIY.
I specifically buy devices that aren't bound to the Internet and I've found a good amount.
(The base station does connect to a WAN server for remote control, but you could always block that at your firewall. The system works entirely fine in local mode.)
I understand that Sonos can stream to multiple "zones" simultaneously but besides the occasion of a house party how often is this necessary?
This news to me is just another reason for me to never buy one.
The idea that you get the same thing from a raspberry pi, an android, and some hacks is strictly for those valuing their own time at zero.
I have it set to wake me to the local streaming radio, so that's a nice feature, but of course it could be replicated.
I can throw my phone into a passing cement mixer and the music keeps playing in the house, and then I can open the controller on my laptop and hit "next track". In this way it's fairly analogous to the Spotify controller where I can change the music from any device in the house, but the "playing" happens on the device itself; it's not actually streamed from my phone or computer.
Also, until sometime in the past couple of years Bluetooth was godawful unusably unreliable. It's gotten REALLY good lately though, and suddenly I love it.
Let's say a new Itunes update comes along, this requires the user to opt into a privacy policy, if there happens to have been a change in said policy since the last update, then accepting the new conditions is required in order to install the update? Same for an update in browsers, iOS, Android, ...
I am not in favour, just confused as to why this specific case is singled out? Especially since not updating critical software (Operating systems, browsers, et cetera) seems to have far more serious consequences than w/ a speaker?
Declining update and choosing to live without new features vs. ceasing to function.
Sounds to me like maybe the bluetooth syncing with a recent smartphone or something of that nature might not work if the speaker didn't receive a certain update, which I consider a possibility. The "cease to function" seems like a nightmarish-description of such. It's not like "not updating" will magically brick the device.
Edit: my point being that not updating software is not about not enjoying certain features, it's in many cases being incredibly vulnerable to certain types of exploits, in the case of an OS or browser a far more serious consequence, is it not?
The right path for Sonos would be to open-source the kernel and hardware drivers, then either keep it up to date with a base feature set or budget $100K or so a year to run a bounty-based community contribution program. That will end the complaints that Sonos risked bricking hundreds of millions of dollars' worth of consumer hardware so its marketing department could harvest email addresses. Not everyone would like that choice, but it would be a better choice than the gradual decay of perfectly fine (and expensive) hardware that should otherwise last for decades.
Unfortunately the economics of service-oriented consumer hardware is all screwed up these days. Should buying a proprietary connected geegaw for a single up-front fee entitle me to eternal updates? In what non-pyramid-scheme economic model does that work? Yet in our world where we're still weaning ourselves from the dot-com years, that's typically the expectation.
* No need to have a second device present to stream music. That is, I can start up music with the controller on my phone, and go to the store, but the music still plays since all the streaming is handled by the sonos speaker.
* No need to pair your phone/computer to the device every time you come home. Just get on the wifi, and you're set.
Those are the big ones for me, and honestly they're huge.
On the flip side, the biggest pain point is that you're beholden to whatever music services they include in their controller, and you have to use their controller to play music. So while they support Apple Music, I can't use the Music app on my phone to play it. I have to drill through the Sonos app and find what I want. It works pretty well, but it's annoying. Also, I can't stream Audible books, or Overcast podcasts since they aren't supported by the Sonos controller.
Also, activating Bluetooth is as frictionless as activating WiFi.
Sonos allows my wife and I to simply play whatever music we want, wherever we want, with no hassle. And it sounds great.
Not in my experience. Bluetooth is functional, but it's hard to call it better.
* Connecting 2 people (phones) to one device is hardly supported.
* Those 2 people can't really manage a single playlist without special effort.
* Auto connecting to a bluetooth speaker when I walk in the house doesn't happen. It does with wifi.
* Connecting to the bluetooth speaker typically kicks off any other connection
* The issue of walking too far from the bluetooth speaker with the playing device still exists.
* Ever run the microwave while streaming audio over bluetooth? It doesn't work.
In my experience, bluetooth is better at one thing: streaming whatever audio from whatever app I want.
Oh my, no. Activating bluetooth is so frustrating that Apple built a special chip to make it easier. That Sonos work over Wifi and not Bluetooth is the primary selling point.
(Disclaimer: I used to work for Sonos.)
In particular, non-technical users appear to be vastly more comfortable with the Sonos system.
Quick review.
iOS users have to use their spotify app which is lousy.
Google Play users can cast to it, thankfully.
Major positive point is that it uses wifi and supports casting from Google Play. But for my gf who uses iphone she hates it.
Overall we prefer the Marshall bluetooth speakers over Sonos because in an apartment there's rarely a need for wifi casting music.
Edit: Chromecast audio is also a viable alternative. Based on how well my regular Chromecast (video) works for me I assume the audio one is as good.
In earnest, what is a better approach here?
I wish MediaTomb or Ampache was a better DLNA server than Plex.
1) RMS is always right.
2) RMS is that you?
To all the people talking about ownership. I find it hard to believe the aux in will cease to work. So worst case is they turn into regular speakers.
What it sounds like is you won't be able to update your firmware. So more likely than not, everything would keep working but random Internet related services (like Spotify Integration) may break over time because, for instance, if Spotify changes their API you won't get the software update to fix it.
And that is why I think it is OK. Software updates over the Internet are always subject to licensing. That is not new and not unique to Sonos.
If I wanted regular speakers, I would have bought regular speakers.
The part of the speaker that is actually a product is unlikely to stop working.
Being able to play music on a 'smart speaker' from my phone isn't a service, it's just a feature of the product. If you sell a product with features that require you to host some servers somewhere publicly accessible on the internet, that doesn't make that product feature a service.
Don't like that? Don't include those features in your product. Design them better so you don't have to forever maintain these servers.
If you want to announce you are compatible with service X as a feature, you should have an agreement of compatibility with service X.
.... until it does happen as they promised ...
> sound systems that may eventually "cease to function".
Not just some part of it but all of it.
They would need to have a remote kill switch and use it.
Secondly, they have done this before (forced agreeing to the terms to do a software update) and every time they have warned "some functionality might cease to work" and every time I kept using it past the cutoff.
Once, I had Spotify stop working on me. I assume because they switched API keys or something.
In fact, for example, they had an update that killed Audible integration. And my Audible integration kept working until I installed the update.