Sonos’ frantic flailing illustrates the stupidity of smart tech
extremetech.com
extremetech.com
The Star Trek tricorder, that can do with a handheld device what would take an entire lab, is smart tech. But that requires lots of research and wouldn't lend itself to a subscription-based or planned-obsolescence business model even if we could build it. :P
It also has downsides; for example, every new device synced to a photo library has to process the whole library from scratch (to identify photo content for searching).
Are you kidding? Why would you stick with TriCorder v17.3 when the new sensor suite in TriCorder v18 can deter hypercalcemia with twice the accuracy!?
- planned obsolescence has no real economic meaning in a world (Star Trek's) where abundance is materially realized — symbolized by the synthetizer technology and omniscient computer (free info / education / content). Everything gets recycled and obsolescence is de facto moot because they can produce orders of magnitude more than they can ever consume (a holistic premise of the fictional universe that is broken numerous times in the short stories of the show, but hey, it's cool it's TV).
You could actually say that in Star Trek, obsolescence is more than planned, it's actually how you deal with anything generic or commodities: you recycle it, and would just rebuild it when needed (unless it's inconvenient and you'd spare some space to store the assembled object, but that's paying a high cost, space isn't infinite unlike synthetizers' production, mastery of matter and energy).
Yup, inventory and logistics truly are the heart of civilization. Note that 4D-space is cool in sci-fi but alas...
- subscription-based: weeeeell, there's no cost associated with living in Star Trek, is there? Everything is just provided by said abundance-tech, nobody pays to eat or clothe or have a roof. So subscriptions, well, there's no point; Netflix would just be free, people there would just do it for the sake of making this cool thing. As everything else is done in the Federation.
That's the core of the real utopia of Star Trek IMHO. Dilithium crystals look very real compared to all of the above, don't they? haha.
And that is what is getting them in trouble currently- Sonos is really a software product that is masquerading as a hardware product. People bought it expecting it be like, well, a speaker, but what they really bought was an experience more like buying a copy of Windows.
pwthornton mentioned that Sonos could becoem a SaaS company and start leasing hardware instead. I see that being a hard sell for customers, but going fully the route of being a SaaS subscription service for their coordination and streaming management services seems possible. That being said, I think they are terrified of becoming just and audio source- they currently charge $450 for the Sonos port, which basically just acts as an audio source for an existing stereo system.
The software they provide that makes their speakers work is the magic as you say, and it really does work better than anything else. It's certainly way better than AirPlay 2. But keeping that software working requires updates.
I don't know the best path forward, but they should at least consider the Eero+ route of selling a subscription service that provides real value that also helps them support devices long term. Eero is incentivized to continue to support my 1st-generation Eeros because I pay them $99 a year.
There is a lot of interesting stuff that they could do with a subscription service that would add value, and would also help them continue to pay for updates on older devices.
It's the walled garden "trust us it'll be fine & powered by magic" spiel that's the issue.
I'm flexible on the whole open source issue - not my crusade, but I do want some semblance of understanding how it works so that I know a corporate decision isn't going to take the magic blue smoke out of my devices for me.
I'm not saying that the behavior Sonos engaged in here is good - I'm just saying that jumping to "it should be illegal" so quickly without considering all of the secondary impacts is a mistake.
Not active and existing APIs, discontinued ones.
The cost to do the work and the cost to ensure compliance with the law would both be real dollar amounts for most small companies. Enough that if you wanted to try out an experimental new service you'd have to take into account that cost in your budgeting.
Telling a company (or individual) that they need to sit down and pay to kill their products means they will think harder before they launch them. That's a barrier, and it will slow down innovation and make sure that only the companies with the resources to support compliance and open sourcing will be the ones to launch new services.
The stifling innovation argument is faulty anyway. If you are really innovating, then there should be no reason to fear releasing the code for something the company considers obsolete. Unless, that is, such innovation is a red herring and the new product is a mere repackaging of the same, old technology, done exclusively for the sake of artificially sustaining cash flow.
Many products contain proprietary elements that are licensed by third-parties who have no reason to agree to their software being open-sourced. The company I work at constantly gets calls to open-source product X that they discontinued, but they can't because of the reliance on code that doesn't belong to them. And predictably enough, they're not willing to go back and renegotiate a new license for a product that they're discontinuing.
And since open sourcing it does not result in revenue, they cannot justify the cost of vetting the provenance.
A way to fix this would be to shorten patent and copyright terms.
Will it build as-is? Probably not.
Will it build if someone really wants it, and is willing to spend some time to get it to work? Likely.
Will it have the exact same functionalities as the original firmware? Possibly not.
Does it give users a way out of planned obsolescence? Yes.
A penalty large enough to impact the companies you care about here would be business destroying to a small organization. Individuals and companies are not going to want to assume that kind of liability. I don't see how anyone could support this kind of measure if they are thinking about the breadth of business out there.
whoever controls the buildsystem controls the hardware
the constant FBI-apple push-pull where the FBI wants apple to release a custom update is a version of this too
After spending several thousands of dollars, I shouldn’t have to feel worried that my investment will become artificially obsolete. Already I can’t directly play music from my iPhone to my speakers like before, I have to use airplay which is clunky at best.
Sonos has been in the news for the last few months, but before that I had hardly heard of them. I don't own any of their stuff and I don't have any bone in this. I just noticed that their name keeps popping up on reddit, in the news, and here on HN. Mostly it's for negative reasons.
I feel like this is part of some marketing campaign, but almost all of the news has been negative.
First I remember some articles about Sonos going to war with Amazon over something or another last October.
Then there was the controversy regarding "Recycle Mode" thing that bricks their speakers.
Then Sonos sues Google for patent infringement.
Most recently Sonos announced it would not provide software updates for legacy products, CEO apologizes, bla bla bla.
And then this article.
This is not normal.
Not simple enough for all consumers yet, but works flawlessly for me every day in my house.
The solution is of course for customers to fight back against this bullshit. But that requires competitors who don't play this game, and that requires VCs who will tolerate a healthy, slowly-growing business rather than one that grows exponentially and flames out. I hope we can return to that world someday.
Curmudgeon likes his 50 year old speaker more than that new fangled tech stuff. Then rants about tech being in our lives too much. (I could only skim it as so boring and repetitive - already read this story too 1000 times).
Smart tech has issues around support, but it is so damn useful. I thought little of it until I started moving everything to it. It also has a network effect - the more you add the more useful it gets.
When I walk away from my building my the thermostat goes down, door deadbolts, lights go down, music and TV all go off etc. Pry it from my cold dead hands.
This kind of seems like saying 'the more microservices you add, the more useful it gets' which only works up to the point where services don't encounter any integration issues.
What happens when your thermometer starts sending messages that cause your door to unlock when the temperature gets too high? Or if your TV ends up shutting down because of malformed packets being sent across your home network?
IoT is an integration nightmare that only gets worse as stuff fades out or ends up receiving no more support from the manufacturer. Those items you bought from 50 years ago will still work just fine even if the company goes under. But the same can't be said for modern technology we're filling our homes with.
And that's even without talking about the privacy issues. I'll take dumb items that I can make smart over smart items I can not make dumb any day of the week.
Generally, most HomeKit devices are LAN or Bluetooth based, and theoretical should work without needing to phone home. (Some might stop if they can’t, but that’s an implementation issue, and i haven’t experienced it yet)
what happens when it decides that it won't connect with next version of iOS/Android unless you upgrade?
Microservices vs Monoliths. Don't put speaker, amplifiers, processor and wifi all in one box.
Put the slow-changing stuff (speakers, amplifier) over here and pay once for it. Put the fast-changing stuff (pandora! spotify! Tidal!) over here, and pay a subscription for it. I pay for Tidal and Roon. Tidal for content, Roon for updating software as iOS updates, linux updates, and SPDIF drivers need updating. As long as Roon supports SPDIF, I'm good.
but in application it does about everything, and most jobs it does haven't changed much.
HN is generally filled with other tech people who have turned luddite in their old age.
Any solution that requires receiver/amp/combo and wires is going to be limited where you can deploy it.
Before wireless audio like Sonos, most people have a stereo in one room in the house. Now many people have speakers in almost every room in the house.
I got rid of my Onkyo receiver and JBL studio series to go with Sonos because it is a better listening experience in the real-world. Audio being synced across your house is the real deal.
But I am a little concerned that when I move to the new iDroid Megapixel XXXL 3000 in late 2025, I'll have to replace them all if I want to be able to send audio from my phone to the speakers.
When having the sync across the whole house is the main reason you like the setup, it's even more galling to need to replace them all. I wish (and will be watching to see if they do, before I add any more speakers to their system) they'd come up with an upgrade path.
It'd seem like a smart thing for them to do. Surely they could come up with some kind of bridge gadget you could connect to your network between the old and new for the cost of about one speaker. That'd spare them having people angrily leave their ecosystem because, if you need to do 5 rooms at once anyway, you might as well shop around.
They are no longer. The iPhone makes a great controller. Spotify and Tidal mean we don't need a local file store. Speakers from Bluesound, Amazon, Google and Apple compete at various price points at the low end. If you do want to spend on a local controller, then Roon, a spin-off of ultra-audiophile Meridian, has a better experience than Sonos, and is supported by brands like Meridian, Naim, Nad, Trinnov and $100 hifiberries.
I have Mission speakers in the bedroom. Meridian speakers and a hifiberry in the lounge. A 7.1 system with another hifiberry in the TV room. Bluesounds in the kitchen and dining room. And the bluesounds have battery packs, so I can carry them into the garden if it will ever stop raining. With Roon, they all play the same music in perfect sync. A new "remote" cost me $49 for a 7" Amazon tablet, but I can also use the phone in my pocket, or the iPad upstairs.
So Sonos is competing with Echo Studios, Bluesounds, Apple and Google, at the low end, and it's being beaten by a pure software company who is integrating with audiophile kit at the high end. That software company charges a subscription. It's not relying on hardware sales, just as those sales are getting killed by cheap competitors.
Meanwhile they've got legacy kit. IIRC the original stuff set up its own wifi network that it controlled in a proprietary fashion. Nowadays, competing devices just work on whatever you wifi you got, and I assume Sonos' newer devices do too. That can't be fun to sort out, hence the "two networks" solution. To be competitive, they have to cut the old stuff loose.
Yeah it's very easy to just give in but the amount of things we toss in the trash that's perfectly fine but simply no longer supported is obscene, and getting worse by the day.
I can't bring myself to give in.
The point is that bad faith actions (forced obsolescence, surveillance, no commitment to security, weak sharing of protocols etc.) are holding back smart home and IoT significantly. This technology has so much potential and is not THAT new but adoption is still hit or miss. It’s a bad sign that so many self described techies are comfortably waiting on the sideline or worse, afraid to have some of these devices in their homes.
I agree that smart tech is beyond useful but when the big IoT players ignore detractors and repeat the same mistakes over and over again it’s just flushing potential (and value) down the toilet.
It would be great to see industry take complaints to heart and begin to build trust in their ecosystems.
This is the response I got from them. Note: I did not accuse them of bricking my devices. Also note: they are promising the gradual degradation of my experience. There is no commitment to support, nor willingness to open source or find other remedies (or even pay them for continued support, which I offered in my letter).
Of last note: we have decided to dump Sonos permanently and will actively avoid doing business with the company again. It’s a fairly tone-deaf response unfortunately.
Curious if others are as unhappy as we are? We are just disappointed.
Anyways, the response:
Thank you for contacting Sonos Customer Care. I understand your frustration with the end of software updates announcement. Please know that we in no way mean to render your system inoperable or brick the devices. The trade-up program you mentioned bricking units is entirely voluntary and a multi-step process to ensure it is not mistakenly activated on the system.
Your devices are not being end-of-lifed by any means; they are simply not receiving new update information beyond a certain version. This is due to the available memory and processing power available in our legacy line of devices. This is specifically what separates them from modern and legacy. The Legacy devices have hit a point where in order to keep allowing new features; we would have to remove old features because there is simply no room left to store the new data for newer technologies or changes in outdated technologies no longer used by our partners. It is similar to having all seats on a plane full and telling someone already boarded that they need to get off. To keep your system functioning as it stands; we need to halt the progression of the software.
Once the change is made, your Sonos system will no longer receive regular updates, so your experience will initially remain the same. However, the functionality of features and services will be impacted as technology evolves over time. The thing to worry about once this takes place is the third party services and partners offered through our platform. Using Spotify as an example; if they change something in the way Sonos accesses their service then your System not receiving new updates will not be able to accommodate this.
However, since this won't be in effect until May of 2020; we are still working on figuring out workarounds and things of that nature for all of our customers who will possibly be affected by this change. Just keep an eye out on alerts from us. Once we do have that information, we will gladly share that with you.
More information here: https://support.sonos.com/s/article/4786 https://support.sonos.com/s/article/4798
Our commitment is to support products with software updates for a minimum of five years after we stop selling them, and we have a track record of supporting for longer.
The Trade-Up eligible devices all have newer counterparts that allow the old connection types while also integrating new features. The best example of this is the Sonos Connect:AMP and AMP devices; the Connect:AMP was not designed with Home Theater in mind where the AMP has an input for a TV source without removing any of the existing connection options that the Connect:AMP had; while also integrating the airplay functionality into the device and system.
There's a lot to play with in a HiFi that you just don't get exposed to with an off-the-shelf system like Sonos.
Also integrating parts that have widely different reasonable life spans into inseparable, non-repairable, non-upgradeable units is not responsible waste management.
And that’s what’s turning some of us into cranky luddites.
Updates cost money, and Niley Patel speculates that Sonos is losing money on older devices with all of the updates they have been providing. But the devices still work, which is a testament to their build quality. Sonos needs to figure out a solution.
There are some solutions:
1) I don't know how separated and abstracted their development process is, but, ideally, they would be able to continue to provide security updates, even if older devices don't get anything beyond that. But how many years should they do? 10? 20?
2) Sonos could sell extended support for devices that people have owned after X amount of years. Even charging $19.99 a year for a household would probably put this back in the black. If people are going to use Sonos products for 10-20 years and often have expensive and complicated setups, they should figure out some long-term support.
3) Sonos could start to become a SaaS company and largely lease their speakers. Microsoft is starting to do this with the Xbox, and it is a way around this issue. As hardware gets too old, Sonos would replace it with newer hardware. There is a reason that more and more software is subscription-based.
3) Sonos could come up with a premium subscription offering that bundles in long-term support. Eero has Eero+, which I find to be a good product, and it's how they get into the black on hardware sales.
I'm working on a longer write up about this for Medium, but this isn't as easy of an issue to solve as many make it out to be. Providing infinite support for networked-based devices will eventually put you out of business. Sonos is the one dealing with this because they have been selling these speakers the longest, and they have people who are able to use their devices for a long time.
One issue I think they are running into is that they used to update products pretty infrequently. They didn't even do small step updates to keep the CPU and networking equipment updated. This is particularly an issue with the 1st gen Play 5 that they sold for about a decade.
People who bought a Play 5 15 years ago should not continue to expect support. People who bought one five years ago should. They have the same exact internals.
That was a mistake, and they should make sure they are upgrading the internals in every device every three years or so. Had Sonos been doing this, they would be talking about far fewer products.
Sonos has devices that range from 32 GBs of ram to 1024 GBs. The delta in their older stuff and their newer stuff in terms of tech is huge. This strategy is on their old CEO, and their new CEO believes Sonos has to update their stuff more often. It's probably going to take a few years for the update cadence to sort itself out, but in the meantime, they are kind of in a quagmire.
(edit) So the (new) Sonos position of basically “OK, we’ll keep the hackers out, just don’t expect us to support the new streaming service released in 2022” seems fine for both sides.
They largely have been doing this, but their decision to keep the original Play 5 on sale for so long means that some people may have bought it within the last five years, despite being a really old device.
Funny you mention this, because they recently launched this: https://flex.sonos.com/