Sonos CEO Says the Old App Can't Be Rereleased (theverge.com) 106
The old Sonos app won't be making a return to replace the buggy new version. According to Sonos CEO Patrick Spence, rereleasing the old app would make things worse now that updated software has already been sent out to the company's speakers and cloud infrastructure. The Verge reports: In a Reddit AMA response posted Tuesday, Sonos CEO Spence says that he was hopeful "until very recently" that the company could rerelease the app, confirming a report from The Verge that the company was considering doing so. [...] Since the new app was released on May 7th, Spence has issued a formal apology and announced in August that the company would be delaying the launch of two products "until our app experience meets the level of quality that we, our customers, and our partners expect from Sonos." "The trick of course is that Sonos is not just the mobile app, but software that runs on your speakers and in the cloud too," writes Spence in the Reddit AMA. "In the months since the new mobile app launched we've been updating the software that runs on our speakers and in the cloud to the point where today S2 is less reliable & less stable then what you remember. After doing extensive testing we've reluctantly concluded that re-releasing S2 would make the problems worse, not better. I'm sure this is disappointing. It was disappointing to me."
Note to self. (Score:5, Insightful)
Buy nothing from Sonos.
Re: (Score:3)
I have one old speaker from them. I use it in the garage occasionally. The new shit is super unreliable. Power cycle the speaker, then force quit the app and and restart it, probably works but any other scenario is probably painful.
Re: Note to self. (Score:5, Funny)
Removed Android playback AFTER I'd purchased $1,000 of speakers. Fuck them. With ScotchBrite.
Re: (Score:2)
Problem #2 - "The old Sonos app won't be making a return to replace the buggy new version." OK fine. Then fix the buggy new version. You wrote the old version, so you obviously know how to do it, so fix whatever is wrong with the new version.
Oh wait
Fucking idiots.
Re: Note to self. (Score:3)
You're assuming that the people who wrote the old app still work for Sonos. This assumption might be wrong.
Re: (Score:2)
Re: (Score:3)
Wouldn't be the first time a company changed its build infrastructure, making it impossible to compile legacy releases again. Source code is always archived (well, almost always) but way too few companies worry about archiving the build environment once a release has shipped.
Could this code be resurrected? Sure. But is it worth the expense? Clearly Sonos leadership doesn't think so.
Re: (Score:2)
Clearly Sonos leadership doesn't think.
FTFY
Re: (Score:3)
Probably comes across as sour grapes, but it was pretty easy once I'd seen their pricing. Plus, I'm not above just clipping the JBL Go 3 that I got on clearance to my belt loop and playing music from my phone. That's all the multi-room audio I need, *smirk*.
Re: (Score:2)
I'm curious about that too -- are Sonos speakers worth it over what is out there? I have been pretty happy with no-name stuff, or brands like Anker or JBL. If I want "serious" speakers, I can spend a couple C-notes for a set of USB monitors with an acoustically flat response and with a volume that is definitely loud enough.
If I need "real" speakers like for a good P/A system, I use a USB audio interface.
Re: (Score:3)
My understanding is that they don't make total trash in hardware terms(as you'd want, for the money); but the reason that they exist is because it's really inconvenient to cut holes and run wires all over to do multi-room/whole home the classic way; and significantly more expensive(and not necessarily less proprietary or brittle) to bring in a systems integrator to crestron you up.
They've been doing their best to change this; but historically t
Re: (Score:2)
Is pulling wires really that expensive?
https://www.ikea.com/fi/en/p/m... [ikea.com]
These things are dirt cheap and let you pull thin speaker cabling pretty much anywhere in the house with minimal to no drilling.
Re: Note to self. (Score:2)
Tenants might not be allowed to do even the small change you described.
Re: (Score:2)
I'm sure there are a whole lot more people who are too lazy to do even that compared to amount of people who are capable and willing and not "allowed" to do that.
So this point is frankly a stupid "muh politics" red herring.
Re: Note to self. (Score:2)
From a merchant's perspective, as long as there are people willing to pay for the product, it doesn't matter that much if it's because they're lazy or forbidden, there is a demand for the product regardless.
Re: (Score:2)
Of course. There are plenty of people buying apple fashion accessories for example.
That wasn't the point of discussion though. The point was aimed at a specific claim that this combination made sense not as a "lazy or forbidden" thing, but because pulling wires is somehow more expensive than getting much more expensive "smart" speakers.
Re: (Score:2)
Is pulling wires really that expensive?
https://www.ikea.com/fi/en/p/m... [ikea.com]
These things are dirt cheap and let you pull thin speaker cabling pretty much anywhere in the house with minimal to no drilling.
At some point the long lengths of thin cabling start to compromise sound quality. Additionally, some power amplifiers may become unstable with the additional capacitive load that long, thin cables represent.
Re: (Score:2)
Buy slightly more expensive shielded and appropriately twisted cables. You're not pulling hundreds of meters, unless you're living in a mansion.
Re: (Score:1)
Over the years I have found that twin and earth copper cables work just fine from installed speakers.
Most recently I went with 2.5mm T&E for a Yamaha Multicast for EV Evid ceiling speakers in my lounge and kitchen, no capacitance worries and plenty of current carrying capacity.
Re: (Score:2)
Well your house already has wires, otherwise how else do you power those devices?
The problem is that when those power wires were installed, noone considered that you might want to run other types of wiring at the same time.
When i had this house renovated (which involved replacing old and dangerous electrical wiring) i had cat6a put everywhere at the same time, and it didn't make much difference to the cost of the work. I had no need for speaker wires, but there's no reason that couldn't have been added too.
Re: (Score:2)
Click the link. It's there for a reason.
Re: (Score:2)
Trunking is ugly, and can be very inconvenient.
Often you'll have to route it a long way around, eg up and over any doorways, as if you ran it along the floor it would get trodden on and damaged.
It's only slightly less ugly than having the bare wires running around.
Re: (Score:2)
Notice how excuses evolved from "it's too expensive" to "you need a permission" to "I don't like how it looks".
It would appear we have arrived at the truth. It's not that it's expensive. It's not that you need a permission. You just have an aesthetic preference, and you're willing to pay through the nose and be severely inconvenienced to be able to match that aesthetic preference.
Re: (Score:2)
You do need permission if you're only renting the property and don't own it. Even putting trunking would probably require permission as it's fixed to the walls, and might cause damage when removed again.
Doing it properly is expensive unless you're already renovating/rewiring. Doing it cheaply is ugly. I don't want to have trunking, exposed wires or pipes all over the house.
Personally i consider it worth the money to do properly, which is why if you read my previous comment you'll see that i did just that, a
Re: (Score:2)
It can also depend on the details of the structure, or what other work you are doing incidentally; I remember my dad having to exploit an unrelated renovation of the (fully finished) rooms on the top floor of the house to get the pulls he
Re: (Score:2)
Have you tried clicking that link yet?
Re: Note to self. (Score:2)
Mod parent through the roof.
Re: (Score:2)
Let me check. Yep, have that too, but it is pretty old.
Re:Note to self. (Score:5, Informative)
Wait... you're only writing that note NOW?
Where were you when Sonos sold user data?
Where were you when they sued Google for suggesting they own exclusive rights to control more than one speaker at a time?
Where were you when Sonos decided that to qualify for a 30% discount on a new device you had to brick the old one irreparably?
Where were you when they sued Google for something wireless?
Where were you when they announced that if you have a legacy Sonos device on your network it will prevent functionality of the new ones you purchase and block their updates?
Where were you when they sued Google for calling a speaker smart?
Where were you when they forced new privacy policies on users or locked them out of their speakers?
Also why hasn't Sonos sued Google yet over their practice of ignoring a "do no evil" policy, it seems like Sonos is the king of that.
Re: (Score:2)
Buy nothing from Sonos.
You waited a while to figure that one out. Anything "app based" has been off my shopping list for years now.
If something requires an internet connection to work, you don't own it.
Re: (Score:2)
Sonos does not require an internet connection to work.
You're still dependent on the vendor for something to make it work, so same principle.
Re: (Score:2)
Sage advice for the last 20 years! The company started on top of a hill and has been rolling down ever since.
Sonos have wasted so much potential fucking around with their customers; anyone who continues to put up with it is flat out crazy, or at the very least extremely susceptible to the sunk cost fallacy.
Did someone accidentally nuke the old source code? (Score:2)
Right. You can't revert the onboard firmware because you monkey patched one too many times and you can no longer guarantee compatibility with the previously (working) released software.
How much is this clown making to stand at the prow of the ship and shout amends for not being able to turn the ship away from the iceberg?
Re: (Score:2)
It should be remembered that in a recent article it was disclosed that development of the app would be overseen by a member of the board. If that's not evidence that the CEO should be fired what could possibly be? The CEO is literally the Chief Executive Officer, responsible to the board. Overseeing the app is an executive responsibility. He is literally incompetent by the board's own admission.
Re: (Score:2)
The only possible scenario I can think of is that it is much worse than what has been revealed, and the board is keeping him around so that they can use him to layoff most of the company and then subsequently take the blame for everything once the truth comes out.
If I was working for Sonos, I'd be screaming and bolting for the exit.
Re:Did someone accidentally nuke the old source co (Score:4, Insightful)
The board is supposed to be there to represent the interests of the shareholders. They are *not* there to actually run the business; that's the CEO's job, and his subordinates, and the two are best not combined (which is why it is not good governance to have the chairman of the board and the CEO be the same person, even though all too often they are). If you have a board member running one of your major projects, you're Doing It Wrong.
Re: (Score:2)
Re: (Score:2)
Right. You can't revert the onboard firmware because you monkey patched one too many times and you can no longer guarantee compatibility with the previously (working) released software.
How much is this clown making to stand at the prow of the ship and shout amends for not being able to turn the ship away from the iceberg?
I suspect the problem is the old cloud software doesn't work with the new version of the app. And since you can't guarantee everyone will "upgrade" to the old version of the app you need to keep the new cloud software running.
And if the old app is even buggier than the new app when talking to the new cloud software... well that's the situation he described, downgrading isn't an option.
Re: Did someone accidentally nuke the old source c (Score:2)
Then the bug is actually in the new cloud software, rather than the new app.
Re: (Score:2)
Re: (Score:2)
Revert the cloud to old software. Add a patch to reject the new app connections, perhaps push a message to the user to apply latest update (which would revert it). Not that complicated, if in fact possible to revert the cloud.
And you're inundated by complaints for devices being temporarily bricked (do you need to downgrade speaker firmware too?). Not to mention, there might be other infrastructure updated for the new software that's now hard to go back (if the database upgraded that could be nasty).
Not to say it's impossible, but it could at least be a month or two to come up with a solid plan and test it out, if not longer. At that point they might as well spend their time fixing the new stuff instead.
Re: (Score:2)
This does give me some comfort that my old S1 system won't magically receive an update in future that takes what functionality it has away from me.
Ineptitude FTW!
Ohhh (Score:1)
Re: Ohhh (Score:2)
Re: (Score:2)
A lot of the time QA is held back and prevented from doing their job by management. When QA stops a new release because they found bugs, that prevents new revenue, so from management's perspective they are worse than a cost center, they are an anti-profit center, to be curbed and held on a tight lash as much as possible.
Re:Ohhh (Score:5, Insightful)
Why do you assume the programmers are at fault? It is extremely common that leadership refuses to spend the time necessary for thorough testing, and forces the software to be rushed-out even if it is known to be buggy.
Furthermore, once the software is stable "enough" that people are buying it, they will often disallow the developers from fixing any bugs that remain, since their number-crunching shows that the expected return-on-investment for the cost of those bug fixes is too low, and some shiny new replacement version of the product has a higher ROI (even with the new bugs it will bring).
In my experience, software developers hate bugs and don't complain when they are asked to spend the time to fix them. They even get agitated when they aren't allowed to fix them. But deadline pressure and corporate bean-counters stand in the way.
Re: Ohhh (Score:2)
If you are triggered by this message, really, eat more vegetables! It is good for you.
Re: Ohhh (Score:2)
At most companies, I just make it a policy to fix every big I encounter when working on a project before writing any features for the new project. Then no bean counters can come to me and say release it as is.
I've seen plenty of devs work on the features first, then they get pressured to release before doing things like writing tests.
Nope, features come last.
Re: (Score:3)
Why do you assume the programmers are at fault? It is extremely common that leadership refuses to spend the time necessary for thorough testing, and forces the software to be rushed-out even if it is known to be buggy.
Furthermore, once the software is stable "enough" that people are buying it, they will often disallow the developers from fixing any bugs that remain, since their number-crunching shows that the expected return-on-investment for the cost of those bug fixes is too low, and some shiny new replacement version of the product has a higher ROI (even with the new bugs it will bring).
In my experience, software developers hate bugs and don't complain when they are asked to spend the time to fix them. They even get agitated when they aren't allowed to fix them. But deadline pressure and corporate bean-counters stand in the way.
This is how it works in modern business, nothing is ever the executives fault. The buck never stops there. When there is a problem, regardless of who caused it (and that is _never_ at the executive level) then someone lower down will be found to take the fall for it.
Re: Ohhh (Score:2)
Are you talking about Crowdstrike?
Re: (Score:3)
Those idiots monkey programmers need to be fired.
Chances are, those programmers pointed for months to the failed tests, still marked "FIXME" code parts and said clearly that the software is not yet ready for release. But some manager, marketing drone or other PHB pushed them to an impossible deadline.
Which when done right is a good thing. We all know the Free Software projects that are still at version 0.9.2beta25.4 because the pure-blood developers just can't get themselves to consider it "done". In the commercial world, someone needs to push them to get
Re: Ohhh (Score:5, Informative)
Programmers are never responsible for quality control. If your organization doesn't have independent quality verification, then it's the CTO who should be fired, not the programmers.
Deployments are always the hardest part (Score:4, Insightful)
SO many developers, and software companies, think that getting the code right is the hardest part. If there is any kind of scale to the product, it's not, not even close. Deployment is a software development project all its own, and (certainly the first iteration) is often equal in size and complexity to the software itself. Many, many companies skimp on proper deployments, because it feels like the software is already "done" when it is functionally complete, and they're in a hurry to get it "out there" and stop spending money on automation that nobody can see.
CrowdStrike is a great illustration of this, and now Sonos. They rolled out the release before it was ready, and now there is no rollback plan. I have no doubt that the lack of a rollback plan, was due to cost "savings." It usually is.
Re: (Score:1, Offtopic)
There should not be a need for a "rollback plan", you talk as though that's a standard thing.
Sonos has existed for decades, its incompetence is a recent development. They are familiar with deployment, their problem is clearly lack of design and test.
Re: (Score:3)
Externally you wouldn't necessarily need to make it obvious that it's a rollback. You roll back the code, increment the version and your customers who don't know any better just assume it's "new". As someone else already said though, Sonos must've really screwed the pooch with the handling of previous versions of their codebase.
Re:Deployments are always the hardest part (Score:5, Informative)
Yeah, "rollback plan" should indeed be a standard thing. For every deployment, my team has both a roll-forward and a rollback automation script (including database updates), each of which have to be tested before going to production. We rarely use the rollback part, but on the few occasions when we did, we--and our customers--were very thankful it was there.
If you don't have a rollback plan, you are walking a tightrope without a net. You are *assuming* that your deployment will be successful, and not giving yourself an option to undo, should problems occur. This approach is common, but extremely risky. "That'll never happen!" Oh yeah it will. Just ask CrowdStrike about that.
Also, and in the same vein, large releases are much riskier than small, incremental releases. This is a key reason why agile works, because it insists on small, incremental releases. It's noteworthy that the Sonos release was a large update. Large updates should be rolled out incrementally, hidden behind a feature flag. Then, the new (major) version can be enabled for progressively larger groups. By the time everybody has it, it has been proven to work.
Re: (Score:3)
> Large updates should be rolled out incrementally, hidden behind a feature flag.
Absolutely, but making that actually work requires some real engineering - and that takes time and money. I'll bet it got skipped for "cost savings" (probably to keep the share price up). Well, that sure worked out well for them!
Re: (Score:2)
Yeah, "rollback plan" should indeed be a standard thing. For every deployment, my team has both a roll-forward and a rollback automation script (including database updates), each of which have to be tested before going to production. We rarely use the rollback part, but on the few occasions when we did, we--and our customers--were very thankful it was there.
We go one step farther. Anything significantly new is flag-guarded, and present in the release simultaneously alongside the old functionality. And we make sure that we can toggle the flag on and off for internal users at will without breaking anything horribly before we even think about running a 1% launch. By the time we do a 1% launch, ~100% of customers have both the old and new code on their devices, and if we need to roll something back, we just tell the server to always say "no" for that flag.
With
Re: (Score:1)
Here's some good tips on deployments, including the need for a rollback plan.
https://codefresh.io/learn/sof... [codefresh.io]
Re: (Score:3)
There should not be a need for a "rollback plan", you talk as though that's a standard thing.
for a system that's to replace an existing deployment? you bet it should.
Sonos has existed for decades, its incompetence is a recent development.
well, it doesn't look like they will exist for much longer. if they had had a rollback plan, however ...
Re: (Score:2)
A rollback plan is a standard thing in computing.
In the two decades I've been required to use Information Technology service management frameworks (e.g. HP ServiceDesk, ServiceNow, FreshService), a rollback plan is a standard artifact. Sometimes, a colleague writes junk in their rollback plan (e.g., "We cannot rollback because..."). Then I have to scold them to think harder. A rollback can be as basic as restoring a server from backup, or taking VM snapshots before the change. Rollback is almost always an o
Re: Deployments are always the hardest part (Score:2)
"rollforward is generally significantly more complicated than rollback"
Although notably, one of the common exceptions is when you have a bunch of scattered devices which may or may not have had firmware updates applied, may or may not be connected to the network, and are completely out of your control.
Re: (Score:2)
That is easily proven as false. Sonos has basically been a company riding a wave of incompetence for nearly as long as it was competent. The problems did start back in 2017 already, and since then they've filed lawsuit after lawsuit, told customers their products would break if they buy a new device without binning the old ones, told customers that if they wanted to have a discount on the new device they had to permabrick the old ones, been tonedeaf on the right to repair, fucked up the rollout of Nest conn
Re: (Score:2)
You contradict yourself in the Sonos case. Either the software is done and functionally complete or it's not ready. It can't be both. The problem here wasn't one of rollout, the problem was that the software itself wasn't done or functionally complete.
You don't make changes like this with a rollback plan in mind. That is effectively admitting failure from the onset. You do have to have a functionally fast development team which can address issues which users raise, but I have no doubt that Sonos *could* rol
Re: (Score:2)
Oh yes, software can certainly be "functionally complete" and "not ready" at the same time! Sonos is a great example. The functions are there, but so are many bugs. The problem is that, once it became clear that it was buggy, they had no way to roll back the buggy release, or to turn off the feature flag (which should always--in massive deployments--be used to hide new functionality that isn't yet proven).
Having a rollback plan is a recognition that, despite human best efforts and practices, sometimes thing
Glad I don't own any of their shit! (Score:2)
Works fine for me. (Score:2)
Another Downside of Patents (Score:5, Informative)
When companies make a habit of demanding patent royalties from their competitors for revenue, they get less thirsty and stop caring about their own competitive advantage as much.
Especially with so many obvious patents being issued we'd be better off without any of them.
I never owned a Sonos but I know people who raved about them a decade ago.
Re: Another Downside of Patents (Score:2)
Yeah, I had a coworker try to sell me on this crap. There are some people who just will never understand why I wouldn't ever buy an Internet connected device.
Fair enough (Score:2)
It makes sense, I was surprised when they said they were considering it. They've clearly reimplemented the whole back end and the old app wouldn't work any more. End users thinking that the app was the entirety of Sonos' systems was a bit naive.
They rolled it out too soon though.
And firmware... (Score:1)
They've clearly reimplemented the whole back end and the old app wouldn't work any more.
I also figured they had updated the backend, but a surprise to me was that a firmware update was involved in all of this... seems like you'd want to change no more than two elements at a time if at all possible. What a scramble!
Worst CI/CD Engineers Ever (Score:1)
Re: (Score:2)
You're speaking about Crowdstrike, Microsoft, or Sonos ?
The competition to the bottom gets tricky....
Re: (Score:1)
Home Assistant / Music Assistant (Score:5, Informative)
Not a Sonos user here, but the above open source software stack claims to support Sonos devices. I use it with my Chromecast, Marantz and WiiM devices. Still a little rough on the edges, but I actually use it daily now with my 80-something speakers. It doesn't support every streaming service. I use it with Qobuz and a local collection over SMB/CIFS.
It also supports Apple Music, Deeper, Spotify, Tidal,, SoundCloud, YouTube Music, as far as streaming platforms go. There are also many local options such as Jellyfin, Plex.
This will work mainly for techies willing to maintain a host or VM running Home Assistant. Not necessarily a lot of overlap with Sonos customer target who want plug and play and are willing to pay for it.
Still better than having a useless app, though.
"the level of quality... (Score:2)
...that we, our customers, and our partners expect from Sonos."
So, basically until it's dog shit, a ripoff, non-functional, bricks things, prevents me from playing music when my spouse's phone leaves the LAN, de-featured from what I purchased, non-functional with my music library or server, nonstandard, proprietary and broken.
Shouldn't take long.
I use Denon Heos (Score:2)
Badge of honor among smart devices (Score:2)
We made our system so damn overly complicated, even we can't figure it out, and we've lost control of our system. But if it doesn't work, remember... you're just holding it wrong.
This is what happens (Score:2)
When you jump out of an airplane with a needle, some thread, and a moth-eaten bedsheet and hope for the best.
It sounds like their system is a tangled mess by design and now they've painted themselves into a corner.
Where "roll forward" comes unstuck (Score:2)
Never buy anything... (Score:5, Insightful)
...that depends on the cloud when it could easily run locally
Re: (Score:2)
Sonos runs locally. They make for great AirPlay speakers, the app is really only needed to setup linked systems.
The Arc/One SL/Sub makes for a great 5.2 ATMOS setup on my home cinema.
Re: (Score:2)
Sonos runs locally. They make for great AirPlay speakers, the app is really only needed to setup linked systems.
...and to get the speaker onto Wi-Fi; they don't provide a means to connect the speaker to a wireless network except through the app...which is account-walled and won't allow a speaker to be connected to a wi-fi network without creating an account, which requires internet connectivity.
So, yeah, it *eventually* can live on a network as an AirPlay speaker, but if you can't connect the speaker to the LAN without asking a Sonos server to let you, it's still dependent on the cloud.
Re: (Score:1)
Sonos does run locally and not depend on the cloud. If you are suggesting that you should never buy anything with added cloud functionality above the core premise then you my as well just say "Don't by anything new ever."
Shit man my car has added cloud functionality.
Re: Never buy anything... (Score:2)
Why would I ever want a speaker to connect the Internet at all? It's not about having cloud features or relying on the cloud, it's about not poisoning one product with a shitty version of another product.
People who buy into this shit are like those who buy a TV/DVD combo then get annoyed when the DVD part breaks, rendering the TV useless.
Those are two separate devices. They should be treated as such.
Swiss army knives make shit knives.
Re: (Score:2)
Why would I ever want a speaker to connect the Internet at all?
Ahhh I take it you've not heard of the concept of streaming. You may not like streaming music but many other people do. For that you need to connect to a music source, and until Spotify sends out their library on a bunch of fat HDDs that necessitates a connection to the internet.
People who buy into this shit are like those who buy a TV/DVD combo then get annoyed when the DVD part breaks, rendering the TV useless.
No one is doing anything of the sort because just like in this case of the Sonos speakers, the TV in your analogy isn't useless and remains a perfectly functioning TV.
Those are two separate devices. They should be treated as such.
But why are they two separate devices? No really where is it mand
Re: Never buy anything... (Score:2)
I stream music all the time without hooking my speaker system up to the Internet.
And I mucked the analogy a bit. I meant to say when the TV dies, the DVD player is useless.
Re: (Score:2)
And when your speaker dies? When you integrate something for a single purpose it doesn't matter if the other device is standalone, it becomes useless. I'm not completely countering my own point from before because the people who buy a separate speaker and streamer bought that streamer to stream something. Integrating the internet into a speaker means if the speaker dies you can't stream Spotify. Keeping it separate means if the speaker dies you can't stream Spotify, or if your amp dies you can't stream Spot
Re: Never buy anything... (Score:2)
When my speaker dies, I replace the speaker...
Outsource all the programming to India. (Score:2, Flamebait)
Like Boeing, they outsourced their primary values to India. They saved a few pennies for a few years and now their core product and brand recognition looks like the average street in Delhi.
How could they fail to see this? (Score:2)
I can see Microsoft failing to check their update on Grub. It is not their product after all. (That is bad too, but okay)
Here Sonos basically has ~5 different product classes with the software: soundbars, speakers, subs, bluteooth, and headphones. They have to test it in a lab. It won't take too much resources or manpower.
Even Best Buy would let you test a Sonos system before you buy. They set up a mock home theater with all the surround speakers, and you can see whether it works... or in this case it does
We burned all the bridges (Score:2)
It's like a restaurant firing their chef, hiring a new one who proceeds to repla
Re: (Score:2)
Maybe they need a new CEO (Score:4, Insightful)
Re: Maybe they need a new CEO (Score:2)
This is not the CEO's job. This is the role of the CTO. A CEO is a high-level strategist and glorified salesman. Details like this are not part of their purview.
Re: (Score:2)
What a wasted opportunity (Score:2)
Any questions software-defined reality isn't real?
It’s not like Sonos didn’t have a hint that their software was its weakest link.Sonos spent years apologizing for its app’s poor UX.
Its Overton window just slammed Sonos out of its standalone audio products. It may salvage what’s left in perpetual service to major brands, rebadging and audio engineering for competent software companies.
Sonos multi room is amazing in a flat (Score:2)
I was living a large flat in London that had multiple Sonos units. I listened to so much more music and the experience of the music blending with the vibe of the flat was next level. I decided that when I get a permanent place I would get a similar setup.
Multi room music listening just hits different as you go about your routine and day.
Re: (Score:2)
Ask Me Anything? (Score:2)