I'd recommend folks read the last section of the announcement where these changes were announced [1]. The section is titled "Mission First", and I'm pretty sure the recent altercations at Google Cloud's offices over the past week [2] motivated much of the writing in this section. This seems like a stark change in how things happened at Google, and to put it explicitly in a blog was something I couldn't imagined to have happened in ~2017-2022.
> We have a culture of vibrant, open discussion that enables us to create amazing products and turn great ideas into action. That's important to preserve. But ultimately we are a workplace and our policies and expectations are clear: this is a business, and not a place to act in a way that disrupts coworkers or makes them feel unsafe, to attempt to use the company as a personal platform, or to fight over disruptive issues or debate politics. This is too important a moment as a company for us to be distracted.
Honestly, good on them. Having a culture of honest and open discussion about work is important, but there's a very vocal minority in many companies that thinks their political opinions are both objective fact and the most important thing to discuss at any given work function.
When that kind of attitude seems to receive official support from the company it actually does make people with different political opinions feel unsafe at work. This is not okay, and it's not healthy for the company, and I'm glad to see Google finally pushing back against the idea that loud political fights are appropriate in the workplace.
When the mission is make money within the existing institutional frameworks it's much easier to be apolitical and that seems to be the change here. But when your culture is explicitly "make the world better" you can't avoid getting mixed up in political issues.
How the devs felt about police for example likely affected the maps feature for reporting cops on the road. Same with Google Maps history at abortion clinics. The read on various news organizations definitely affected the choices for Google News partnerships. How you feel about government surveillance and "the deep state" likely affected how they built their messaging apps. Even down to the sign-up form where you're asked your name/sex. A conservative Google would have made very different decisions.
It's really hard to do anything non-trivial that doesn't end up brushing up against political issues de jour.
I'm not so sure. You point to a vocal minority, I see something amiss for the company.
If I were "corporate" I would be asking myself why it is a group of my employees faced arrest, job loss to make a statement about company policies. I would want to know if it suggests a bigger problem down the road for the company.
Google employs ~180k people. At that scale, even 50 employees involved in these protests would be 0.02%. That sure sounds like a very vocal minority to me.
> If I were "corporate" I would be asking myself why it is a group of my employees faced arrest, job loss to make a statement about company policies. I would want to know if it suggests a bigger problem down the road for the company.
I think that's exactly what corporate is doing. Corporate thought it through and came to the conclusion that this happened because Google created a culture where a certain type of employee thought that Google was the place to push their personal politics. They would like people who feel that strongly about a political issue to move on to another workplace and let their coworkers get on with their lives.
I find the use of "this is a business" logic to dissuade political discussion about a company's own business practices to be extremely troubling. Yes, there can be a discussion about the proper way to have these conversations, but dismissing discussion of ethics with an attitude of "this is a business" has lead to some horrific outcomes throughout history.
TJ Watson (of IBM) had a similar "this is a business" outlook: “I’m an internationalist. I cooperate with all forms of government, regardless of whether I can subscribe to all of their principles or not.” IBM's machines were extremely important part of Nazi Germany's Holocaust efforts, and there is evidence that IBM was actively working with Nazi Germany after the invasion of Poland [1].
Anyone notice the last few days tons of people bringing up the IBM/Holocaust connection? I've seen maybe half a dozen people in different threads bring it up. It almost feels coordinated, like those are the marching orders that were given out somewhere: "If this topic comes up, compare it to IBM and the holocaust, make these points, etc."
Google is being accused of writing software that facilitates the intentional killing of civilians. Whether you agree or disagree, that's what the accusation is. In that context, for people who accept the accusations, IBM seems like an obvious choice for an analogous company. Do you have a better comparison?
Stepping past all the AI hype, as an engineer in what was Play/Android/Chrome, I'm excited about being closer to hardware. The fact that Pixel was in another division was always weird and felt like it was an artifact of legacy decisions rather than the right way for things to work today.
>The fact that Pixel was in another division was always weird
Why? Having a separate division also has upsides, like keeping it away from terrible exec interference.
A lot of products we take for granted today succeeded because they were a sort of skunkworks project away form the reach of the mothership that's full of execs who would have tried to push their own agenda in the product or shut it down due to their lack of vision.
The original PlayStation, first Xbox, DirectX, Gameboy, etc.
I think there is value in having core android team that focused on AOSP separate from the product oriented pixel or apps teams.
And yes, in the end everyone end up using GMS (besides those who are banned, ie. Huawei). But still, it's better to have a separation that is imperfect, vs not having one at all
> The fact that Pixel was in another division was always weird
Was there a mechanism to communicate this weirdness to decision-makers? In a company that is in control of all of its teams, what is stopping a faster reorg?
Yes this was a well understood phenomenon. These sorts of things grow slowly over time though, and the orgs are large, so understandable why it took a while.
It was probably kept apart while Google was deciding if Pixel was going to survive and to not antagonize 3rd party Android phone OEMs. But all that is more locked-in now.
So the team that is in charge of the OS which is licensed to Hardware vendors in the world is the same team that's in charge to create competing Hardware?
I'd say that creates a huge conflict of interest.
That's one of the big reasons why Nokia Series60 didn't take off as a licensed OS: Whatever Samsung or LG or Lenovo wanted to build on that platform to differentiate, they had to involve Nokia during the development (who then developed the needed OS-feature in parallel to the Nokia product that will make use of it).
Google is either very secure that their grip on all these HW-vendors is strong enough forcing them to stay, or they are no longer part of Google's long-term strategy for Android.
The first priority for Android is competing against iPhone. Any self-dealing to get Pixel to have more of the Android pie would be far down the list, and probably counter-productive. It was already possible under the previous structure anyway.
Correct, and it's clear that this isn't an outdated strategy from Google, just in the current cycle we saw Circle to Search launch on both Galaxy and Pixel, and most of the new AI stuff that differentiates Pixel is now coming to Galaxy as well. This might be a headscratcher if you think Google is trying to make Pixel the dominant Android phone, but that's not it. Google wants Android to be the dominant phone OS, and despite it being massively popular globally, in the US the numbers are dire, with 50-60% overall going to Apple, and as high as 80-90% of young people choosing iPhone. I love Pixel, but it accounts for approximately 5% of the market in the US, with Samsung at 22%. Those stats about young people nearly universally picking iPhone is a bad sign for Android and a bad sign for competition in the phone market as a whole.
>as high as 80-90% of young people choosing iPhone
It's not really choosing, it's more like being handed over form parents or being force to due to iMessage network effect with teens in the US. Which teens wants to choose to be left out of group conversations?
As an adult you can give fewer fucks about normie conformism, bubble colors and people being petty over it, but as a teen it would be a death sentence for your social life. Hence why the regulatory bodies are starting to twist Apple's arm over it.
It's almost like everyone who comes in to say "people all just use Whatsapp" are going from anecdotal evidence, while the reality is Whatsapp, Snapchat, Instagram, Telegram, iMessage, SMS and Tiktok all have a share of that messaging pie-- these people just live in message app bubbles.
That's the situation NOW, but it can certainly change in the future. Work with Google as a hardware vendor, grow the Android market with them, and eventually they don't need you and cut you out. It happens.
Let's not fetch too far, this becomes a strawman argument.
They weren't compatible operating systems because applications compiled for one of them were unable to be executed on the other without heavy modifications.
At "underlying layers" the OS of a Tesla is compatible with that of a Nintendo Switch, and yet no one would say they have a compatible OS.
> It is like telling Samsung, Huawei or Xiomi aren't Android, because they use another GUI framework on top of AOSP.
No it's not, because they all use the same GUI framework as AOSP, hence they can run the same precompiled application.
"In terms of software, Satio uses the Symbian OS 9.4 operating system, which is created collaboratively under the stewardship of the Symbian Foundation as "Symbian^1".[4] It is Sony Ericsson's first non-UIQ Symbian device, after UIQ's development closed down earlier that year. It uses the PlayNow service, Sony Ericsson's mobile content platform, and is part of the company's new Entertainment Unlimited service.[5] In terms of connectivity, it is Wi-Fi-enabled and has a GPS chip for navigation and location-based services. It also supports full Flash for video playback.
They made a total of 2 devices with Symbian^1 (+1 refresh) before ultimately moving on to Android.
Symbian^1 was the attempt to harmonize the Symbian flavors into one open platform in order to compete, after Nokia, Sony Ericsson, Motorola and other contributed all IP into the Symbian Foundation.
Nokia called it "S60 5th Edition", for others it was "Symbian^1", and the first "common" Symbian OS.
So yes, I wasn't precise enough in the later comments, and you're correct.
Conflict of interest is a human invention, and not a law of the universe. There is no "conflict" unless you see it as one.
You can play chess against yourself. AlphaGo can, because it wasn't brainwashed about this notion. ChatGPT can debate against itself. You can too, if you don't see it as a conflict. Humans might find it hard, only because they were brainwashed from a child that they need to pick sides. Your neural net is capable of operating on both sides simultaneously if you let it.
The market is big enough for Google to create hardware AND other companies to create hardware.
> The market is big enough for Google to create hardware AND other companies to create hardware.
You obviously didn't read the comment you're replying to. No one is challenging that.
Having the same TEAM in charge of the OS and in-house hardware is an entirely different story.
It's a conflict of interest because the person Samsung is talking to to have a feature implemented into the OS baseline may be the same person in charge of defining the competitive featureset for the next Google hardware.
Now this person knows that the product he and his team is designing will compete with a yet-to-be-announced Samsung-product with a new feature.
So his interest to support a licensee being successful with his product is in conflict with his interest to create a more successful competing product.
And even if he isn't, for SAMSUNG just the potential of this situation to happen can be enough to NOT cooperate with this team and scale back communications with the Android team as a whole.
Weren't the 2 teams already part of the same company? And OEMs do make custom modifications to Android before shipping their devices, so they don't have to share everything they intend to do with the Android team.
>It's a conflict of interest because the person Samsung is talking to to have a feature implemented into the OS baseline may be the same person in charge of defining the competitive featureset for the next Google hardware.
The person can run two threads in their brain, one that deals with Samsung and one that deals with the internal product.
> Now this person knows that the product he and his team is designing will compete with a yet-to-be-announced Samsung-product with a new feature.
So? You're talking about the person, which is just the host hardware. There can be multiple threads running on that hardware at the same time in containers.
> So his interest to support a licensee being successful with his product is in conflict with his interest to create a more successful competing product.
They can both be successful at the same time. He can operate with an interest to optimize for an overall better world rather than interest to win over and kill Samsung. He can build a successful product AND help Samsung build a successful product at the same time.
> The person can run two threads in their brain, one that deals with Samsung and one that deals with the internal product.
Great find! Does this also apply to a police officer investigating a crime where their spouse is a suspect? Or to a judge presiding over a court case involving themselves?
>The person can run two threads in their brain, one that deals with Samsung and one that deals with the internal product.
There's a reason we don't make a police chief investigate their own misconduct.
"It's a perfectly OK thing to do. The person is just the host hardware. There can be multiple threads running on that hardware at the same time. They can run two threads in their brain, one that deals with investigating the case and a seperate one that might or might not did it".
"He didn't just declare himself innocent of misconduct and embezzelment out of self-interest. The independent investigating "thread" must have arrived to an impartial decision".
"In any case, it's not blatant misconduct, you only see it as such. There's no notion of misconduct in nature, it's a made up thing we invented".
If there's accusation of misconduct, there's a bug in the system so you isolate it and investigate it from the outside.
There's no investigation happening here, just two happy parties trying to create great products that can both be successful and be even happier. Lawyers can stay out of this happiness, inventing and injecting "conflicts" that never existed in the first place.
And in any case who (and by whom) is going to be assigned from outside to investigate any misconduct in this Google-other Android vendors case? Any misconduct of prioritizing their phones other third party hw in Android wouldn't be a crime, just a bad deal for the third party vendors.
Maybe confused? I'm not the person you quoted and replied to (that was dheera).
I'm the person disagreeing with dheera in this thread, and I expanded upon what you asked, adding (to further refute dheera) that there wouldn't be any investigation from the outside.
He may also "operate" with 50% of his bonus depending on Google Hardware doubling in market-share.
How much would you bet to win against me in a card-game if you have to show me all your cards and I show you none?
Rest assured, I will maintain the task to beat you in another container than all the details I need to beat you.
> So? You're talking about the person, which is just the host hardware. There can be multiple threads running on that hardware at the same time in containers.
>Conflict of interest is a human invention, and not a law of the universe. There is no "conflict" unless you see it as one.
That could be said for anything in the moral and judicial sphere. "There's no theft, property is a human invention", "There's no rape, animals don't have that concept", and so on.
Conflict of interest has been a thing way before lawyers and HR types existed, they understood it and tried to prevent it at any point in history, from ancient Babylon to Rome, and from Amazon native tribes to imperial China.
It's of course also the explicitly expressed reasoning for why there are independent branches of government (legislative, executive, and judicial in the US).
>I'm an optimistic engineer, believe in win-win situations, and don't see everything as a conflict.
> Conflict of interest is a human invention, and not a law of the universe.
Lucky for us, we're discussing this in the context of humans building stuff for other humans to buy in a human society with human governments and markets, not in some metaphysical 'but what does meaning means' context.
It feels like a large bundle to me, so they probably want to go for selling phones and computers much more heavily? Sounds interesting if you are okay with going all in with Google stuff, I'm not sure it's good news if you are using Android or Chrome otherwise.
Also, it feels like this merger will lead to a similar article to Hixie's in about 5 years:
> A symptom of this is the spreading contingent of inept middle management. Take XYZ, for example, who manages the department that somewhat arbitrarily contains (among other things) Flutter, Dart, Go, and Firebase. Her department nominally has a strategy, but I couldn't leak it if I wanted to; I literally could never figure out what any part of it meant, even after years of hearing her describe it.
I attended the Android Dev summit a few years back, shortly after Flutter had made some public noise.
During a Q&A, I was literally laughed at when I asked the head of Android whether developers should take Flutter seriously. His eventual answer equated to "well Google is really big so we can't say".
I think that was the moment I understood just how deep the mismanagement at Google actually is. Just shocking.
None of this reorganization matters as long as the executive suite remains the same. Google's problem is top-down. This rearranging the deck chairs isn't going to cut it.
Much of Google's value proposition was in their ability to innovate. The current leadership has proven catastrophically inept at this. Without fresh leadership, Google is on its way to becoming some mid-tier ad placement agency at best. Engineering excellence won't save them as long as the top leadership is incapable of adjusting to circumstances.
There's still time for Google to leverage its competence at managing large infrastructure to regain its position as a leading technology company. But the window of opportunity is shrinking. Alphabet's board of directors needs to fire the executive suite, that's the easy part. The harder part is finding a replacement CEO and executive suite who will make the deep cuts and rearrangements necessary to get Google back on track. The longer they put that off, the less chance for Google to be relevant in the future.
It remains to be seen, but I don't really see it this way.
To me, this feels like Hiroshi went to Sundar and said that he wanted to step down or wanted to do something else or whatever. Sundar then had to choose whether to elevate one of Hiroshi's reports or elevate one of the other SVPs to lead two orgs together. Sundar chose Rick.
This feels more like a question of upper management politicking rather than mission change.
I guess it makes them more like Apple, having a vertically integrated division for making phones. TFA says it might make other phone manufacturers struggle. Although I get the impression they are already struggling with the Open Handset Alliance terms from Google that they don't like. Maybe the best outcome is that AOSP gets multiple active forks supported by manufacturers, Google apps stop being distributed by default, and the phone software ecosystem gets more decentralized in general.
Forks supported by manufacturers don't work, because they only earn money when selling hardware. So they can't each operate a huge platform maintenance team on their own.
Also, the only glue that actually holds Android in place as a single platform is Google's CTS (compatibility test suite).
Without it being mandated for Googles Mobile Services (GMS) and its revenue-share, Android will stop being a single platform.
It will start drifting apart as soon as all vendors have to implement the next display/camera/sensor/form-factor support in the OS in parallel of each other...
"Merging teams" is such a Google thing. Merging DeepMind and Google AI, Waze and Maps, Fitbit & Nest and Pixel come to mind. I don't remeber reading such stories from other companies. Is my perception off or is "merging teams" something that Google likes to announce loudly and other companies don't or do other companies "merge teams" less often? I would like to have some input to this.
I think most other companies have not such a high amount of redundant teams to merge them all the time. So while merging happens occasionally in other companies, we usually only hear about them in context of one company buying out another one.
Not many other companies have the unique blend of being out of touch enough to think that anyone knows what this means and reality, and self-important enough to think that people care.
Are you anthropomorphising companies now? Google the company can't be "out of touch" or "self-important". The engineers maybe, but they don't make PR decisions like this.
In this case the original source is an announcement on a "inside-google" blog, and starts with "Hi Googlers", so the target are clearly mostly Google engineers. It ended up on HN because it was repeated by arstechnica, a respected and well-known portal. And here we are discussing it. So maybe people in fact do care about this?
Companies are groups of people, they're maintained by people, all their decisions are made by people and they only ever do things through their people's hands. They are as anthropomorphic as it gets.
I don't necessarily disagree, but structurally these moves can be very impactful, particularly in a huge company as sprawling and with as much overlap as Google.
If there's a lot of redundant work being done, maybe the teams will benefit from working together on the same problems.
You're right that most of us don't care, necessarily, but I think it sends some signals that the company is attempting to focus a bunch of less focused lasers at the same point.
It's less common in tech companies than in massive multinationals like GE or Sony that span a lot of different industries.
I'm not sure how I feel about this, as a user of google's services and as an owner of google's hardware (pixel 5, nest hub 2). I'm probably cautiously optimistic, seeing how high quality yet unique/quirky their hardware has been. However Tensor/Samsung fabs have had their issues, but maybe factors may have been out of the hands of those in charge?
While I think Google needs a better clear vision in many cases, god help us that the people who have screwed up every hardware launch for a decade now get to run the OS too.
I'm curious to see how the new HardChromeDroid division will fare. I cannot imagine that lumping hardware and software together in one division is a particularly good idea, but I'm open to surprises.
They already did, in DSPA. The (dysfunctional) PA that brought out the Nest products, etc.
Now they're swallowing Android and Chrome.
In theory, this could lead to executing in a consistent way. E.g. not having 5+ distributions of Linux and competing platforms fighting over who will rule on some new product.
My thoughts on this, is how will the Hardware convenience, with the Chrome-track-every-breath will dictate the Android/Operating System.
Android as-is does enough monitoring, and reporting back home. I fear it will make it worse for the Android users of non-Google devices, in 3-to-5 years.
In terms of revenue relative to their search / ads biz, how big of a deal is this? I understand there's name recognition for Chrome and Android, but the Pixel product line is nothing when compared to the iPhone or even Samsung.
Is this more of just Google being Google, as the ads side of the house continues to print more money than God?
I read it as an erosion of the implicit assumption that ads bring in the money, and projects like Android and Chrome spend it, indirectly empowering the ads division. "Why can't we just focus on eating the cake? Those guys in the bakery with their funny hats and food-grade tooling should grow up and learn to stand on their own feet without our handouts." Pixel hardware is the closest thing Google has to a division that could sustain itself from revenue streams unrelated to ads.
Probably makes a good impression on shareholders who'd prefer to see all divisions disinvested but the money counting one.
Pixel hardware fails because they don't have good distribution channels. No mobile contracts include a pixel phone. Many mobile contracts have iphones or samsungs attached.
They are failing in being present in markets; Google won't sell in more than half of EU countries at all; never mind the countries outside EU, Google sells only in a handful.
Meanwhile, Samsung products are widely available and marketed everywhere.
Consumer products are demand-driven. This is like saying my shrimp-flavored cola is failing because I don’t have distribution deals.
Pixel is great hardware, but Google has failed to generate enough demand for consumers to be going into stores and the n leaving because Pixel isn’t sold there. That’s what gets distribution in the long run.
Pixel phones have lots of geoblocked features. For example, they lose most of their 5G/VoLTE/VoWifi abilities if you use a sim from a phone network in an unsupported country, even though other models of phone don't have that issue.
A bunch of other features are geoblocked in the EU, or geoblocked to the US and Canada only.
Sure, but my point is that 1) Pixel has low market share in markets it is available, and 2) there does not appear to be any clamoring for availability from the masses in other markets.
Pixel is a great product for tech enthusiasts. It may even be a good product for the mass market. But bemoaning the lack of sufficient distribution is totally missing the point.
>Pixel hardware fails because they don't have good distribution channels.
No, Pixel HW fails because it's shit: weak call signal reception and 5G performance due to poor Samsung modem, low performance compared to Apple and Qualcomm, high battery drain, overheating, software bugs that get in the way of you dialing calls, dimples in OLED display, poor design with ultra thick visor, questionable warranty support, AI features only exclusive to the US market but not EU making them useless, etc. Much of the issues that do not exist on Apple and Samsung phones.
Carriers go with Apple and Samsung because those brands are a slam dunk with consumers, are well known and mostly reliable by this point with fewer lemons, and when they do fail, there's service centers for them everywhere. Pixels are mostly bought by enthusiasts who will forgive the rough edges for the chance of being Google's software beta testers.
Pixels can still have better photo/video quality and less shutter lag than Samsungs and a cleaner, more unified SW ecosystem with only the Google Apps, versus Samsung who duplicates those apps with their own: Samsung Pay, Samsung Browser, Samsung Notes, etc. Some people prefer Samsung's own apps to Google's(their browser is really nice), others find it bloatware.
The main problem is that due to their pricing, bugs and issues, Google's recent Pixels are no longer a slam dunk reccomendation like they used to be in the old days when they rocked Qualcomm chips and were compact, well built, dirt cheap phones.
Google needs to properly absorb Android into Chrome before any antitrust measure forces it (along with the Play Store) to be spun out, as would be most healthy for absolutely everyone except the Google ads business.
There has been so much potential squandered with Android over the last decade it is amazing.
I think that this unification simply took place because Google Android in the future will only be the top layer and Fuchsia OS will manage phones, etc., so such a merger is needed to prepare for new changes.
I think that this unification simply took place because Google Android in the future will only be the top layer and Fuchsia OS will manage phones, etc., so such a merger is needed to prepare for new changes.
What for? The market share of Google’s hardware is minuscule, and other Android OEMs are unable to take advantage of Google stuff since they use their own chips.
I hope ideas like "regulators break them up" never come true. I'll never understand why people crave the destruction of productive organizations. Android has been a stunning success. Valuable for consumers who enjoy the platform, profitable for the investors who bet on the platform and lucrative for the employees who work on the platform. The only people that seem to have a problem with Android are misguided ideologues who think that "big company == bad".
I don't think your understanding of this is complete. A break up doesn't destroy anything; on the contrary, it creates value. Where once there was a single monopolist, there are now two (or three or twelve) organizations that can proceed independently of each other, and be much more focused on their core product.
For example, maybe Google can keep the play store, but not Android. Android would become an independent entity, and can develop in ways that benefit all Android stakeholders instead of just Google. Maybe then Android will finally be able to focus on competing with iOS in ways that Google would never dare to (since Google's relationship with Apple is a bit sus)
A breakup can also be good for investors, because illegal monopolies are inefficient. Android could potentially be much more profitable on its own than under Google's umbrella, and the play store could be more profitable if they're actually forced to compete. This can lead to innovations which increase revenue, and which never would've happened under Google.
The world needs a third party software and hardware stack that isn’t controlled by big tech walled garden monopolists / authoritarians. Not just for phones but laptops and computers too. As far as phones go, unfortunately the best alternative I’ve heard of is Graphene and the best phone for Graphene is the Pixel series. And I assume using it as a daily driver is problematic without access to various apps or maybe if websites block them or even carriers - not sure.
This is why my next tech purchase will be a Raspberry Pi 5 and a Wacom One 13" touch screen --- my testing with a Raspberry Pi 4 and Wacom One (gen 1, no touch) went well, so I'm hopeful that this will work as well.
At least it's a company which will allow alternative OSs.
I'm still quite annoyed that I set aside and stopped using my Samsung Galaxy Book 12 because I couldn't figure out how to get it to boot something other than Windows.
It is unsurprising to see that RickO beat Hiroshi for the title of grand poobah of devices and platforms. Hiroshi always made the impression of a very smart guy. Rick always made the impression of a good politician. Politicians always win
The 'consumer data extractor' (hardware) team is teaming up with the 'consumer manipulator' (AI) team, and all free of any inconvenient 'dont be evil' policy.
I recently got a new Pixel phone and Google's much hyped new AI features just seem so... gimmicky. One of the setup examples shows how you can circle a tent on the left side of a picture and move it more to the center. Neat, I guess? It's kind of a fun toy but I'm not sure what problem this is actually solving. I'm sure there are some usecases for this out there, but it's not a capability I have ever found myself wishing for.
Meanwhile the rest of the phone is surprisingly buggy and annoying. Basic functionality I use every day is worse than on any other recent phone.
Google has never been a great product org, but this desperate need to be seen as one of the cool kids in AI is making things worse. Granted I think of phones/computers more as a tool than a toy and put much higher value on usability and reliability versus novelty; perhaps I'm outlier in that.
>I recently got a new Pixel phone and Google's much hyped new AI features just seem so... gimmicky
Not just that, but their biggest crime is that almost none of those fancy AI features Google paraded at the Pixel launch even actually run on-device but need to be sent to their cloud for processing, despite all the gloating about their new Tensor 3 chip's AI capabilities being the most important (since that chip sucks at CPU and GPU benchmarks compared to Apple and Qualcomm). Also, their Tensor 3 can't even run Google's smallest LLM. Absolutely embarrassing.
They REALLY need to unify the HW and SW development efforts to create a coherent and functional product, instead of designing them separately bazaar style then jerry rigging them together like some underfunded start-up making products for Kickstarter.
I can't stand all the "AI" junk, especially when things worked better in the past. My pet peeve: I used to be able to ask google maps while I was driving "What's the E.T.A.?" and it would respond with, you know, the answer. It's been broken for many years now and responds with nonsense.
Another one: I can't tell my phone to change it's name to what I want. Basic "AI" fail.
honestly, it seems like every phone has its broken quirks. I recently switched from iphone to android and there's still a random collection of everyday things I do that are simply... broken.
Maybe these devices have become so complicated they're simply too challenging to work out all of the edge cases out of. New features are easier.
>Maybe these devices have become so complicated they're simply too challenging to work out all of the edge cases out of. New features are easier.
With the amount of telemetry and data Google is collecting I doubt they can't catch edge cases, let alone recurrent bugs that impact multiple users.
I wanted to buy a Pixel on sale last week but I watched a 6 month long term review of the Pixel and the reviewer complained that every new update fixed some bugs but added it's own new bugs.
It's why I'm still gonna keep using a phone that stopped getting updates over a year ago: it's finally stable and no more new bugs are being introduced by updates, as my mind and muscle memory has already adapted to the old bugs.
Maybe I'm getting old but while 10 years ago I couldn't wait for new major updates to arrive on my phone, I feel like phone SW has peaked a few years ago and has been on a constant decline ever since, with new updates just adding useless crap that bugs you and changing things for the sake of change without improving them, and I would much rather have a phone that only updates security but nothing else. Basically I don't want my phone to be a Googler's playground and me being the beta-tester.
Not to mention all the Google shit on Android is just constantly harassing you. LOOK HERE, LOOK THERE, SET UP THIS, SET UP THAT, TURN THIS ON, TURN THAT ON, GIVE US FEEDBACK, SYNC NOW, SIGN IN TO GOOGLE, LET US SCAN YOUR PHONE FOR YOUR SAFETY, SIGN IN FOR SECURITY, SYNC YOUR PHOTOS, SYNC YOUR DRIVE, USE AI FOR THIS, HERES HOW YOU DO THIS THING BECAUSE ITS NOT OBVIOUS AND WE SUCK AT UI, GET UPDATES, WE DISABLED PERMISSIONS ON OLD APPS, WE DID THIS FOR YOU, WE DID THAT FOR YOU, POST PICTURES OF YOUR RECENT HOME DEPOT TRIP.
Jesus christ, I've had to dismiss at least 20 different popup things just in the Messages app since I reset my phone a few days ago. Just fuck off already!
And guess what. After resetting the phone, I still can only make a successful outgoing phone call 1 out of every 3 tries, and it will only work after a reboot. It worked fine after the reset for about a day. Now, again, it barely works as a phone.
Rodney Dangerfield was right. There is no fucking respect for the people using the phones. There is only respect for the stocks going up. Fuck you and give us money, that's what smartphones are all about.
Yesterday I ordered a Nokia flip phone. I'm done with iOS and Android. It has added nothing to my life except distractions and maintenance. I spent 3 days trying to get this piece of trash to work as a phone. Just a total waste of my life.
> Under Rick Osterloh, a new platforms and devices team will be dedicated to bringing AI to your phone, your TV, and everything else that runs Android.
Who is asking for this? Why can't they just make their search engine work again?
Nah, I want them to first fix the basic shit I actually (would) use.
For example, when I'm driving and a timed phone alarm goes off for the Android phone in my pocket, I ask it to silence the alarm, yet instead rebukes me by falsely claiming no alarms are active right now.
It's fixed now that I checked, but for a while it would also secretly ignore the date that I already specified for a scheduled event while it was prompting me to clarify the time of day.
You mean the thing that requires an internet connection or it doesn't work?
And probably will continue needing internet for the foreseeable future, regardless of how many mobile Tensor-chips they develop, because cloud data and compute power will always be orders of magnitude better than your phone?
That's the thing they need specialized mobile-hardware teams involved for?
Google cannot learn that kind of lesson, the people who were in Google+ are probably long gone from Google for greener pastures most likely. Second, google is an amorphous giant with no other goals than increase market share, no wonder their products are terrible. But still, with all this in mind, we should remember that great things go come from google, it's just that they're not capable of capitalizing on them.
insightful comment. I'd like to think that Google will integrate AI in an at least mostly useful way, but it's clear that large scale reorgs like this around chasing a competitor can and do end badly.
Yawn. Get back to me when something of actual note gets launched.
Mobile OSes are now a boring, stable environment. All this noise about AI seems like an attempt to convince investors that some paradigm-shifting change is on the way. It isn't. A mildly better Google Assistant is on the way.
The comments here are trending towards "stop cramming 'AI' into everything". I am curious how the end-user consumer (versus 'AI' for enterprise/business) differ in experience and use. We are in the beginning of this AI-fication, and it seems deep learning models are doing really well and that DL can predictably scale [1.]; therefore, do we have to wait a bit for really life-changing AI for the end-user consumer?
I can see AI in enterprise/business being extremely useful in different industries, but at the same time, is the current 'AI' actually good/useful for the end-user consumer?
Hiroshi, who's been an Android lead since before it shipped, is no longer leading Android.
When Jony Ive and Scott Forstall and the other big Apple execs left, that was news. Hiroshi may not have Jony's profile, but it's still a major change in how Android is governed.
Treat yourself to multiple UI lockups, missing home buttons, stuck notifications in the middle of the screen, etc every single day if you disable animations.
"Google's previous head of software platforms like Android and ChromeOS, will be headed to "some new projects" at Google"
Ok so they're more than willing and able to shut down divisions and relocate human resources to "new projects".
Why not just do that with Nimbus? At least ChromeOS didn't put Google in the position of being complicit in a plausible ........
Strange decision making, shutdown and kill things people love (e.g Google Domains, RSS, etc.) but not shutdown a project that puts the company at risk and that some of their own workers actively protest against.
It's not nearly as momentous as removing Andy Rubin. Android has evolved a lot. From an OS technology PoV it's less distinctive and more about competitive parity with iOS. Kind of boring.
Tesch and Stabenow probably made a lot of money with their government contracts also back in the day. Doesn't negate the fact that the way they made that money was wrong.
I'd recommend folks read the last section of the announcement where these changes were announced [1]. The section is titled "Mission First", and I'm pretty sure the recent altercations at Google Cloud's offices over the past week [2] motivated much of the writing in this section. This seems like a stark change in how things happened at Google, and to put it explicitly in a blog was something I couldn't imagined to have happened in ~2017-2022.
[1] - https://blog.google/inside-google/company-announcements/buil...
[2] - https://www.cnbc.com/2024/04/17/google-workers-arrested-afte...
> We have a culture of vibrant, open discussion that enables us to create amazing products and turn great ideas into action. That's important to preserve. But ultimately we are a workplace and our policies and expectations are clear: this is a business, and not a place to act in a way that disrupts coworkers or makes them feel unsafe, to attempt to use the company as a personal platform, or to fight over disruptive issues or debate politics. This is too important a moment as a company for us to be distracted.
Honestly, good on them. Having a culture of honest and open discussion about work is important, but there's a very vocal minority in many companies that thinks their political opinions are both objective fact and the most important thing to discuss at any given work function.
When that kind of attitude seems to receive official support from the company it actually does make people with different political opinions feel unsafe at work. This is not okay, and it's not healthy for the company, and I'm glad to see Google finally pushing back against the idea that loud political fights are appropriate in the workplace.
When the mission is make money within the existing institutional frameworks it's much easier to be apolitical and that seems to be the change here. But when your culture is explicitly "make the world better" you can't avoid getting mixed up in political issues.
How the devs felt about police for example likely affected the maps feature for reporting cops on the road. Same with Google Maps history at abortion clinics. The read on various news organizations definitely affected the choices for Google News partnerships. How you feel about government surveillance and "the deep state" likely affected how they built their messaging apps. Even down to the sign-up form where you're asked your name/sex. A conservative Google would have made very different decisions.
It's really hard to do anything non-trivial that doesn't end up brushing up against political issues de jour.
I'm not so sure. You point to a vocal minority, I see something amiss for the company.
If I were "corporate" I would be asking myself why it is a group of my employees faced arrest, job loss to make a statement about company policies. I would want to know if it suggests a bigger problem down the road for the company.
Google employs ~180k people. At that scale, even 50 employees involved in these protests would be 0.02%. That sure sounds like a very vocal minority to me.
> If I were "corporate" I would be asking myself why it is a group of my employees faced arrest, job loss to make a statement about company policies. I would want to know if it suggests a bigger problem down the road for the company.
I think that's exactly what corporate is doing. Corporate thought it through and came to the conclusion that this happened because Google created a culture where a certain type of employee thought that Google was the place to push their personal politics. They would like people who feel that strongly about a political issue to move on to another workplace and let their coworkers get on with their lives.
Putting stuff in blogs has been the norm for several years now, as basically any email from Sundar inevitably leaks anyway.
I find the use of "this is a business" logic to dissuade political discussion about a company's own business practices to be extremely troubling. Yes, there can be a discussion about the proper way to have these conversations, but dismissing discussion of ethics with an attitude of "this is a business" has lead to some horrific outcomes throughout history.
I highly recommend listening to this podcast about IBM's role in Nazi Germany https://hbr.org/podcast/2019/11/lessons-from-ibm-in-nazi-ger...
TJ Watson (of IBM) had a similar "this is a business" outlook: “I’m an internationalist. I cooperate with all forms of government, regardless of whether I can subscribe to all of their principles or not.” IBM's machines were extremely important part of Nazi Germany's Holocaust efforts, and there is evidence that IBM was actively working with Nazi Germany after the invasion of Poland [1].
[1] https://www.sfgate.com/opinion/article/The-business-of-makin...
Anyone notice the last few days tons of people bringing up the IBM/Holocaust connection? I've seen maybe half a dozen people in different threads bring it up. It almost feels coordinated, like those are the marching orders that were given out somewhere: "If this topic comes up, compare it to IBM and the holocaust, make these points, etc."
Google is being accused of writing software that facilitates the intentional killing of civilians. Whether you agree or disagree, that's what the accusation is. In that context, for people who accept the accusations, IBM seems like an obvious choice for an analogous company. Do you have a better comparison?
because it's a wildly obvious connection to make? more apt comparison than BMW or Chase
Stepping past all the AI hype, as an engineer in what was Play/Android/Chrome, I'm excited about being closer to hardware. The fact that Pixel was in another division was always weird and felt like it was an artifact of legacy decisions rather than the right way for things to work today.
>The fact that Pixel was in another division was always weird
Why? Having a separate division also has upsides, like keeping it away from terrible exec interference.
A lot of products we take for granted today succeeded because they were a sort of skunkworks project away form the reach of the mothership that's full of execs who would have tried to push their own agenda in the product or shut it down due to their lack of vision.
The original PlayStation, first Xbox, DirectX, Gameboy, etc.
I think there is value in having core android team that focused on AOSP separate from the product oriented pixel or apps teams.
And yes, in the end everyone end up using GMS (besides those who are banned, ie. Huawei). But still, it's better to have a separation that is imperfect, vs not having one at all
I don’t think most of GMS works behind the great firewall right? So technically, all Android phones sold in China should be using something else.
> The fact that Pixel was in another division was always weird
Was there a mechanism to communicate this weirdness to decision-makers? In a company that is in control of all of its teams, what is stopping a faster reorg?
- Politics
- moats
- "if we do this, XYZ is going to leave"
- weak leadership
- comfort zone
- too many managers risking being spotted as obviously redundant as a result of a reorg :)
What else?
Yes this was a well understood phenomenon. These sorts of things grow slowly over time though, and the orgs are large, so understandable why it took a while.
You assume this is 'just weirdness', but in the thread You have multiple examples that there were upsides to this separation as well
Was done as pixel made negative profit while android made positive profit. Even now I’m not sure pixel is bringing any meaningful revenue to Google.
Google sold just 10 million pixel phones last year. That’s not even 10% of Samsung.
It was probably kept apart while Google was deciding if Pixel was going to survive and to not antagonize 3rd party Android phone OEMs. But all that is more locked-in now.
So the team that is in charge of the OS which is licensed to Hardware vendors in the world is the same team that's in charge to create competing Hardware?
I'd say that creates a huge conflict of interest.
That's one of the big reasons why Nokia Series60 didn't take off as a licensed OS: Whatever Samsung or LG or Lenovo wanted to build on that platform to differentiate, they had to involve Nokia during the development (who then developed the needed OS-feature in parallel to the Nokia product that will make use of it).
Google is either very secure that their grip on all these HW-vendors is strong enough forcing them to stay, or they are no longer part of Google's long-term strategy for Android.
The first priority for Android is competing against iPhone. Any self-dealing to get Pixel to have more of the Android pie would be far down the list, and probably counter-productive. It was already possible under the previous structure anyway.
Correct, and it's clear that this isn't an outdated strategy from Google, just in the current cycle we saw Circle to Search launch on both Galaxy and Pixel, and most of the new AI stuff that differentiates Pixel is now coming to Galaxy as well. This might be a headscratcher if you think Google is trying to make Pixel the dominant Android phone, but that's not it. Google wants Android to be the dominant phone OS, and despite it being massively popular globally, in the US the numbers are dire, with 50-60% overall going to Apple, and as high as 80-90% of young people choosing iPhone. I love Pixel, but it accounts for approximately 5% of the market in the US, with Samsung at 22%. Those stats about young people nearly universally picking iPhone is a bad sign for Android and a bad sign for competition in the phone market as a whole.
>as high as 80-90% of young people choosing iPhone
It's not really choosing, it's more like being handed over form parents or being force to due to iMessage network effect with teens in the US. Which teens wants to choose to be left out of group conversations?
As an adult you can give fewer fucks about normie conformism, bubble colors and people being petty over it, but as a teen it would be a death sentence for your social life. Hence why the regulatory bodies are starting to twist Apple's arm over it.
> as a teen it would be a death sentence
No wonder teens these days have extremely high rates of anxiety & depression.
Honestly it's weird how this is a thing in the US. In what I believe is most of the world, people just use WhatsApp, no sms, no iMessage.
Same is true for any demography as well. Maybe add some Instagram for the younger groups.
Younger groups mostly message on Snapchat not Instagram. Instagram is still mostly for pictures and "influencing".
It's almost like everyone who comes in to say "people all just use Whatsapp" are going from anecdotal evidence, while the reality is Whatsapp, Snapchat, Instagram, Telegram, iMessage, SMS and Tiktok all have a share of that messaging pie-- these people just live in message app bubbles.
Well at least where I grew up Snapchat never really caught on. Instagram kinda held its place.
I sure hope you're right!
That's the situation NOW, but it can certainly change in the future. Work with Google as a hardware vendor, grow the Android market with them, and eventually they don't need you and cut you out. It happens.
An american company is never going to be able to produce phones as cheaply as a korean or a chinese company.
Let me assure you, Apple is capable to produce a cheaper device than any other smartphone vendor in the world today.
In concert with Foxconn and others.
Further, much of android could be forked.
It took off for Sony and Ericsson.
What took off?
(Sony) Ericsson used UIQ, a pen-based OS built on top of the core of Symbian foundation.
Nokia developed Series60, a key-based OS built on top of a Symbian core.
They were not compatible operating systems, and most of all Ericsson didn't license it from Nokia.
As Nokia alumni I disagree.
They weren't compatible at UI widgets level, but were at the underlying layers.
It is like telling Samsung, Huawei or Xiomi aren't Android, because they use another GUI framework on top of AOSP.
And as many Android developers are painfully aware, that isn't the only customisations to AOSP standard behaviours.
Let's not fetch too far, this becomes a strawman argument.
They weren't compatible operating systems because applications compiled for one of them were unable to be executed on the other without heavy modifications.
At "underlying layers" the OS of a Tesla is compatible with that of a Nintendo Switch, and yet no one would say they have a compatible OS.
> It is like telling Samsung, Huawei or Xiomi aren't Android, because they use another GUI framework on top of AOSP.
No it's not, because they all use the same GUI framework as AOSP, hence they can run the same precompiled application.
Maybe you are mixing up your Sony Ericsson phones? There were loads of key based S60 phones too from them, not just UIQ.
Name one please.
Since you are asking for one.
Sony Ericsson Satio
"In terms of software, Satio uses the Symbian OS 9.4 operating system, which is created collaboratively under the stewardship of the Symbian Foundation as "Symbian^1".[4] It is Sony Ericsson's first non-UIQ Symbian device, after UIQ's development closed down earlier that year. It uses the PlayNow service, Sony Ericsson's mobile content platform, and is part of the company's new Entertainment Unlimited service.[5] In terms of connectivity, it is Wi-Fi-enabled and has a GPS chip for navigation and location-based services. It also supports full Flash for video playback.
https://en.wikipedia.org/wiki/Sony_Ericsson_Satio
Good find, all praise to you.
They made a total of 2 devices with Symbian^1 (+1 refresh) before ultimately moving on to Android.
Symbian^1 was the attempt to harmonize the Symbian flavors into one open platform in order to compete, after Nokia, Sony Ericsson, Motorola and other contributed all IP into the Symbian Foundation.
Nokia called it "S60 5th Edition", for others it was "Symbian^1", and the first "common" Symbian OS.
So yes, I wasn't precise enough in the later comments, and you're correct.
Conflict of interest is a human invention, and not a law of the universe. There is no "conflict" unless you see it as one.
You can play chess against yourself. AlphaGo can, because it wasn't brainwashed about this notion. ChatGPT can debate against itself. You can too, if you don't see it as a conflict. Humans might find it hard, only because they were brainwashed from a child that they need to pick sides. Your neural net is capable of operating on both sides simultaneously if you let it.
The market is big enough for Google to create hardware AND other companies to create hardware.
> The market is big enough for Google to create hardware AND other companies to create hardware.
You obviously didn't read the comment you're replying to. No one is challenging that.
Having the same TEAM in charge of the OS and in-house hardware is an entirely different story.
It's a conflict of interest because the person Samsung is talking to to have a feature implemented into the OS baseline may be the same person in charge of defining the competitive featureset for the next Google hardware.
Now this person knows that the product he and his team is designing will compete with a yet-to-be-announced Samsung-product with a new feature.
So his interest to support a licensee being successful with his product is in conflict with his interest to create a more successful competing product.
And even if he isn't, for SAMSUNG just the potential of this situation to happen can be enough to NOT cooperate with this team and scale back communications with the Android team as a whole.
Weren't the 2 teams already part of the same company? And OEMs do make custom modifications to Android before shipping their devices, so they don't have to share everything they intend to do with the Android team.
>It's a conflict of interest because the person Samsung is talking to to have a feature implemented into the OS baseline may be the same person in charge of defining the competitive featureset for the next Google hardware.
The person can run two threads in their brain, one that deals with Samsung and one that deals with the internal product.
> Now this person knows that the product he and his team is designing will compete with a yet-to-be-announced Samsung-product with a new feature.
So? You're talking about the person, which is just the host hardware. There can be multiple threads running on that hardware at the same time in containers.
> So his interest to support a licensee being successful with his product is in conflict with his interest to create a more successful competing product.
They can both be successful at the same time. He can operate with an interest to optimize for an overall better world rather than interest to win over and kill Samsung. He can build a successful product AND help Samsung build a successful product at the same time.
> The person can run two threads in their brain, one that deals with Samsung and one that deals with the internal product.
Great find! Does this also apply to a police officer investigating a crime where their spouse is a suspect? Or to a judge presiding over a court case involving themselves?
>The person can run two threads in their brain, one that deals with Samsung and one that deals with the internal product.
There's a reason we don't make a police chief investigate their own misconduct.
"It's a perfectly OK thing to do. The person is just the host hardware. There can be multiple threads running on that hardware at the same time. They can run two threads in their brain, one that deals with investigating the case and a seperate one that might or might not did it".
"He didn't just declare himself innocent of misconduct and embezzelment out of self-interest. The independent investigating "thread" must have arrived to an impartial decision".
"In any case, it's not blatant misconduct, you only see it as such. There's no notion of misconduct in nature, it's a made up thing we invented".
If there's accusation of misconduct, there's a bug in the system so you isolate it and investigate it from the outside.
There's no investigation happening here, just two happy parties trying to create great products that can both be successful and be even happier. Lawyers can stay out of this happiness, inventing and injecting "conflicts" that never existed in the first place.
> If there's accusation of misconduct, there's a bug in the system so you isolate it and investigate it from the outside.
Why should the investigation be from the outside?
Yep.
And in any case who (and by whom) is going to be assigned from outside to investigate any misconduct in this Google-other Android vendors case? Any misconduct of prioritizing their phones other third party hw in Android wouldn't be a crime, just a bad deal for the third party vendors.
You haven't answered my question.
Maybe confused? I'm not the person you quoted and replied to (that was dheera).
I'm the person disagreeing with dheera in this thread, and I expanded upon what you asked, adding (to further refute dheera) that there wouldn't be any investigation from the outside.
He may also "operate" with 50% of his bonus depending on Google Hardware doubling in market-share.
How much would you bet to win against me in a card-game if you have to show me all your cards and I show you none?
Rest assured, I will maintain the task to beat you in another container than all the details I need to beat you.
> So? You're talking about the person, which is just the host hardware. There can be multiple threads running on that hardware at the same time in containers.
...what?
>Conflict of interest is a human invention, and not a law of the universe. There is no "conflict" unless you see it as one.
That could be said for anything in the moral and judicial sphere. "There's no theft, property is a human invention", "There's no rape, animals don't have that concept", and so on.
That's true, but there are good reasons for calling theft and rape crimes in civilization.
Conflict of interest, on the other hand, was invented by some lawyer and HR types just to make life harder for the rest of us.
I'm an optimistic engineer, believe in win-win situations, and don't see everything as a conflict.
Conflict of interest has been a thing way before lawyers and HR types existed, they understood it and tried to prevent it at any point in history, from ancient Babylon to Rome, and from Amazon native tribes to imperial China.
It's of course also the explicitly expressed reasoning for why there are independent branches of government (legislative, executive, and judicial in the US).
>I'm an optimistic engineer, believe in win-win situations, and don't see everything as a conflict.
Yes, it's called naivety :)
Civilization is already an unnatural situation, in the sense that we have to coexist in groups (and with other groups) bigger than Dunbar's number.
> Conflict of interest is a human invention, and not a law of the universe.
Lucky for us, we're discussing this in the context of humans building stuff for other humans to buy in a human society with human governments and markets, not in some metaphysical 'but what does meaning means' context.
It feels like a large bundle to me, so they probably want to go for selling phones and computers much more heavily? Sounds interesting if you are okay with going all in with Google stuff, I'm not sure it's good news if you are using Android or Chrome otherwise.
Also, it feels like this merger will lead to a similar article to Hixie's in about 5 years:
> A symptom of this is the spreading contingent of inept middle management. Take XYZ, for example, who manages the department that somewhat arbitrarily contains (among other things) Flutter, Dart, Go, and Firebase. Her department nominally has a strategy, but I couldn't leak it if I wanted to; I literally could never figure out what any part of it meant, even after years of hearing her describe it.
https://ln.hixie.ch/?start=1700627373&count=1
I attended the Android Dev summit a few years back, shortly after Flutter had made some public noise.
During a Q&A, I was literally laughed at when I asked the head of Android whether developers should take Flutter seriously. His eventual answer equated to "well Google is really big so we can't say".
I think that was the moment I understood just how deep the mismanagement at Google actually is. Just shocking.
None of this reorganization matters as long as the executive suite remains the same. Google's problem is top-down. This rearranging the deck chairs isn't going to cut it.
Much of Google's value proposition was in their ability to innovate. The current leadership has proven catastrophically inept at this. Without fresh leadership, Google is on its way to becoming some mid-tier ad placement agency at best. Engineering excellence won't save them as long as the top leadership is incapable of adjusting to circumstances.
There's still time for Google to leverage its competence at managing large infrastructure to regain its position as a leading technology company. But the window of opportunity is shrinking. Alphabet's board of directors needs to fire the executive suite, that's the easy part. The harder part is finding a replacement CEO and executive suite who will make the deep cuts and rearrangements necessary to get Google back on track. The longer they put that off, the less chance for Google to be relevant in the future.
Who could be good for this? Demis Hassabis maybe?
It remains to be seen, but I don't really see it this way.
To me, this feels like Hiroshi went to Sundar and said that he wanted to step down or wanted to do something else or whatever. Sundar then had to choose whether to elevate one of Hiroshi's reports or elevate one of the other SVPs to lead two orgs together. Sundar chose Rick.
This feels more like a question of upper management politicking rather than mission change.
I guess it makes them more like Apple, having a vertically integrated division for making phones. TFA says it might make other phone manufacturers struggle. Although I get the impression they are already struggling with the Open Handset Alliance terms from Google that they don't like. Maybe the best outcome is that AOSP gets multiple active forks supported by manufacturers, Google apps stop being distributed by default, and the phone software ecosystem gets more decentralized in general.
Forks supported by manufacturers don't work, because they only earn money when selling hardware. So they can't each operate a huge platform maintenance team on their own.
Also, the only glue that actually holds Android in place as a single platform is Google's CTS (compatibility test suite).
Without it being mandated for Googles Mobile Services (GMS) and its revenue-share, Android will stop being a single platform.
It will start drifting apart as soon as all vendors have to implement the next display/camera/sensor/form-factor support in the OS in parallel of each other...
The other manufactures are fine I think as google still build a quality phone to house their cutting edge tech.
Samsung is the dominant Android manufacturer by far.
Would it be realistic for app developers?
Google execs found incapable of speaking two sentences without mentioning AI
If this article was written 4 years ago the title would have been,
"Google is combining …, and it's all about Blockchain"
Quantum Cyber Blockchain of Things, actually.
Blockchain is so 2023 for Google. Speaking of blockchain, what happened to the crypto accelerator Google launched in 2023?
It's google, what do you think happened? :)
"Merging teams" is such a Google thing. Merging DeepMind and Google AI, Waze and Maps, Fitbit & Nest and Pixel come to mind. I don't remeber reading such stories from other companies. Is my perception off or is "merging teams" something that Google likes to announce loudly and other companies don't or do other companies "merge teams" less often? I would like to have some input to this.
I think most other companies have not such a high amount of redundant teams to merge them all the time. So while merging happens occasionally in other companies, we usually only hear about them in context of one company buying out another one.
Not many other companies have the unique blend of being out of touch enough to think that anyone knows what this means and reality, and self-important enough to think that people care.
Are you anthropomorphising companies now? Google the company can't be "out of touch" or "self-important". The engineers maybe, but they don't make PR decisions like this.
In this case the original source is an announcement on a "inside-google" blog, and starts with "Hi Googlers", so the target are clearly mostly Google engineers. It ended up on HN because it was repeated by arstechnica, a respected and well-known portal. And here we are discussing it. So maybe people in fact do care about this?
> Are you anthropomorphising companies now?
Companies are groups of people, they're maintained by people, all their decisions are made by people and they only ever do things through their people's hands. They are as anthropomorphic as it gets.
> Google the company can't be "out of touch" or "self-important".
Such descriptions are precisely what is meant by the term "company culture"
I don't necessarily disagree, but structurally these moves can be very impactful, particularly in a huge company as sprawling and with as much overlap as Google.
If there's a lot of redundant work being done, maybe the teams will benefit from working together on the same problems.
You're right that most of us don't care, necessarily, but I think it sends some signals that the company is attempting to focus a bunch of less focused lasers at the same point.
It's less common in tech companies than in massive multinationals like GE or Sony that span a lot of different industries.
>self-important enough to think that people care.
Bingo.
I'm not sure how I feel about this, as a user of google's services and as an owner of google's hardware (pixel 5, nest hub 2). I'm probably cautiously optimistic, seeing how high quality yet unique/quirky their hardware has been. However Tensor/Samsung fabs have had their issues, but maybe factors may have been out of the hands of those in charge?
While I think Google needs a better clear vision in many cases, god help us that the people who have screwed up every hardware launch for a decade now get to run the OS too.
I'm curious to see how the new HardChromeDroid division will fare. I cannot imagine that lumping hardware and software together in one division is a particularly good idea, but I'm open to surprises.
They already did, in DSPA. The (dysfunctional) PA that brought out the Nest products, etc.
Now they're swallowing Android and Chrome.
In theory, this could lead to executing in a consistent way. E.g. not having 5+ distributions of Linux and competing platforms fighting over who will rule on some new product.
In practice, it's not going to go well.
Is Fuchsia still a thing?
Yes. https://fuchsia.googlesource.com/fuchsia/+log
My thoughts on this, is how will the Hardware convenience, with the Chrome-track-every-breath will dictate the Android/Operating System.
Android as-is does enough monitoring, and reporting back home. I fear it will make it worse for the Android users of non-Google devices, in 3-to-5 years.
> I cannot imagine that lumping hardware and software together in one division is a particularly good idea
In my experience not doing that is a bad idea. Hardware and software depend on each other. There should be good communication between them.
Exactly. This vertical integration is often pointed to as a reason Apple has been so successful.
Google is a one trick pony. I've seen the same panic behaviour with Social / Google Plus. Now they're doing nothing but AI.
I hope ChromeOS gets some love. We need some competition to Windows. They've even experimented with ads in Explorer. They have no limits anymore :)
In terms of revenue relative to their search / ads biz, how big of a deal is this? I understand there's name recognition for Chrome and Android, but the Pixel product line is nothing when compared to the iPhone or even Samsung.
Is this more of just Google being Google, as the ads side of the house continues to print more money than God?
I read it as an erosion of the implicit assumption that ads bring in the money, and projects like Android and Chrome spend it, indirectly empowering the ads division. "Why can't we just focus on eating the cake? Those guys in the bakery with their funny hats and food-grade tooling should grow up and learn to stand on their own feet without our handouts." Pixel hardware is the closest thing Google has to a division that could sustain itself from revenue streams unrelated to ads.
Probably makes a good impression on shareholders who'd prefer to see all divisions disinvested but the money counting one.
Pixel hardware fails because they don't have good distribution channels. No mobile contracts include a pixel phone. Many mobile contracts have iphones or samsungs attached.
They are failing in being present in markets; Google won't sell in more than half of EU countries at all; never mind the countries outside EU, Google sells only in a handful.
Meanwhile, Samsung products are widely available and marketed everywhere.
Consumer products are demand-driven. This is like saying my shrimp-flavored cola is failing because I don’t have distribution deals.
Pixel is great hardware, but Google has failed to generate enough demand for consumers to be going into stores and the n leaving because Pixel isn’t sold there. That’s what gets distribution in the long run.
They are not even sold in a lot of the EU and they barely market the phones. Kinda hard for there to be consumer demand if you cannot even buy it.
This isn’t true at all. Many products have very high demand in markets they’re not available.
Pixel phones have lots of geoblocked features. For example, they lose most of their 5G/VoLTE/VoWifi abilities if you use a sim from a phone network in an unsupported country, even though other models of phone don't have that issue.
A bunch of other features are geoblocked in the EU, or geoblocked to the US and Canada only.
Sure, but my point is that 1) Pixel has low market share in markets it is available, and 2) there does not appear to be any clamoring for availability from the masses in other markets.
Pixel is a great product for tech enthusiasts. It may even be a good product for the mass market. But bemoaning the lack of sufficient distribution is totally missing the point.
Because those are phones people want. Their ads are splashed everywhere. Pixel phones are not enough to get people to sign up with a carrier.
>No mobile contracts include a pixel phone.
What country is this? In the US at least you can get pixel phones through the major carriers.
>Pixel hardware fails because they don't have good distribution channels.
No, Pixel HW fails because it's shit: weak call signal reception and 5G performance due to poor Samsung modem, low performance compared to Apple and Qualcomm, high battery drain, overheating, software bugs that get in the way of you dialing calls, dimples in OLED display, poor design with ultra thick visor, questionable warranty support, AI features only exclusive to the US market but not EU making them useless, etc. Much of the issues that do not exist on Apple and Samsung phones.
Carriers go with Apple and Samsung because those brands are a slam dunk with consumers, are well known and mostly reliable by this point with fewer lemons, and when they do fail, there's service centers for them everywhere. Pixels are mostly bought by enthusiasts who will forgive the rough edges for the chance of being Google's software beta testers.
Not to mention that Pixel doesn't provide anything that you couldn't already get on a Samsung.
Pixels can still have better photo/video quality and less shutter lag than Samsungs and a cleaner, more unified SW ecosystem with only the Google Apps, versus Samsung who duplicates those apps with their own: Samsung Pay, Samsung Browser, Samsung Notes, etc. Some people prefer Samsung's own apps to Google's(their browser is really nice), others find it bloatware.
The main problem is that due to their pricing, bugs and issues, Google's recent Pixels are no longer a slam dunk reccomendation like they used to be in the old days when they rocked Qualcomm chips and were compact, well built, dirt cheap phones.
Google needs to properly absorb Android into Chrome before any antitrust measure forces it (along with the Play Store) to be spun out, as would be most healthy for absolutely everyone except the Google ads business.
There has been so much potential squandered with Android over the last decade it is amazing.
I think that this unification simply took place because Google Android in the future will only be the top layer and Fuchsia OS will manage phones, etc., so such a merger is needed to prepare for new changes.
I think that this unification simply took place because Google Android in the future will only be the top layer and Fuchsia OS will manage phones, etc., so such a merger is needed to prepare for new changes.
Pretty much every LLM is a google competitor now. I guess Meta's unleashing of AI is a huge thread to google's moneymaker.
Maybe they see the writing on the wall and want to become Apple 2. Bad news for Samsung?
I don't see how Google could threaten Samsung while at the same time not antagonizing every other OEM out there.
And even if they went for "all out war" I still don't think it would be that one-sided.
What for? The market share of Google’s hardware is minuscule, and other Android OEMs are unable to take advantage of Google stuff since they use their own chips.
I don't know anything involved in the process, but I hope this doesn't make it any harder for regulators to break them up when the day finally comes.
I hope ideas like "regulators break them up" never come true. I'll never understand why people crave the destruction of productive organizations. Android has been a stunning success. Valuable for consumers who enjoy the platform, profitable for the investors who bet on the platform and lucrative for the employees who work on the platform. The only people that seem to have a problem with Android are misguided ideologues who think that "big company == bad".
> destruction of productive organizations.
I don't think your understanding of this is complete. A break up doesn't destroy anything; on the contrary, it creates value. Where once there was a single monopolist, there are now two (or three or twelve) organizations that can proceed independently of each other, and be much more focused on their core product.
For example, maybe Google can keep the play store, but not Android. Android would become an independent entity, and can develop in ways that benefit all Android stakeholders instead of just Google. Maybe then Android will finally be able to focus on competing with iOS in ways that Google would never dare to (since Google's relationship with Apple is a bit sus)
A breakup can also be good for investors, because illegal monopolies are inefficient. Android could potentially be much more profitable on its own than under Google's umbrella, and the play store could be more profitable if they're actually forced to compete. This can lead to innovations which increase revenue, and which never would've happened under Google.
Does this mean we’re going to see a new secure OS?
The world needs a third party software and hardware stack that isn’t controlled by big tech walled garden monopolists / authoritarians. Not just for phones but laptops and computers too. As far as phones go, unfortunately the best alternative I’ve heard of is Graphene and the best phone for Graphene is the Pixel series. And I assume using it as a daily driver is problematic without access to various apps or maybe if websites block them or even carriers - not sure.
This is why my next tech purchase will be a Raspberry Pi 5 and a Wacom One 13" touch screen --- my testing with a Raspberry Pi 4 and Wacom One (gen 1, no touch) went well, so I'm hopeful that this will work as well.
^ totally oblivious that he is trapped in the Broadcom Pi walled garden.
It would be nice to use some totally open firmware, but that's a big swing in terms of finance relative to the inexpensiveness of the rPi.
That said, I believe I can live with:
https://raspberrypi.stackexchange.com/questions/561/is-the-b...
https://forums.raspberrypi.com/viewtopic.php?t=340024
If someone would make a decent tablet with opensource underpinnings and Wacom EMR, I'd be glad to consider it.
Nevermind open firmware, just ask for an ecosystem that isn't completely controlled by one company named Broadcom.
Broadcom Pi.
It's called Broadcom Pi.
The fruit names are just a naked attempt at BigTech-laundering.
At least it's a company which will allow alternative OSs.
I'm still quite annoyed that I set aside and stopped using my Samsung Galaxy Book 12 because I couldn't figure out how to get it to boot something other than Windows.
The world is reactive, never proactive
Does this mean Android/Google will scale down with all the cringe YouTube tutorials.
I wonder if Google is anticipating that they will be broken up.
It is unsurprising to see that RickO beat Hiroshi for the title of grand poobah of devices and platforms. Hiroshi always made the impression of a very smart guy. Rick always made the impression of a good politician. Politicians always win
The 'consumer data extractor' (hardware) team is teaming up with the 'consumer manipulator' (AI) team, and all free of any inconvenient 'dont be evil' policy.
What a time to be alive!
[flagged]
I'll wait for Advanced Platforms and Devices second edition.
I recently got a new Pixel phone and Google's much hyped new AI features just seem so... gimmicky. One of the setup examples shows how you can circle a tent on the left side of a picture and move it more to the center. Neat, I guess? It's kind of a fun toy but I'm not sure what problem this is actually solving. I'm sure there are some usecases for this out there, but it's not a capability I have ever found myself wishing for.
Meanwhile the rest of the phone is surprisingly buggy and annoying. Basic functionality I use every day is worse than on any other recent phone.
Google has never been a great product org, but this desperate need to be seen as one of the cool kids in AI is making things worse. Granted I think of phones/computers more as a tool than a toy and put much higher value on usability and reliability versus novelty; perhaps I'm outlier in that.
>I recently got a new Pixel phone and Google's much hyped new AI features just seem so... gimmicky
Not just that, but their biggest crime is that almost none of those fancy AI features Google paraded at the Pixel launch even actually run on-device but need to be sent to their cloud for processing, despite all the gloating about their new Tensor 3 chip's AI capabilities being the most important (since that chip sucks at CPU and GPU benchmarks compared to Apple and Qualcomm). Also, their Tensor 3 can't even run Google's smallest LLM. Absolutely embarrassing.
They REALLY need to unify the HW and SW development efforts to create a coherent and functional product, instead of designing them separately bazaar style then jerry rigging them together like some underfunded start-up making products for Kickstarter.
>Absolutely embarrassing.
So sad but this continues to be the case for Google's incursion into AI. Why do they still keep Pichar around?
As a customer running on device is fairly low on my priorities, and I assume that's the case for at least 90% of users.
Would it be nice? Sure, but I much prefer useful features now that could run on device later on if it adds value.
I can't stand all the "AI" junk, especially when things worked better in the past. My pet peeve: I used to be able to ask google maps while I was driving "What's the E.T.A.?" and it would respond with, you know, the answer. It's been broken for many years now and responds with nonsense.
Another one: I can't tell my phone to change it's name to what I want. Basic "AI" fail.
honestly, it seems like every phone has its broken quirks. I recently switched from iphone to android and there's still a random collection of everyday things I do that are simply... broken.
Maybe these devices have become so complicated they're simply too challenging to work out all of the edge cases out of. New features are easier.
>Maybe these devices have become so complicated they're simply too challenging to work out all of the edge cases out of. New features are easier.
With the amount of telemetry and data Google is collecting I doubt they can't catch edge cases, let alone recurrent bugs that impact multiple users.
I wanted to buy a Pixel on sale last week but I watched a 6 month long term review of the Pixel and the reviewer complained that every new update fixed some bugs but added it's own new bugs.
It's why I'm still gonna keep using a phone that stopped getting updates over a year ago: it's finally stable and no more new bugs are being introduced by updates, as my mind and muscle memory has already adapted to the old bugs.
Maybe I'm getting old but while 10 years ago I couldn't wait for new major updates to arrive on my phone, I feel like phone SW has peaked a few years ago and has been on a constant decline ever since, with new updates just adding useless crap that bugs you and changing things for the sake of change without improving them, and I would much rather have a phone that only updates security but nothing else. Basically I don't want my phone to be a Googler's playground and me being the beta-tester.
Not to mention all the Google shit on Android is just constantly harassing you. LOOK HERE, LOOK THERE, SET UP THIS, SET UP THAT, TURN THIS ON, TURN THAT ON, GIVE US FEEDBACK, SYNC NOW, SIGN IN TO GOOGLE, LET US SCAN YOUR PHONE FOR YOUR SAFETY, SIGN IN FOR SECURITY, SYNC YOUR PHOTOS, SYNC YOUR DRIVE, USE AI FOR THIS, HERES HOW YOU DO THIS THING BECAUSE ITS NOT OBVIOUS AND WE SUCK AT UI, GET UPDATES, WE DISABLED PERMISSIONS ON OLD APPS, WE DID THIS FOR YOU, WE DID THAT FOR YOU, POST PICTURES OF YOUR RECENT HOME DEPOT TRIP.
Jesus christ, I've had to dismiss at least 20 different popup things just in the Messages app since I reset my phone a few days ago. Just fuck off already!
And guess what. After resetting the phone, I still can only make a successful outgoing phone call 1 out of every 3 tries, and it will only work after a reboot. It worked fine after the reset for about a day. Now, again, it barely works as a phone.
Rodney Dangerfield was right. There is no fucking respect for the people using the phones. There is only respect for the stocks going up. Fuck you and give us money, that's what smartphones are all about.
Yesterday I ordered a Nokia flip phone. I'm done with iOS and Android. It has added nothing to my life except distractions and maintenance. I spent 3 days trying to get this piece of trash to work as a phone. Just a total waste of my life.
>, I still can only make a successful outgoing phone call 1 out of every 3 tries
Which phone? Pixel?
Harassment as a service. That comes from the advertising mindset..
> Under Rick Osterloh, a new platforms and devices team will be dedicated to bringing AI to your phone, your TV, and everything else that runs Android.
Who is asking for this? Why can't they just make their search engine work again?
> Who is asking for this?
Anyone who regularly uses Siri or "Hey, Google".
Nah, I want them to first fix the basic shit I actually (would) use.
For example, when I'm driving and a timed phone alarm goes off for the Android phone in my pocket, I ask it to silence the alarm, yet instead rebukes me by falsely claiming no alarms are active right now.
It's fixed now that I checked, but for a while it would also secretly ignore the date that I already specified for a scheduled event while it was prompting me to clarify the time of day.
You mean the thing that requires an internet connection or it doesn't work?
And probably will continue needing internet for the foreseeable future, regardless of how many mobile Tensor-chips they develop, because cloud data and compute power will always be orders of magnitude better than your phone?
That's the thing they need specialized mobile-hardware teams involved for?
> You mean the thing that requires an internet connection or it doesn't work?
The AI? Or the phone?
So nobody.
(It's only good for pranks.) Hey Siri, call the police. Ok Google, call the police. Alexa, call the police.
I can't help feeling that Google is trying hard to turn itself into the Sirius Cybernetics Corporation.
Share and enjoy!
Or maybe, is it even a possible scenario that different teams are working on different products?
It seems like Google didn't learn the lesson from Google+ and is going to do another fear-driven reorg.
Google cannot learn that kind of lesson, the people who were in Google+ are probably long gone from Google for greener pastures most likely. Second, google is an amorphous giant with no other goals than increase market share, no wonder their products are terrible. But still, with all this in mind, we should remember that great things go come from google, it's just that they're not capable of capitalizing on them.
insightful comment. I'd like to think that Google will integrate AI in an at least mostly useful way, but it's clear that large scale reorgs like this around chasing a competitor can and do end badly.
"Google is paninicking and chasing the latest buzzword" would be a more descriptive title
Also seems like some employees/exec has figured out the latest promotion mill.
Yawn. Get back to me when something of actual note gets launched.
Mobile OSes are now a boring, stable environment. All this noise about AI seems like an attempt to convince investors that some paradigm-shifting change is on the way. It isn't. A mildly better Google Assistant is on the way.
[dead]
It's going to be a tremendous success, I can feel it already.
I believe I speak for everyone when I say: As long as the AI does something helpful I don't care how you structure the team.
But the current generative craze with "AI generated backgrounds" is a dead end.
Give me better AI autocomplete, AI image correction, AI noise cancellation...
>Give me better AI autocomplete, AI image correction, AI noise cancellation..
We have all those, some of it we've had a long time, we just didn't call it AI.
aren't they doing those things?
[flagged]
Buy a Shenzhen Android phone, then. For $300, you get a stock Android phone that has all of those things.
The comments here are trending towards "stop cramming 'AI' into everything". I am curious how the end-user consumer (versus 'AI' for enterprise/business) differ in experience and use. We are in the beginning of this AI-fication, and it seems deep learning models are doing really well and that DL can predictably scale [1.]; therefore, do we have to wait a bit for really life-changing AI for the end-user consumer?
I can see AI in enterprise/business being extremely useful in different industries, but at the same time, is the current 'AI' actually good/useful for the end-user consumer?
[1.] https://arxiv.org/abs/1712.00409
How is a reorg at Google any kind of news ?
Last I spoke to folks working there, these seem to happen every y-ending day.
A reorg at this scale?
It doesn't even happen annually.
A reorg of two teams of 10 people? Sure. Google is a ~180k person company.
There's also a buried lede:
Hiroshi, who's been an Android lead since before it shipped, is no longer leading Android.
When Jony Ive and Scott Forstall and the other big Apple execs left, that was news. Hiroshi may not have Jony's profile, but it's still a major change in how Android is governed.
https://en.wikipedia.org/wiki/Hiroshi_Lockheimer
Even small changes by a $2 trillion company with billions of daily users are a big deal.
I basically forgot that Google made Android.
Are they really making major contributions? Seems like Mobile OS are basically stagnant.
It gets less and less stable with every major upgrade, if that counts. Achieving that takes effort.
Do you have some basis for this that you can share? My perception is the opposite.
Treat yourself to multiple UI lockups, missing home buttons, stuck notifications in the middle of the screen, etc every single day if you disable animations.
On a completely stock Pixel phone.
Nah, that's just entropy.
Making UIs progressively worse beyond levels you though imaginable, like Reddit or GNOME -- that's art.
"Google's previous head of software platforms like Android and ChromeOS, will be headed to "some new projects" at Google"
Ok so they're more than willing and able to shut down divisions and relocate human resources to "new projects".
Why not just do that with Nimbus? At least ChromeOS didn't put Google in the position of being complicit in a plausible ........
Strange decision making, shutdown and kill things people love (e.g Google Domains, RSS, etc.) but not shutdown a project that puts the company at risk and that some of their own workers actively protest against.
It's not nearly as momentous as removing Andy Rubin. Android has evolved a lot. From an OS technology PoV it's less distinctive and more about competitive parity with iOS. Kind of boring.
There are a handful of exciting movements, imo, but they're few and far between.
The Android Virtualization Framework, for one, is something I'm really excited to see in the next few years
>puts the company at risk
You think Numbus puts Google at risk, as opposed to strengthening its corporate standing with the US government?
Is that the dichotomy here? Strange worldview.
I mean at risk of being complicit in genocide.
Because Nimbus makes lots of money.
Tesch and Stabenow probably made a lot of money with their government contracts also back in the day. Doesn't negate the fact that the way they made that money was wrong.