This is a big shift from the Google of old. People in this industry talk, even when they work for the companies that make these products. Previously, Google was very cautious about doing anything that would create a rift between itself and all the vendors that made Android what it is today. Very little was held back because Google needed to keep Samsung happy, and Samsung wouldn’t be happy if a cool new Android thing didn’t work on the next Galaxy phone.
Now Google is building all these cool things but calling them Pixel features. Features that will probably never come to a Galaxy phone or any other brand of phone. And it’s building the hardware to make them even better and to unleash even cooler things in the future. Things that are Pixel features. Things that will never be on a Galaxy phone.
You can’t even really call Android an open source mobile operating system anymore at this point, and it seems the latest few Pixels are really starting to drive the point home that for Google, Android is not really their mobile brand anymore – it’s Pixel. We’ll see how far they’re willing to take this, but I wouldn’t be surprised if they’ve barely even started.
What’s the life expectancy of AOSP?
Google doesn’t care now just like Microsoft doesn’t care with Windows 11 and things that work on Surface.
Like what Samsung gonna do? Make its own OS?? lol.
Go will Open Source Android till Google makes changes in the main Android that kills making apps work on both.
I don’t know why people trusted Google. They are a company like any other company.
“I don’t know why people trusted Google. They are a company like any other company.”
What else were we supposed to do? For all its faults, Android was the most open at the time. Windows mobile and iOS were and are thoroughly closed source. Maemo, moblin, tizen, etc. never went anywhere.
Android closing up is the fault of the IHV Android ecosystem itself. Manufacturers trying to put themselves on top of the Android ecosystem with forks. Abysmal update track records, because phone manufacturers care more about shipping the latest and greatest at a premium, than taking care of their customers whose money they already got. The only way Google could partially get out of that mess, was to make most of the juicy bits of Android part of their own play services. Which phone makers need to ship if they want to make use of the Android trademarks and associated marketing and brand recognition. So an ugly work around, but an effective one.
Companies like Samsung could have made their own OS (which they had) just like Apple as they have almost Apple level money.
But no they took the cheap way out and now are stuck (and so are customers)
As far as putting themselves on top, this is a business and so they want to make money. That is the way you make money, you have to differentiate yourself. You see what happened to HTC and LG (Among others) they didn’t do that and now are out.
It’s just funny to me because Google changes like the wind, they are an ad company that makes a bunch of stuff that they can cancel at any time if it does bring in ad revenue. Not dependable at all.
In the end Samsung will be fine, not sure about other Android phone and tablet makers though.
I suspect that it will be Apple, Google and Samsung. Everyone else will be struggling going forward.
No they didn’t. Samsung smartphones featured no less that 3 oses at certain moment: Android, Tizen and WindowsPhone. Tizen got the same top shelf devices as Android did and Samsung even poured serious cash at developers that were any interested, but there’s only as far as you can go on a dead horse.
This might be a bit long, but there were many outside reasons that pushed Android to be closer over time, and I will try to list some (those are in addition to of course “the usual suspects”, I will not go there).
(This is my personal observation and guesses only).
First, there was a proliferation of forks from various competitors. Amazon with their Fire phones came to my mind. But there were also many Chinese competitors, and yes of course Samsung too (will come to that soon).
In that race, Google was in a losing position. Any new feature, regardless of how much they have spent developing that, would be available on competitors day one. However competitors like Samsung were free to pursue their own proprietary features, and not give them back to the open source ecosystem. So, that meant any purely Android Google phone would always be behind the competitors in terms of features (by definition).
Second, there was a “thinly veiled” attempt to move away from Android within the headset manufacturers. Samsung again, had completely skinning the Android UI as “Galaxy”, and also released their own Tizen based Z series: https://docs.tizen.org/platform/what-is-tizen/profiles/mobile/. The likely guess is their plan would be “swapping out” Android for Tizen in the “Galaxy” series without customers noticing any difference (this would only work if all Android APIs were available open source, and Samsung had a viable store to replace the existing one).
Anyway, these two forces, and many others, in my opinion, pushed Google to have “exclusive” features on Android (how they were ultimately used is another discussion).
And here we are today.
(Also as far as I know Pixel and Android teams have a “firewall” in between, and communicate like as any other outside OEM).
sukru,
I understand what you are saying. Although I would say that key portions of android have already been made proprietary for a very long time now. If you run AOSP, it’s incomplete and not merely because apps aren’t installed, but because OS functionality is missing. Even with lineageos a lot of apps won’t work without “google play services”. As for alternatives including microg, google’s signature verification needs to be cracked, which is not endorsed by lineageos but is ultimately necessary to use alternatives to google’s proprietary software.
https://android.stackexchange.com/questions/228879/lineageos-is-it-possible-to-install-and-run-apps-without-using-gapps-or-microg
This whole situation is unfortunate for FOSS, but google designed it this way on purpose.
The decline of FOSS in mobile could lead to more pain points for us in the future. but yeah I get your point about why google wants more of android and pixel to be proprietary.
Alfman,
I think this started about 10-ish years ago, but can’t exactly tell. At that point Android was a providing a “democratizing” software, and companies like HTC would be competing on the hardware. It has gone downhill ever since. (HTC / T-Mobile G1 could be the “only” pure AOSP model).
But I am not sure what is the best “solution”. Many FOSS companies run into this when their users become “competitors”, or they just plain need more money.
Like ElasticSearch which was very popular, and decided to ask money off AWS users. Amazon countered by offering OpenSearch, which is actually good since we not have competition.
IBM on the other hand locked down RHEL when Oracle was copying it almost verbatim. (And of course all those “lost revenue on CentOS”). Which was not a good move for the competition, nor the open source community.
Walking this path is not always easy. Some can avoid this by making backend services into a subscription (Ubuntu with Ubuntu One), some just pull out from open source entirely (Oracle/SUN ZFS).
I am sure Google might have done better, but I am not sure how much so.
edit: not have competition -> now have competition (@Thom: please bring back that button)
sukru,
Yeah, I don’t know the answer either despite having contemplated these issues for many years. There may not be a good answer. Theoretically FOSS is more compatible with communism than capitalism, but that’s its own baggage, haha. It’s messy, but maybe FOSS software could be funded as public works, much like public roads are.
microg, in my non-technical perspective, is a no-go.
I prefer much more using Google’s APK, with their version pinned and running as normal (not system) apps, inside a wrapper, as GrapheneOS does.
Weasel6418,
I understand. But google’s play store implementation is also a clear no-go when it comes to FOSS. When forced chose between two “no-go” solutions, I’ll personally opt for the FOSS to the extent that I can. While some people may be comfortable running google’s proprietary software on GrapheneOS, we should note that GrapheneOS doesn’t fundamentally solve the FOSS issue. The debate of proprietary vs open source still applies. Given how tons of software ends up needing the platform libraries, I’d far rather those libraries be FOSS.
I do realize that microg’s implementation isn’t functionally 1:1 with google’s play services, Microg isn’t technically at fault, they’re dealing with vendor locking problems that google created to impede alternatives. I realize not everyone is willing or able to stand up to this, but supporting vendor locked implementations goes against my principals and so I choose migroG to protest that.
I wish pure FOSS mobile platforms were more viable. The network effects end up roping all of us into the duopoly in one form or another. Grapheneos and lineageos that run android software are the closest most of us can get to FOSS alternatives without entirely giving up on mainstream smartphone apps altogether.
Sadly for any alternative to have a chance the competition needs to unite behind it so it can gain enough market traction to get apps (chicken/egg).
There are plenty of perfectly good mobile OS out there that could fit the bill which failed to break the market alone.
WebOS, KaiOS, Blackberry OS, Sailfish, Windows Phone(?) etc
If one of them got fully open sourced and a consortium backing it would have a chance. But sadly each manufacturer wants to control the “new Android” and so they inevitably fail as they can’t get market share alone. So we are stuck with Android, an OS that lost its soul long ago.
I was hoping LG did something with WebOS (at least try it) on Mobile phones but no dice.
WebOS started on mobile phones! With a very very cringy ad campaign that likely sealed its fate: https://www.youtube.com/watch?v=tSMj5RoYdEI
I think the success or failure of each of those had only to do with how well they were marketed, how much buzz they could generate (and a bit more about how much developer support they had – but devs follow buzz too). iPhone and Droid (Verizon) had a lot of buzz early on, Samsung continued the buzz with its full court press of Galaxy products, while Motorola was destroyed by essentially zigging when it should have zagged, and trying to carve out the middle market without also capturing the buzz at the top end (their heart was in the right place, but that doesn’t work – you have to tell the aspirational story too). Motorola also tried to go direct to consumer and misunderstood how important the buzz was at the point of sale in shops like Best Buy (where devices are sold by spec sheets – that’s all the sales people there have to go on, so that’s what moves devices).
Blackberry became old fashioned – a deathnell in this industry, webos was a cringy middle market “me too” or worse “catch up” product, Windows Phone was, well, Windows, at the height of Vista, no one wanted that crap on their phone, and everyone was sick of Microsoft at that point. Hard to generate buzz after 30 years of MS being MS.
How do you define what is part of an OS? AOSP already has a minimal camera (and gallery, and music app, and SMS app), so why should the Pixel Camera be considered part of the OS and not a pre-installed value add-on the OEM branch of Google puts in the device?