zlacker

>>akyuu+(OP)
Why was it that in the early PC days, IBM was unable to keep a lid on 'IBM compatible', allowing for the PC interoperability explosion, yet today, almost every phone has closed drivers, closed and locked bootloaders, and almost complete corporate control over our devices? Why are there not yet a plethora of phones on the market that allow anyone to install their OS of choice?

>>Subicu+hn
You're getting a lot of indirect responses. If you've ever tried to mod your android phone the answer is simple. Its google play services and hardware attestation for things like banking websites.

Its really easy to make a custom rom but hard to do serious "real life" stuff; companies don't want to make it easy. To most regular users, if they cant download apps from the google play store, and they can't use venmo\cashapp, then the OS is dead in the water from day 1

>>cons0l+Bw
This just shows that the barrier of entry of a new phone OS is more than $0. You can pay app developers to port their apps off of play services, you can pay developers to add support for your attestation keys. Considering how many billions of dollars Android makes for Google, there is a room for a return on investment for an alternate OS to enable investments into a new OS.

>>charci+zP
How much do you want to pay? Who will be paying? Big companies will probably laugh such an effort out of the room, nay, they will not even let you into the room to talk with them.

>>zelphi+FU
$10 million dollars per app. The creator of the new OS will pay. If you offer enough cash they will stop laughing.

>>charci+DX
Have you ever tried to pay a bank to do something for you ?

Trying to get some scale, you're hypothesizing about giving 10 millions to HSBC to make business with your startup, when they're throwing away 500+ millions every year just to cover their money laundering.

https://www.investopedia.com/stock-analysis/2013/investing-n...

And we're discussing doing this for basically every major banks.

>>makeit+x11
But what what I'm asking for is only a small amount of engineering time to add 1 line to their gradle and change 1 line in their app's code. This isn't a deal spanning many engineering years doing on going work and having to measure how effective things are. It's a small change plus the overhead of making a deal and getting through the beurocracy.

>>charci+931
The issue is to have them do anything at all.

I see it akin to the proverbial "not getting out of bed for less than XXXXX". You're getting out of bed every day, for free. But having someone make you do it for a specific reason will be an exponentially harder proposition.

> 1 line in their app

Aren't you asking them to maintain compatibility outside of Play Services and be on available on your platform ? That's a whole project, including their (or their contracting shop's) validating the whole new stack from a security and technical perspective, and a legal and business check on what that actually means to them.

Perhaps we can look at it from a darker perspective: if a random guy came to the bank to ask them support for their parralel phone ecosystem, the bank would at least want to know what they're getting into and what's in it for them. Especially if they're offered 10 millions for allegedly one line of code.

>>makeit+k71
>not getting out of bed for less than XXXXX

I just made up the figure. Perhaps 10 billion dollars is more enticing. Perhaps you have to purchase the company outright and then dictate they add support. My point is that it's not impossible to get the apps people need to work on an alternate Android OS. It is a matter of funding conpatibility. You can find a niche audience of people to start out with to make a competitive OS for them. And then overtime expand that audience more and more.

>Aren't you asking them to maintain compatibility

Typically the complaints about banks is that they use the Play Integrity library which doesn't trust other operating systems. So the ask is to support the Android API for integrity and to trust the key of the OS provider. This would be done via a new library to make integration easier and more foolproof.