> Unfortunately due to their market dominance and muddling of (hardware, OS, software distribution, and platform ecosystem), it's almost impossible to disentangle.
I think it is more complicated, and nuanced, than this.
I have a theory that the world can only support a maximum of 2 consumer computing platforms at a time, due to the cost of writing and porting software. Therefore causing a natural tendency towards either a monopoly (seen in the 90s and early 2000s) or a duopoly (the current smartphone era).
The fact is writing software is expensive. Developing cross platform frameworks is also incredibly difficult and in the case of Mobile, has taken a massive 3rd party entity (Facebook) the better part of a decade (React Native) to get even close to "working well" (with other solutions also being decade long projects). One can argue that during the first decade of Android vs iPhone, (2007 to 2017) that the cross platform solutions were all pretty terrible, thus the massive shakeup that React Native (for all its warts) caused.
Then there is the fact that developing a consumer OS is hard and expensive. Very few companies have the resources needed to make a consumer OS. Not just writing software, but localization, documentation, SDKs, UX work, security, update infrastructure, and so on and so forth.
Honestly I'd say today's current duopoly may even be a bad thing for small software companies, basically doubling development costs. Compare this to the 90s when releasing consumer software just meant compiling for a Microsoft OS and never worrying again because Microsoft handled forwards compatibility for you!
Contrast that to now days where you see applications to control smart appliances apps being discontinued left and right because companies cannot justify keeping 2 dev teams staffed so they can patch an app once every couple years when app store guidelines force changes.
If we adopted standards instead of making walled gardens, things would work. If the onus was on Apple and Google to make their platforms standards compliant, and that the egg would be on their face if they didn't, they would be the ones doing the rigorous testing, bug fixing, and optimization.
If Microsoft can make their platform work for 20+ years of software, Apple and Google can be on the hook for HTML, WASM, and a standardized UI and hardware abstraction layer.
Yes, but then all apps would be poorly-written webapps in a single-threaded language with a bunch of performance kludges bolted on (like WASM, which STILL inexplicably can't directly edit the DOM).
There's frankly not much of a technical reason why Android phones and iPhones can't run each others' apps except for the malignant IP enforcement of both Google and Apple.
This is a business problem - take away Apple's 30% cut and see how quickly they change their tune on "security."
The video game industry has already solved this problem with engines like Unreal that can compile to PC, Xbox, Playstation and Nintendo. There's certainly no reason why you shouldn't be able to just compile the same phone app from the same IDE to both iOS and Android. I believe the architectures are basically the same but with different operating systems.
There's also no good reason why you should have to distribute your software on iOS through a monopoly app store or why Android should hide the ability to install software from non-Google app stores or the internet behind a scary "security" warning. That "security" warning, while better than what Apple does, is itself a monopolistic practice that should be illegal.
Honestly, even outside of gaming, we could probably go back to web apps again, now that hardware is better. Zuck's famous quote about the company's biggest mistake being going with HTML5 instead of native for mobile is 12 years old.
It isn't just the barrier to entry, Microsoft famously paid the cost to enter the smartphone market.
It is how many platform ecosystems a market can support problem. Creating a platform ecosystem creates higher costs and lowers profits for everyone building in the market (the market in this case being "all smartphone users").
1 more platform is going to be at least a 30% increase in dev costs (if not a full on 50% increase, and for large complex apps it can be even worse), and now if that platform gains market share, that is less sales on the other 2 platforms for anyone who doesn't support all three.
So to make the same money as before, a company now needs to spend 30%-50% more on development expenses.
There is a natural limit to the number of platforms that can exist when the cost of building across multiple platforms is non-trivial.
This by the way was why even the tiny European market was able to support so many different 8-bit platforms back in the day. Porting costs were minimal, one or two people for a couple of months, and developer costs were much lower back then. Jump to the 90s and now suddenly you have just 2 consoles, and with Xbox losing market share year over year, we may soon be down to just one in another decade or so.
If AGI ever becomes a thing and you can simply spend electricity to write the new operating system, that would be pretty insane, and it seems like it would disrupt duopoly.
I think it is more complicated, and nuanced, than this.
I have a theory that the world can only support a maximum of 2 consumer computing platforms at a time, due to the cost of writing and porting software. Therefore causing a natural tendency towards either a monopoly (seen in the 90s and early 2000s) or a duopoly (the current smartphone era).
The fact is writing software is expensive. Developing cross platform frameworks is also incredibly difficult and in the case of Mobile, has taken a massive 3rd party entity (Facebook) the better part of a decade (React Native) to get even close to "working well" (with other solutions also being decade long projects). One can argue that during the first decade of Android vs iPhone, (2007 to 2017) that the cross platform solutions were all pretty terrible, thus the massive shakeup that React Native (for all its warts) caused.
Then there is the fact that developing a consumer OS is hard and expensive. Very few companies have the resources needed to make a consumer OS. Not just writing software, but localization, documentation, SDKs, UX work, security, update infrastructure, and so on and so forth.
Honestly I'd say today's current duopoly may even be a bad thing for small software companies, basically doubling development costs. Compare this to the 90s when releasing consumer software just meant compiling for a Microsoft OS and never worrying again because Microsoft handled forwards compatibility for you!
Contrast that to now days where you see applications to control smart appliances apps being discontinued left and right because companies cannot justify keeping 2 dev teams staffed so they can patch an app once every couple years when app store guidelines force changes.