But then you can also go to the store and ask "how well does this Linux computer you want to sell me, run these Metal or DirectX only games". Fragmentation reduces the overall amount of games and probably hurts small players like Linux the most, but for Apple it kind of makes sense.

They have their own graphics framework which is optimized for their OS and their hardware and very probably works quite well and they don't bother about anyone else because...

It's a bet. They'll lose some software which might not support Metal but the others will lose too. In the end it might be a win for Apple with developers who literally are only capable of programming for Apple.

Which would make them less appealing because much like learning a dead programming language (and not a popular one like COBOL), there's no real use to learning an API that nobody uses. Like how anyone who specialized in Brushed Metal or Carbon is probably not going to mention that on their resume.

I can attest to this. Unix-based tools + not half-baked interface is a really sweet combo for development. Windows is pretty annoying to develop in, and efforts like powershell, while appreciated, tend to only go halfway. I used to use linux but usually the "best" interface is only the best one for a few years, and then you have to find something else to migrate to. It gets disruptive after a while.

And yeah, the frickin vendor lock-in where you have to build on a mac to support mac or ios.

This is why i started shifting to expading my backend knowledge. I'm not really big on pretty buttons and things, anyway, and if i really need to code for someone other than myself (since i code on my own), they can find the UI guy. That's not to say I don't have some pretty cool tools and ideas that i've built myself that would be fun to make a UI with, but I can't be bothered all the time to constantly rewrite my same frontend every time something gets deprecated and replaced: i never get anywhere with the backend.

Another fun little project of mine was to write tutorials for beginners, to help more people get into programming. I found ways to help people who can't pass https://blog.codinghorror.com/separating-programming-sheep-from-non-programming-goats/]this test[/url[ and help them pass it. The problem is, i need a stable platform. And nothing other than clib is stable. The problem is, the trick to doing this is going lower than C, and invocation of clib isn't even stable, let alone the platform on which you would do the invoking. I thought about going the VM route, making my own virtual machine to be stable, but then that too must be maintained, and it isn't as simple as hitting the compile button on the next platform, because even C++ code is changing, and i thought i read that some language constructs are now deprecated as well.

So, at this point, i figure that it makes most sense, to me, to say "the hell with everyone else, i'll do my own thing unless someone pays me to do it." What kills me is that there's so many things that could and should be universal at this point, but are not, like 2 button mice. I connect one to my android tablet, and right click sends me to the desktop. Why!? Why can't it act as a hold left click, to provide the same functionality as it does in desktops, which is presumably why i connected a damn mouse?

IMO Apple loses the most, since they've deprecated OpenGL we have Windows and Linux supporting it, not to mention Vulkan. Specially when some proprietary software still uses OpenGL or OpenCL like Adobe.

Also, AMD users on Linux have a much better performance with OpenGL on open source drivers than the Windows counterpart, in terms of Vulkan it seem to still needs some improvement but it is still good enough for games. Then there's Gallium3D Nine for DirectX 9 games which offers near native performance