> I get more and better work done with AI, to my own satisfaction and to the benefit of my employer
In the short run, because Anthropic and other providers are heavily subsidizing coding agents to maximize user base. Will your employer still benefit and be satisfied in a couple of years when Anthropic jacks up the price by 5x and dumbs down Opus to the point where 50% of changes are easier to do manually than via an agent?
> Lower income countries where something that was very inexpensive and worked enough had legs
As a lower income country citizen who had a Lumia 635, that's exactly how they got me. It was something like 150-160 U$S and at the time, in my country that would've bought me a Galaxy J2 or some other bottom of the barrel phone with a laggy Android UI. In contrast the Lumia 635 was just as compute/memory-starved but Windows Phone could handle it way, way better. My dad even used it into 2019 after I upgraded because it was just there and worked better than any cheap Android phone.
The developer experience of the platform was weird. As someone with (a bit of) C# experience I found XAML+C# way easier to get into than Android's kludgy layout system, but as expected the ecosystem of libraries was just not there, and while it was easy to get a free developer account as a student it was a bit pretentious of Microsoft to expect normal people to pay 25(?) USD for the privilege of publishing in a marketplace of between 0 and 5 users.
>The AppleTV is already in a similar tier to modern consoles, as far as specs and benchmarks go
What benchmarks are you talking about? CPU-wise the A15 Bionic just barely beats the Ryzen 3700X in single-core and gets absolutely destroyed in multi-core (Geekbench). As for the GPU, the Radeon RX 7600 (closest thing I can find to a "modern console") does >10x the TFLOPS in FP32.
The only reason why they look like they're "in a similar tier" in ported games is because the A15 Bionic is usually tested on 5-6" screens that can be upscaled from 360p without any measurable loss in visual quality, with a massive downgrade in model and texture quality for the same reason. The only modern console the Apple TV "may be" similar to is the Switch 1
>Right and that means people have to send their data to an external service.
Nothing in this proposal claims it has to be a local AI. That just happens to be the implementation by Chrome and Edge (for now at least, I'd imagine Google will eventually start moving this API towards hosted Gemini).
> You need something beefy if you're serious/professional about those
But you can get way better results with the lowest end computers than you could years ago.
Back in the 90s my grandfather used 3DS Max to map out his future apartment's rooms and start planning furniture, using renders to get an idea of how sunshine would look like at different times, etc. At the time, he did this on an expensive 486 that would take an entire day to render some of those visuals. Nowadays I can do the same with a free copy of Blender and any reasonably modern integrated GPU in probably under an hour.
Who is the target audience for this? I can't imagine that many modern applications support OS/2 the way that they support e.g. MorphOS, and $139 is a steep price for a borderline useless OS that doesn't have a community like the Amiga-derived OSes do.