As mentioned by flohofwoe in the parent comment, Dawn and wgpu are both native implementations of WebGPU for use outside the browser. In particular wgpu-rs will allow users to target both the browser API and run natively.
I think it's debatable whether WebGPU is less capable than OpenGL 4.6 with AZDO. Even if it is less capable, it doesn't mean WebGPU is not a solution nor only usable by browsers.
Regardless it's the closest portable solution at the moment, and it will continue to add capabilities. The WebGPU CG has already had public meetings with Khronos to talk about running WebGPU on top of Vulkan, and running Vulkan Portability on top of WebGPU. So I think there's a lot of interest in making WebGPU succeed on both web and native targets.
What many of us want is a high level standard that doesn't require being a driver engineer to take advantage of modern GPUs.
If WebGPU ends up being as castrated as WebGL, then we either keep using OpenGL, or change to modern middleware engines (which is what most are doing), with the added benefit that they make the actual 3D API less relevant, it is just a checkbox.
OpenGL really is a legacy API in 2020. There is a massive mismatch between modern GPU design and the OpenGL API design. The only reason it wasn't left behind a decade ago is that there wasn't a decent replacement.
Today, with Vulkan and soon WebGPU available, there really is very little reason to hang on to OpenGL for anything but legacy.
The only platform that has marked it as legacy doesn't support Vulkan (directly) either so I'm not convinced it's dead yet. The simpler API for the minor performance trade-off is still compelling for lots of uses.
This I/O device thing seems to be a feature of one specific wasm engine that just exposes a simple framebuffer, so, it has nothing to do with anything else.
That really doesn't resolve anything, and just further demonstrates the really weird relationship with race and nationality that Americans have.
Namely: Why would it be called "white EUROPEAN" when basically none of the people it is applied to are European in any reasonable sense of the word? They may have great-great-great-grandparents who were European, but that does not make them European.
Americans do this all the time - they call themselves "german", "irish", or "italian", even when they are not born there, do not speak the language, have never even visited the country, and have zero exposure to the culture of that country.
This is a debate about language conventions, I can assure you that 0% of Americans claiming to be European in this sense actually think they live in Europe or are actual Europeans. There are some that speak the language and have exposure to the culture, but that is not what they are saying in this context.
In America, a nation of immigrants, people are often proud of where their family came from. This can trickle down a few generations. There is an increasing number of people that identify as "American" instead of where their family originally immigrated from, particularly in Appalachia and the South, so this could eventually change.
I understand the frustration from a European point of view, but I just wanted to provide some context to the phenomenon.
> have zero exposure to the culture of that country.
Even when that’s true—-and it often isn’t--they usually mean that they’re part of a specific American subculture.
Italian-Americans, for example, traditionally celebrate Christmas Eve with a “Feast of the Seven Fishes.” This dinner isn’t traditional in Italy, but it’s also not common in other American cultures either. An Irish-American family might have a turkey instead but they’re more likely to “observe” St. Patrick’s Day by eating soda bread and corned beef.
