[jimmaswell]

"Any browser that doesn't implement WebGL will soon be considered crippled"

Really? I don't think WebGL is that important. What's it really good for besides laggy browser games? What's with the hype on browser games anyway? They're always going to perform much slower than native code. I don't see WebGL becoming such a critical aspect of browsing that the average user would consider IE "crippled" for not supporting it.

[troymc]

Your challenge, should you choose to accept it, is to dream up an amazing and popular application of WebGL, now that a majority of web browsers will be supporting it. It might not be a game. Warning: imagination required.

[pbsdp]

It's hard to dream big about the browsers getting the modern iteration of something we've had in some form since 1992.

[ZeroGravitas]

How does that not apply to every single thing on the web, whether text, image, animation, video, or interaction?

[andybak]

You are assuming that a delivery mechanism can't be transformative.

[nailer]

ASM is only 2x slower than native (it was 10x prior to) and will be 1.5 soon. With a way better distribution model. Have you seen UE3 or UE4 demos?

[anon1385]

>ASM is only 2x slower than native

Only in wilfully misleading benchmarks. Allow use of SIMD and multithreading and asm.js can be as much as 50 times slower:

http://cdn.arstechnica.net/wp-content/uploads/2013/05/native...

http://cdn.arstechnica.net/wp-content/uploads/2013/05/classi...

[nailer]

Good point: re multithreading. I wonder if web workers could be used to resolve that?

[nitrogen]

Here's one application: http://youtu.be/zB7QEFCR1Zk?t=16m59s

[nawitus]

It will soon be used for numerous business apps as companies are porting their native applications to work on browsers.

[jimmaswell]

I just hope the desktop versions don't go away.