News for hackers. Stuff that matters.
Re:Google Introduces Portable Native Client by duaneb on Saturday May 18, 2013 @03:31AM (#5728178)
I don't see the need for asm.js and emscripten. They seemed like a great idea, but it has now been overtaken by PNaCL.

We can all make unsubstantiated claims, and not all of us want to use the ghetto that is modern javascript.

Re:Google Introduces Portable Native Client by zobzu on Saturday May 18, 2013 @05:21AM (#5728342)
I think one big difference is that if right now I play the GL demo on Firefox using asm.js, it works. If I play the same demo on Chrome.. it works.

Now if I play nacl stuff on ANYTHING else than chrome is does NOT work.

For me its not about having a nice language, it's about being standardized.

Else, I can just provide C binaries, they're faster than nacl anyway.

Re:Google Introduces Portable Native Client by duaneb on Monday May 20, 2013 @01:22AM (#5736046)
This is by far the best argument, but how is the performance of asm.js under browsers other than firefox?
Re:Google Introduces Portable Native Client by vanderZwan on Saturday May 18, 2013 @05:25AM (#5728347)
> Now if I play nacl stuff on ANYTHING else than chrome is does NOT work.

That's purely an adoption-by-browser problem.

> Else, I can just provide C binaries, they're faster than nacl anyway.

The whole point of both nacl and asm.js was sandboxing.

Re:Google Introduces Portable Native Client by karlmdavis on Saturday May 18, 2013 @08:07AM (#5728629)
> That's purely an adoption-by-browser problem.

... which is a pretty big problem, right? Especially since I don't think FF is likely to pick up NaCL unless/until hell freezes over.

Re:Google Introduces Portable Native Client by zobzu on Tuesday May 21, 2013 @09:25PM (#5748224)
I don't see where the point of asm.js is sandboxing. In fact, that's pretty much beside the point. Sandboxing in NaCL is just a necessity.