4 ms·

> The reason that Google Native Client is not set fair to become an "open Web standard" is because three of the four non-MS vendors oppose it for reasons of sel

by tiles 15y ago

> The reason that Google Native Client is not set fair to become an "open Web standard" is because three of the four non-MS vendors oppose it for reasons of self-interest. As a user, I don't want to see NaCl's compile-once-per-each-architecture model become part of the web. It's true browsers oppose it out of self interest, but NaCl has its own problems that haven't been solved. PNaCl is a good start and much more in the vein of an open web that isn't tied to a particular platform. And whatever benefits an Open Web, Mozilla will inevitably support, as it sustains their business model. NaCl doesn't do anything for browser makers, but it also doesn't help me as a user.

leoc 15y ago
So what remaining problems with NaCl's compile-once-per-each-architecture model do you believe can't be solved by the combination of multi-architecture compilers (llvm, gcc ...) on the developer/server end and PNaCl on the client? Obviously the implementation is incomplete and immature, but Mozilla and Opera aren't even pretending that their opposition to NaCl is based on the immaturity of the implementation. > And whatever benefits an Open Web, Mozilla will inevitably support, as it sustains their business model. Mozilla inevitably supports and benefits from an Open Web only to the extent that you accept Mozilla's rather Newspeak definition of the Open Web as "a Web whose client API and runtime is under the shared control of a small group of browser vendors". If you have a less Orwellian kind of openness in mind, then it's clearly not true: Mozilla benefits from its shared control over the Web platform in largely the same way that MS benefits from its control over the Windows platform, and just like MS it has a strong self-interest in not seeing its platform "commoditised". As if that weren't perfectly clear already, then Chromeless underlines it. I assume you accept that Mozilla has a self-interest in seeing Chromeless succeed? The reason that Chromeless has a chance of hitting the big-time on the desktop is almost solely because Javascript and friends are popular on the Web client, and the ultimate reason for that is because JS and friends are bolted in and difficult to avoid on the web client.
- bzbarsky 15y ago
  Well, for a start there's the problem that LLVM itself is architecture-dependent, last I checked. It's not as arhictecture-dependent as assembly but the last time I looked you couldn't generate the same LLVM on x86 and x86-64, say. Has that changed? It looks like PNaCl aims to "solve" this problem by using 32-bit-targeted LLVM... including 32-bit pointer alignment. It's not immediately obvious how portable this would be to some hardware, and more importantly I'm not aware of any plans to support anything other than x86 and ARM initially (well, and x86-64 using its ability to run 32-bit code as far as I can tell). Am I just missing something on this front? If I'm not missing anything, then this project's main impact if it caught on would be to restrict the set of hardware on which you can access web content, which is NOT a good thing.
  - daeken 15y ago
    > Well, for a start there's the problem that LLVM itself is architecture-dependent, last I checked. It's not as arhictecture-dependent as assembly but the last time I looked you couldn't generate the same LLVM on x86 and x86-64, say. Has that changed? LLVM bytecode is 100% architecture and platform independent, unless you call into platform-specific bits or into blobs of machine code.
    - bzbarsky 15y ago
      OK, so in that case what http://llvm.org/docs/FAQ.html#platformindependent http://llvm.org/docs/FAQ.html#platformindependent is telling me is that it's not possible to produce said LLVM bytecode from C or C++ code in general, right? And if I read http://llvm.org/docs/LangRef.html http://llvm.org/docs/LangRef.html correctly, some bitcode is "not supported by all targets". How is that reconciled with "100% platform and architecture independent"?
      - daeken 15y ago
        Hmm, you're quite right. A better way of putting this would be: a large subset of LLVM IR is platform independent, but not all. My mistake.
    - deleted 15y ago

by tiles 15y ago

leoc 15y ago

So what remaining problems with NaCl's compile-once-per-each-architecture model do you believe can't be solved by the combination of multi-architecture compilers (llvm, gcc ...) on the developer/server end and PNaCl on the client? Obviously the implementation is incomplete and immature, but Mozilla and Opera aren't even pretending that their opposition to NaCl is based on the immaturity of the implementation. > And whatever benefits an Open Web, Mozilla will inevitably support, as it sustains their business model. Mozilla inevitably supports and benefits from an Open Web only to the extent that you accept Mozilla's rather Newspeak definition of the Open Web as "a Web whose client API and runtime is under the shared control of a small group of browser vendors". If you have a less Orwellian kind of openness in mind, then it's clearly not true: Mozilla benefits from its shared control over the Web platform in largely the same way that MS benefits from its control over the Windows platform, and just like MS it has a strong self-interest in not seeing its platform "commoditised". As if that weren't perfectly clear already, then Chromeless underlines it. I assume you accept that Mozilla has a self-interest in seeing Chromeless succeed? The reason that Chromeless has a chance of hitting the big-time on the desktop is almost solely because Javascript and friends are popular on the Web client, and the ultimate reason for that is because JS and friends are bolted in and difficult to avoid on the web client.

bzbarsky 15y ago
Well, for a start there's the problem that LLVM itself is architecture-dependent, last I checked. It's not as arhictecture-dependent as assembly but the last time I looked you couldn't generate the same LLVM on x86 and x86-64, say. Has that changed? It looks like PNaCl aims to "solve" this problem by using 32-bit-targeted LLVM... including 32-bit pointer alignment. It's not immediately obvious how portable this would be to some hardware, and more importantly I'm not aware of any plans to support anything other than x86 and ARM initially (well, and x86-64 using its ability to run 32-bit code as far as I can tell). Am I just missing something on this front? If I'm not missing anything, then this project's main impact if it caught on would be to restrict the set of hardware on which you can access web content, which is NOT a good thing.
- daeken 15y ago
  > Well, for a start there's the problem that LLVM itself is architecture-dependent, last I checked. It's not as arhictecture-dependent as assembly but the last time I looked you couldn't generate the same LLVM on x86 and x86-64, say. Has that changed? LLVM bytecode is 100% architecture and platform independent, unless you call into platform-specific bits or into blobs of machine code.
  - bzbarsky 15y ago
    OK, so in that case what http://llvm.org/docs/FAQ.html#platformindependent http://llvm.org/docs/FAQ.html#platformindependent is telling me is that it's not possible to produce said LLVM bytecode from C or C++ code in general, right? And if I read http://llvm.org/docs/LangRef.html http://llvm.org/docs/LangRef.html correctly, some bitcode is "not supported by all targets". How is that reconciled with "100% platform and architecture independent"?
    - daeken 15y ago
      Hmm, you're quite right. A better way of putting this would be: a large subset of LLVM IR is platform independent, but not all. My mistake.
  - deleted 15y ago