5 ms·

> Show me a counterexample, that's all I ask. I'd give all the rhetoric in the world for a single damned counterexample of a program format that's lasted since

> Show me a counterexample, that's all I ask. I'd give all the rhetoric in the world for a single damned counterexample of a program format that's lasted since the 80s. That's what's at stake. ELF was defined in the late 80s/early 90s as part of SVR4. COFF was defined as part of SVR3. Both still exist today. The JVM bytecode was defined in 1995 and has been evolved compatibly. I'm curious, though. Do you require the same longevity of web formats that were defined in the 90s? Things have changed quite a bit, and attempting to browse the modern web with a 90s browser is futile. Given that, I don't really understand what you're arguing for. > Regardless of my opinion of NaCl, this is stuff that should be getting defined slowly and in the open by bearded old men that have spent their lives in glacial, trusted industries like telecoms, or just about any other mature technology industry. Us 'bearded old men' aren't on the web, and its only improvements like SPDY, WebSockets, and NaCL that have made me the slightest bit interested in the web as an application platform.

pyre 14y ago
> attempting to browse the modern web with a > 90s browser is futile I think the inverse is what we're talking about. Will we be able to browse NaCl sites in future browsers (possibly after an architecture change away from x86)? As it stands, we would probably be able to browse a 90's site on modern browsers. It sounds like if NaCl compiles things down to ASM x86 (or really AMD64) (?). If that's true, then it's possible for things to not be forward compatible. I'll admit that it doesn't seem likely, especially after what happened with Itanium and AMD64/x86-64, but we could see things move towards ARM.
- pifflesnort 14y ago
  > I think the inverse is what we're talking about. Will we be able to browse NaCl sites in future browsers (possibly after an architecture change away from x86)? As it stands, we would probably be able to browse a 90's site on modern browsers. Given PNaCL, the answer should be 'yes'.
  - mbetter 14y ago
    For future Google browsers, maybe.
prodigal_erik 14y ago
> attempting to browse the modern web with a 90s browser is futile And it's hard to imagine stronger evidence that the people designing the "modern web" are not the few who are willing and able to do a competent job of it.

pyre 14y ago

> attempting to browse the modern web with a > 90s browser is futile I think the inverse is what we're talking about. Will we be able to browse NaCl sites in future browsers (possibly after an architecture change away from x86)? As it stands, we would probably be able to browse a 90's site on modern browsers. It sounds like if NaCl compiles things down to ASM x86 (or really AMD64) (?). If that's true, then it's possible for things to not be forward compatible. I'll admit that it doesn't seem likely, especially after what happened with Itanium and AMD64/x86-64, but we could see things move towards ARM.

pifflesnort 14y ago
> I think the inverse is what we're talking about. Will we be able to browse NaCl sites in future browsers (possibly after an architecture change away from x86)? As it stands, we would probably be able to browse a 90's site on modern browsers. Given PNaCL, the answer should be 'yes'.
- mbetter 14y ago
  For future Google browsers, maybe.

prodigal_erik 14y ago

> attempting to browse the modern web with a 90s browser is futile And it's hard to imagine stronger evidence that the people designing the "modern web" are not the few who are willing and able to do a competent job of it.