3 ms·

On one side you have companies like Google creating entirely new programming languages (Carbon) that aren’t really solving a problem and aren’t that different f

On one side you have companies like Google creating entirely new programming languages (Carbon) that aren’t really solving a problem and aren’t that different from one or more existing languages (Rust, Go), and on the other you have people like OP writing a COBOL compiler in C#. What a world.

jay-barronville 4y ago
Lol. This is why I love HN though!
pjmlp 4y ago
Carbon solves one problem, ISO doesn't care about what Google thinks regarding ABI break.
- gymbeaux 4y ago
  Carbon is incredibly niche. I can't imagine anyone is using it for greenfield stuff. The idea that I can refactor my code base one class or even one method at a time is neat, but if I am the guy in the C-Suite dropping $10M+ each year on SWE salaries, I would rather wait for C++25 or rewrite my codebase in something "mainstream" like Rust than spend man-hours rewriting C++ in a baby programming language- created and maintained by Google of all companies.
  - pjmlp 4y ago
    Carbon team is the first one to assert that Carbon is for Google internal purposes, or anyone that eventually shares their point of view regarding C++, everyone else should go with Rust or any other safer alternatives. They also admit it is an experiment that might fail, and they will go back writing C++ on the codebases that they fail to migrate. So it is kind of strage that people hype it for stuff the authors never planned for.
    - gymbeaux 4y ago
      I don't know that I have seen anyone hype it, I just think it's funny that Google invented yet another programming language, and that (to me anyway) it's really just to gain access to features that are probably coming down the pipe in C++ in due time anyway.
      - jasode 4y ago
        >Google invented yet another programming language, [...] just to gain access to features that are probably coming down the pipe in C++ in due time anyway. That's the opposite of what Google perceives. There is no timetable for a future C++23 or even a later version that will make their experiments with Carbon redundant. Are you already familiar with the Chandler Carruth blog post? https://github.com/carbon-language/carbon-lang/blob/trunk/docs/project/difficulties_improving_cpp.md https://github.com/carbon-language/carbon-lang/blob/trunk/do... That post also has a link to the Titus Winters pdf explaining some of the ABI breakage that the committee does not currently want to prioritize. I'm skeptical Carbon will spur industry-wide usage outside of Google's internal work. Nevertheless, their rationale for Carbon is in response to the current committee's position of prioritizing ABI backward compatibility over performance improvements.
        gymbeaux 4y ago
        Interesting. I haven't seen that blog post. Thank you for sharing it! True enough, there's no guarantee C++ will eventually get the features Google is getting now with Carbon, and I realize they are a software company, and if anyone has a reason to invent their own internal-use programming language, it's them.

jay-barronville 4y ago

Lol. This is why I love HN though!

pjmlp 4y ago

Carbon solves one problem, ISO doesn't care about what Google thinks regarding ABI break.

gymbeaux 4y ago
Carbon is incredibly niche. I can't imagine anyone is using it for greenfield stuff. The idea that I can refactor my code base one class or even one method at a time is neat, but if I am the guy in the C-Suite dropping $10M+ each year on SWE salaries, I would rather wait for C++25 or rewrite my codebase in something "mainstream" like Rust than spend man-hours rewriting C++ in a baby programming language- created and maintained by Google of all companies.
- pjmlp 4y ago
  Carbon team is the first one to assert that Carbon is for Google internal purposes, or anyone that eventually shares their point of view regarding C++, everyone else should go with Rust or any other safer alternatives. They also admit it is an experiment that might fail, and they will go back writing C++ on the codebases that they fail to migrate. So it is kind of strage that people hype it for stuff the authors never planned for.
  - gymbeaux 4y ago
    I don't know that I have seen anyone hype it, I just think it's funny that Google invented yet another programming language, and that (to me anyway) it's really just to gain access to features that are probably coming down the pipe in C++ in due time anyway.
    - jasode 4y ago
      >Google invented yet another programming language, [...] just to gain access to features that are probably coming down the pipe in C++ in due time anyway. That's the opposite of what Google perceives. There is no timetable for a future C++23 or even a later version that will make their experiments with Carbon redundant. Are you already familiar with the Chandler Carruth blog post? https://github.com/carbon-language/carbon-lang/blob/trunk/docs/project/difficulties_improving_cpp.md https://github.com/carbon-language/carbon-lang/blob/trunk/do... That post also has a link to the Titus Winters pdf explaining some of the ABI breakage that the committee does not currently want to prioritize. I'm skeptical Carbon will spur industry-wide usage outside of Google's internal work. Nevertheless, their rationale for Carbon is in response to the current committee's position of prioritizing ABI backward compatibility over performance improvements.
      - gymbeaux 4y ago
        Interesting. I haven't seen that blog post. Thank you for sharing it! True enough, there's no guarantee C++ will eventually get the features Google is getting now with Carbon, and I realize they are a software company, and if anyone has a reason to invent their own internal-use programming language, it's them.