3 ms·

I see many downsides to mixing languages in a codebase, and no upside. JS and Python have eaten the world and can do everything every other language can do (so

I see many downsides to mixing languages in a codebase, and no upside. JS and Python have eaten the world and can do everything every other language can do (something which is true of all languages because of Turing). There's not enough benefit in your new language to overcome the switching costs -- and that's not personal, it's true of most every language anyone can or will suggest. We've hit peak language and passed it. Web Assembly and ChatGPT/Copilot actually work against the invention of new languages because with WASM all existing languages can be packaged up inside JS apps and run in the browser. And with LLMs, you get better results when there are more examples of code in the corpus. There won't be any examples of MSN2 nor of other invented languages. Finally, MSN2 is a terrible name. It was the name of a failed Microsoft "easy onramp" product like AOL that served as a home page for the internet decades ago.

unsolicited 2y ago
Even if I agreed with your opinions, I am not sure this would be the most constructive way to reply to someone who has put a lot of effort into a project like this. > I see many downsides to mixing languages in a codebase, and no upside. You have never heard of someone using Java and SQL in the same project before? Are you completely against any form of language-oriented programming? Are the Racket researchers simply wasting their time? > We've hit peak language and passed it. Review your comment in 50 years and see if you still feel that way. Obviously, many advancements will have been made by then. Languages are regularly being released with novel ideas. New languages are consistently seeing adoption. > Web Assembly and ChatGPT/Copilot actually work against the invention of new languages because with WASM all existing languages can be packaged up inside JS apps and run in the browser. Ignoring that this is an extremely narrow view—not everyone is trying to run something in a browser—it would be a tragedy, in more ways than one, to stop the development of new ideas because LLMs are only capable of regurgitating old ones. > Finally, MSN2 is a terrible name. True, but there are better ways to say this.
- dtagames 2y ago
  I've invented a few unsuccessful software products myself and worked on successful ones, too. I don't think it's any service to the OP to pull punches here. Anyone who invents a language is obviously bright and probably well intentioned. It would be best to redirect those talents now, methinks.

unsolicited 2y ago

Even if I agreed with your opinions, I am not sure this would be the most constructive way to reply to someone who has put a lot of effort into a project like this. > I see many downsides to mixing languages in a codebase, and no upside. You have never heard of someone using Java and SQL in the same project before? Are you completely against any form of language-oriented programming? Are the Racket researchers simply wasting their time? > We've hit peak language and passed it. Review your comment in 50 years and see if you still feel that way. Obviously, many advancements will have been made by then. Languages are regularly being released with novel ideas. New languages are consistently seeing adoption. > Web Assembly and ChatGPT/Copilot actually work against the invention of new languages because with WASM all existing languages can be packaged up inside JS apps and run in the browser. Ignoring that this is an extremely narrow view—not everyone is trying to run something in a browser—it would be a tragedy, in more ways than one, to stop the development of new ideas because LLMs are only capable of regurgitating old ones. > Finally, MSN2 is a terrible name. True, but there are better ways to say this.

dtagames 2y ago
I've invented a few unsuccessful software products myself and worked on successful ones, too. I don't think it's any service to the OP to pull punches here. Anyone who invents a language is obviously bright and probably well intentioned. It would be best to redirect those talents now, methinks.