Just like LaTeX allows you to write documents precisely and reproducibly, this computational English would make automatic machine translation extremely difficult. Even a small snippet of computation could drastically change the meaning if it is misinterpreted. Over time, anyone who values precision might stick to English by default, creating a global lock-in for formal communication.
Casual conversation could still happen in any language, but for technical writing, legal documents, instructions, or rigorous journalism, computational English could become the universal standard. Its adoption would depend on its ability to guarantee exact meaning rather than ease of learning.
Would this lock-in ensure that English will be the universal language of the world forever, just like LaTeX for scientists?
In software development, we even refined it greatly to meet our needs. That's what programming languages are.
You mean restrained. More specifically what you're proposing can formally be referred to as controlled natural language with executable semantics. Some attempts similar to this have been Attempto Controlled English and ClearTalk. (And Logos that someone showed here recently.)
>text would not just require reading; it would require executing algorithms embedded in the language itself
Arguably mathematics is just that.
>just like LaTeX for scientists
Future doesn't look very bright for LaTeX with Typst getting traction.