the javascript splitting proposal
Regarding this: https://devclass.com/2024/10/22/should-javascript-be-split-into-two-languages-new-google-driven-proposal-divides-opinion/
It's important to remember that when Google says "the usecases have not materialized", what they *mean* is "we do not have a usecase for this".
BigInts have, in fact, seen widespread use in applications that would otherwise be very difficult to implement correctly, just most of them do not run in a browser, and so Google doesn't give a shit about them.
I do not trust this proposal one bit. This whole thing is primarily going to benefit those with a well-established developer tooling division, which not coincidentally includes Google.
the javascript splitting proposal (2)
A non-exhaustive list of things where having BigInts at your disposal (as opposed to floats) is really quite helpful and/or necessary:
- Cryptography
- Database development
- Fixed-precision decimals (you know, like working with amounts of money...)
- Parsers and protocol handlers
- Internal state tracking
the javascript splitting proposal (3)
And to be clear, I'm speaking from a position where I think too much stuff has been added to JS that doesn't need to be there. Again, not coincidentally, largely because of Google. I *want* a more restricted set of stuff in the language core.
But *what* this presentation proposes as the things to split off, is extremely telling about what their true intentions are, and I absolutely do not believe that they are good for anyone but Google.