Progress on TypeScript 7 – December 2025(devblogs.microsoft.com) |
Progress on TypeScript 7 – December 2025(devblogs.microsoft.com) |
- Go-to-Definition: For variables, jump to the declaration. For functions/classes/types, jump to its types or implementation.
- Go-to-Type-Definition: For variables, jump to its type declaration instead of the variable declaration.
- Go-to-Implementation: For interfaces/abstruct classes, jump to (multiple) implementations of the method. For compiled TypeScript file (like them under node_modules), jump to its implementation rather than .d.ts files.I really don’t see TypeScript destined for long term success with the rise of Wasm on the web. It is only where it is currently because it has operated at a huge advantage of being the only viable solution to a problem that every web developer has needed.
I say this as someone who writes a lot of typescript professionally and a lot of Dart for everything else I possibly can. The difference between the two is huge on almost every factor I can think of where Dart wins by a considerable margin. I’m thankful that Typescript existed at a time when it did but I don’t think it’s in anyway actually a good or sensible language (because it’s so closely to JavaScript) and I can’t wait to not have to use it in the future because better options exist and are becoming viable for more use cases.
Having used Swift and Kotlin in the past, their typesystems don’t come close to the flexibility with which you can wrangle data in TS. Many concepts that I would consider fundamental are either not present or require complicated syntax eg.: product types A & B, sum types A | B, { …spreading }.
And where Swift needs custom syntax such as `guard let` and `case let` to narrow types, Typescript is able to analyze normal control flow.
For practical reasons, there are a ton of ugly workarounds, but I’d rather live with those than go back to the rigid “Java++” languages.
https://github.blog/developer-skills/programming-languages-a...
1. Chrome/v8 takes TS code, compiles it down to JS internally, erases types, and then runs it like normal. This isn't going to be too hard to do, but also isn't going to be very meaningful. Compiling is a one-step process in any case, and plenty of tooling exists to make it seamless.
2. Chrome/v8 actually understands TS types at runtime, and throws exceptions for mismatches. This isn't going to be possible without a major rewrite of the v8 engine and the ECMAScript spec itself.
And a big challenge for both of these is that TypeScript is iterating at too fast a pace for something like Chrome to keep up. It's best to just leave versioning and compilation for the developer to manage and give end users a consistent JavaScript experience.
I don't think any type of understanding TS would require changing ECMAScript spec. Would a TypeScript-understanding parser not be able to handle normal ECMAScript correctly? It could switch between two modes based on the file type.
For option 1 the speed of TS development is not an issue, as Chrome would only need to include some up-to-date compiler, and the TS files could specify their TS version. But doing TS compilation in the browser would only be a small nice thing for devs, for website users it would be a downgrade, as the page load would be slower because of the compiling and the larger file sizes (JS files can already be very big these days).
That’s not how TypeScript works. You execute it after erasing the types.
There's effort towards "erasable syntax" but in practice not many really want to run anything more than basic TS in the browser, simply because you lose things like 'import _ from "lodash", which requires further setup (import maps) or is impractical in larger projects.
The real reason however is that TS is just a language, not part of any standard, and browsers are not just going to support random languages headed by corporations (Microsoft) and that are not community projects.
I think importmaps are reasonably practical even for larger projects today. I know Firefox supporting multiple importmaps in the same HTML is still a wishlist item for some to avoid more scenarios that need an importmap compile step in very large projects, but importmaps are relatively easy to automate (it's just JSON wrapped in a script tag; merging them can be just a matter of a couple Object.assign calls or object spread notation; the multiple importmaps standard that Firefox is behind on is just that, a basic Object.assign merge).
> The real reason however is that TS is just a language, not part of any standard
Right, which is why the TC-39 (committee that owns JS) proposal for "erasable syntax" for "type annotations" isn't Typescript-specific and isn't a pure superset of Typescript syntax. The node implementation of erasable syntax doesn't use the Typescript compiler and doesn't support the full Typescript syntax and is relatively close to the TC-39 proposal.
> not just going to support random languages headed by corporations (Microsoft) and that are not community projects.
Nitpicking somewhat, but Typescript has been open source since it was publicly launched and a fair enough chunk over the years of contributors aren't/haven't been Microsoft employees. It's fairly reasonable to say it is a community project. I appreciate the point that it is still seen as a specific-vendor solution and the point of separating the JS language standard proposal from being "just do what Microsoft decides is right for Typescript" (as TC-39 is trying to do), but to be fair to Typescript, it is a also bit wilder and more open than that "just a Microsoft solution" at this point (and arguably also for most of its history).
This is not only due to implementation complexity, but also to keep TS to be able to change. Or even to build an entire different JS superset. With that proposal, even flow could be natively executed.
1. Hidden classes can't be created from TS interfaces because they don't represent the full data of the underlying object
2. You don't really ever want to compile code the first time you see it, because that takes a lot of memory and extra CPU cycles. By the time code has run enough to be worth compiling, you probably have enough profile data to optimize better than you could with data from the types anyway.
3. Many of the juiciest optimizations come from types that aren't representable in TS, like integers.
4. Including all the types for all your code and deps (literally all the .d.ts) is huge, and the size increase alone might nullify any performance benefit.
those are intersection types and union types and are much more unique and rare in type systems eg all typed functional languages have product/sum types but typescript is the only mainstream language I know to have unions and intersection.
But it's a double edged sword:
- You can brand types with additional information. For example, React Query does this to keep track of which type is stored under which cache-key. This is where most strongly typed languages just give up.
- You can progressively extend/narrow types without having to worry about inheritance. Invariants can be statically checked by encoding them in a type: eg IssuedInvoice = Invoice & { … }. A function that needs only a subset of a domain-object can specify this rather than requiring the email-address to verify a phone number.
- You can even emulate nominal types using keyed unions.
The first two alone have saved me from more errors than structural typing introduced.