News for hackers. Stuff that matters.
Re:The history of C# and TypeScript with Anders Hejlsberg | ... by jolux on Sunday February 01, 2026 @11:07AM (#46847123)
Checked exceptions are bad because people just catch them and rethrow RuntimeException. Proper errors as data would be much preferable.
Re:The history of C# and TypeScript with Anders Hejlsberg | ... by lateforwork on Sunday February 01, 2026 @11:12AM (#46847159)
> people just catch them and rethrow

People don't have to. There is a lot of misinformation out there about checked exceptions, and people are responding to the misinformation.

Re:The history of C# and TypeScript with Anders Hejlsberg | ... by vips7L on Sunday February 01, 2026 @12:53PM (#46847916)
Checked exceptions are proper errors as data. There is no difference between Result[T, E] and T throws E. The same people that just rethrow as RuntimeException would call unwrap() on Result and panic. The main problem with _Java_'s implementation of checked exceptions is that they're not fully in the type system because of Java's weak type system. That is solvable with a strong type system though, like in Scala or Swift.
Re:The history of C# and TypeScript with Anders Hejlsberg | ... by bigstrat2003 on Sunday February 01, 2026 @08:47PM (#46851452)
You can misuse the errors as data paradigm just like you can misuse checked exceptions. That doesn't make either paradigm bad, it makes the programmers who misuse them bad.