Hacker News Reader

Comment by hultner | Hacker News Reader

A strongly typed dynamic language will raise an type error if you for instance try to add two incompatible types, for instance in python

  Python 3.9.0 (default, Oct 10 2020, 11:40:52)
  [Clang 11.0.3 (clang-1103.0.32.62)] on darwin
  Type "help", "copyright", "credits" or "license" for more information.
  >>> 1+"1"
  Traceback (most recent call last):
    File "<stdin>", line 1, in <module>
  TypeError: unsupported operand type(s) for +: 
  'int' and 'str'

While a weakly typed dynamic language like JS will allow for it

  Welcome to Node.js v16.0.0.
  Type ".help" for more information.
  > 1 + "1"
  '11'

View on HN

If the language allows it the types are (ipso facto) not incompatible; “weak typing” is more a subjective statement about how well a language lines up with the speakers mental model of what types should be compatible than an objective one about falsifiable properties of the type system.

Agree. Weakly typed is so loosely defined and subjective to be a completely useless adjective. It basically a synonym for 'language I dislike'.

How about some languages allow implicit type conversions, and others do not.

Is Scala weakly typed then because it has implicits?

I don't know, but JS and PHP implicit conversions are widely acknowledged to cause plenty of issues, which Python and Ruby do not have since you have much fewer cases of implicit conversion.

I contend there is no such thing as "weakly-typed" because it suggests it exists in opposition to strongly-typed systems - but instead I feel "weakly-typing" should be considered the absence of type-correctness checking by either the compiler nor the runtime - and that the term "weakly typed" and related should not be used to refer to languages with plenty of implicit conversion between types, because in those languages (like JS) the implicit conversions are very, very, well-defined and predictable ahead of time - TypeScript models JavaScript's implicit type conversions accurately, too.

JavaScript's more forgiving, and often surprising, implicit conversions are still rigidly-defined .

Remember that implicit conversions should never (ostensibly) throw any exception or cause any errors because implicit conversion should always succeed because it should only be used to represent an obvious and safe narrowing-conversion. A narrowing-conversion necessarily involves data-loss, but it's safe because all of the data needed by the rest of the program is being provably retained.

> I contend there is no such thing as "weakly-typed" because it suggests it exists in opposition to strongly-typed systems - but instead I feel "weakly-typing" should be considered the absence of type-correctness checking by either the compiler nor the runtime

This doesn't feel like a very useful definition. What languages would fall under the category of "weakly typed" by your definition? I can only think of C, and even that is only true for a subset of the language (when you're messing about with pointers).

> Remember that implicit conversions should never (ostensibly) throw any exception or cause any errors because implicit conversion should always succeed because it should only be used to represent an obvious and safe narrowing-conversion.

    > const fun = (x, y) => (x + y) / y
    > fun(1, 2)
    1.5
    > fun("1", 2)
    6

No exception? Yes. No error? Nope, implicit conversion absolutely just caused an error. If I got that "1" from an input field and forgot to cast it to an int, I want to be told at the soonest possible juncture, I don't want to rely on me noticing that the result doesn't look right. As it is, "weak" feels like a good word for the types here. They don't hold their form well.

Weakly typed / strongly typed are not really useful terms - there are too many separate definitions, all of which are pretty much equally valid.

I thought we were all agreed that the strong - weak type axis was to what degree the language automatically converts types?

'Indeed, the phrases are not only poorly defined, they are also wrong, because the problem is not with the “strength” of the type checker but rather with the nature of the run-time system that backs them. The phrases are even more wrong because they fail to account for whether or not a theorem backs the type system.'

29.5 Types Versus Safety

"Programming and Programming Languages"

https://papl.cs.brown.edu/2020/safety-soundness.html#%28part...

That's one of the 7 definitions listed on wikipedia.

I think it's more germane to refer to that as weak implicit type casting.

Reply on news.ycombinator.com