As a TypeScript dev, TypeScript is not pleasant to work with at all. I don’t love Java or C# but I’d take them any day of the week over anything JS-based. TypeScript provides the illusion of type safety without actually providing full type safety because of one random library whose functionality you depend on that returns and takes in any instead of using generic types. Unlike pretty much any other statically typed language, compiled TypeScript will do nothing to ensure typing at runtime, and won’t error at all if something else gets passed in until you try to use a method or field that it doesn’t have. It will just fail silently unless you add type checking to your functions/methods that are already annotated as taking in your desired types. Languages like Java and C# would throw an exception immediately when you try to cast the value, and languages like Rust and Go wouldn’t even compile unless you either handle the case or panic at that exact location. Pretty much the only language that handles this worse is Python (and maybe Lua? I don’t really know much about Lua though).
TLDR; TypeScript in theory is very different from TypeScript in practice and that difference makes it very annoying to use.
I have next to no experience with TypeScript, but want to make a case in defence of Python: Python does not pretend to have any kind of type safety, and more or less actively encourages duck typing.
Now, you can like or dislike duck typing, but for the kind of quick and dirty scripting or proof of concept prototyping that I think Python excels at, duck typing can help you get the job done much more efficiently.
In my opinion, it’s much more frustrating to work with a language that pretends to be type safe while not being so.
Because of this, I regularly turn off the type checking on my python linter, because it’s throwing warnings about “invalid types”, due to incomplete or outdated docs, when I know for a fact that the function in question works with whatever type I’m giving it. There is really no such thing as an “invalid type” in Python, because it’s a language that does not intend to be type-safe.
That’s entirely fair for the usecase of a small script or plugin, or even a small website. I’d quickly get annoyed with Python if I had to use it for a larger project though.
TypeScript breaks down when you need it for a codebase that’s longer than a few thousand lines of code. I use pure JavaScript in my personal website and it’s not that bad. At work where the frontend I work on has 20,000 lines of TypeScript not including the HTML files, it’s a massive headache.
I wholeheartedly agree: In my job, I develop mathematical models which are implemented in Fortran/C/C++, but all the models have a Python interface. In practice, we use Python as a “front end”. That is: when running the models to generate plots or tables, or whatever, that is done through Python, because plotting and file handling is quick and easy in Python.
I also do quite a bit of prototyping in Python, where I quickly want to throw something together to check if the general concept works.
We had one model that was actually implemented in Python, and it took less than a year before it was re-implemented in C++, because nobody other than the original dev could really use it or maintain it. It became painfully clear how much of a burden python can be once you have a code base over a certain size.
Pretty much the only language that handles this worse is Python (and maybe Lua? I don’t really know much about Lua though).
This is the case for literally all interpreted languages, and is an inherent part of them being interpreted.
However, while I recognize that can happen, I’ve literally never come across it in my time working on Typescript. I’m not sure what third party libraries you’re relying on but the most popular OAuth libraries, ORMs, frontend component libraries, state management libraries, graphing libraries, etc. are all written in pure Typescript these days.
This is the case for literally all interpreted languages, and is an inherent part of them being interpreted.
It’s actually the opposite. The idea of “types” is almost entirely made up by compilers and runtime environments (including interpreters). The only thing assembly instructions actually care about is how many bits a binary value has and whether or not it should be stored as a floating point, integer, or pointer (I’m oversimplifying here but the point still stands). Assembly instructions only care about the data in the registers (or an address in memory) that they operate on.
There is no part of an interpreted language that requires it to not have any type-checking. In fact, many languages use runtime environments for better runtime type diagnostics (e.g. Java and C#) that couldn’t be enforced at runtime in a purely compiled language like C or C++. Purely compiled binaries are pretty much the only environments where automatic runtime type checking can’t be added without basically recreating a runtime environment in the binary (like what languages like go do). The only interpreter that can’t have type-checking is your physical CPU.
If you meant that it is inherent to the language in that it was intended, you could make the case that for smaller-scale languages like bash, Lua, and some cases Python, that the dynamic typing makes it better. Working with large, complex frontends is not one of those cases. Even if this was an intentional feature of JavaScript, the existence of TypeScript at all proves it was a bad one.
However, while I recognize that can happen, I’ve literally never come across it in my time working on Typescript. I’m not sure what third party libraries you’re relying on but the most popular OAuth libraries, ORMs, frontend component libraries, state management libraries, graphing libraries, etc. are all written in pure Typescript these days.
This next example doesn’t directly return any, but is more ubiquitous than the admittedly niche libraries the code I work on depends on: Many HTTP request services in TypeScript will fill fields in as undefined if they’re missing, even if the typing shouldn’t allow for that because that type requirement doesn’t actually exist at runtime. Languages like Kotlin, C#, and Rust would all error because the deserialization failed when something that shouldn’t be considered nullable had an empty value. Java might also have options for this depending on the serialization library used.
As a TypeScript dev, TypeScript is not pleasant to work with at all. I don’t love Java or C# but I’d take them any day of the week over anything JS-based. TypeScript provides the illusion of type safety without actually providing full type safety because of one random library whose functionality you depend on that returns and takes in
anyinstead of using generic types. Unlike pretty much any other statically typed language, compiled TypeScript will do nothing to ensure typing at runtime, and won’t error at all if something else gets passed in until you try to use a method or field that it doesn’t have. It will just fail silently unless you add type checking to your functions/methods that are already annotated as taking in your desired types. Languages like Java and C# would throw an exception immediately when you try to cast the value, and languages like Rust and Go wouldn’t even compile unless you either handle the case or panic at that exact location. Pretty much the only language that handles this worse is Python (and maybe Lua? I don’t really know much about Lua though).TLDR; TypeScript in theory is very different from TypeScript in practice and that difference makes it very annoying to use.
Bonus meme:
I have next to no experience with TypeScript, but want to make a case in defence of Python: Python does not pretend to have any kind of type safety, and more or less actively encourages duck typing.
Now, you can like or dislike duck typing, but for the kind of quick and dirty scripting or proof of concept prototyping that I think Python excels at, duck typing can help you get the job done much more efficiently.
In my opinion, it’s much more frustrating to work with a language that pretends to be type safe while not being so.
Because of this, I regularly turn off the type checking on my python linter, because it’s throwing warnings about “invalid types”, due to incomplete or outdated docs, when I know for a fact that the function in question works with whatever type I’m giving it. There is really no such thing as an “invalid type” in Python, because it’s a language that does not intend to be type-safe.
That’s entirely fair for the usecase of a small script or plugin, or even a small website. I’d quickly get annoyed with Python if I had to use it for a larger project though.
TypeScript breaks down when you need it for a codebase that’s longer than a few thousand lines of code. I use pure JavaScript in my personal website and it’s not that bad. At work where the frontend I work on has 20,000 lines of TypeScript not including the HTML files, it’s a massive headache.
I wholeheartedly agree: In my job, I develop mathematical models which are implemented in Fortran/C/C++, but all the models have a Python interface. In practice, we use Python as a “front end”. That is: when running the models to generate plots or tables, or whatever, that is done through Python, because plotting and file handling is quick and easy in Python.
I also do quite a bit of prototyping in Python, where I quickly want to throw something together to check if the general concept works.
We had one model that was actually implemented in Python, and it took less than a year before it was re-implemented in C++, because nobody other than the original dev could really use it or maintain it. It became painfully clear how much of a burden python can be once you have a code base over a certain size.
This is the case for literally all interpreted languages, and is an inherent part of them being interpreted.
However, while I recognize that can happen, I’ve literally never come across it in my time working on Typescript. I’m not sure what third party libraries you’re relying on but the most popular OAuth libraries, ORMs, frontend component libraries, state management libraries, graphing libraries, etc. are all written in pure Typescript these days.
It’s actually the opposite. The idea of “types” is almost entirely made up by compilers and runtime environments (including interpreters). The only thing assembly instructions actually care about is how many bits a binary value has and whether or not it should be stored as a floating point, integer, or pointer (I’m oversimplifying here but the point still stands). Assembly instructions only care about the data in the registers (or an address in memory) that they operate on.
There is no part of an interpreted language that requires it to not have any type-checking. In fact, many languages use runtime environments for better runtime type diagnostics (e.g. Java and C#) that couldn’t be enforced at runtime in a purely compiled language like C or C++. Purely compiled binaries are pretty much the only environments where automatic runtime type checking can’t be added without basically recreating a runtime environment in the binary (like what languages like go do). The only interpreter that can’t have type-checking is your physical CPU.
If you meant that it is inherent to the language in that it was intended, you could make the case that for smaller-scale languages like bash, Lua, and some cases Python, that the dynamic typing makes it better. Working with large, complex frontends is not one of those cases. Even if this was an intentional feature of JavaScript, the existence of TypeScript at all proves it was a bad one.
This next example doesn’t directly return
any, but is more ubiquitous than the admittedly niche libraries the code I work on depends on: Many HTTP request services in TypeScript will fill fields in as undefined if they’re missing, even if the typing shouldn’t allow for that because that type requirement doesn’t actually exist at runtime. Languages like Kotlin, C#, and Rust would all error because the deserialization failed when something that shouldn’t be considered nullable had an empty value. Java might also have options for this depending on the serialization library used.