I don’t think that’s quite right. I haven’t written C++ since the 90s, and I use IDEs (Emacs and Zed), but I still sometimes reach a mental threshold where I look at my screen and see way too many names to have to hold in my mental buffer, then decide to make them more explicit.
IDK what the state of Emac/Zed is in terms of type information (I'm sure it depends on the language in question). For Jetbrains/Eclipse/Netbeans if there's a question about a type you can ctrl+click on the type and immediately pull all information about it.
In java, I rarely pay attention to the `import` section (and I know most devs at my company).
You can look up `using namespace std;` in google and you'll find a lot of articles saying it's a bad practice in C++. Everyone recommends writing the full `std::cout` rather than `cout`.
All modern editors do that pretty well with language servers now. Specifically, Emacs and Zed do this perfectly with Rust.
I do think it’s down to personal preference. With the fully qualified names, I can look at the screen and follow the flow without having to mouse over the various names in play. For that matter, I could print it out if I wanted to and still have all the information.
I don’t think you’re objectively wrong. It’s more that we have different approaches to managing the complexity when it gets hairy.