Why does the java compiler not shorten the name by default? (performance and confusion)

I can't understand why the java compiler doesn't shorten the names of variables, parameters and method names by replacing them with some unique IDs

For example, a given course

public class VeryVeryVeryVeryVeryLongClass {

    private int veryVeryVeryVeryVeryLongInt = 3;

    public void veryVeryVeryVeryVeryLongMethod(int veryVeryVeryVeryVeryLongParamName) {
        this.veryVeryVeryVeryVeryLongInt = veryVeryVeryVeryVeryLongParamName;
    }
}

The compiled file contains all these very long names:

A simple unique ID won't speed up parsing. Will it provide the first confusion?

Solution

You assume that confusion is always needed, but it is not:

>Reflection breaks, JavaBeans and many frameworks rely on it > stack traces become completely unreadable > if you try to encode against a compiled jar, you end up with code like string name = p.a1() instead of string name = p.getname()

Confusion is usually the last step when you deliver a completed application, even so, it will not be used often unless the target platform has severe memory constraints