It works, but why would this make ANY difference, compared to just doing it like we've done it before ?

The best answer I can come up with, using the info in that link, is that static blocks are initialized once when the class is first loaded (not every time a new object is created) and initialized before anything else. Because static members are shared among every object of that class, putting the code to initialize static members wouldn't make sense anywhere else because you could re initialize something by accidently by over writing values.Lets say that you had a static count and initialized it to 0 in the constructor. You then create an object and use it to increase count a couple of times in your code (for whatever reason). Then you create a new object of the same type expecting it to have the same count, but the constructor reinitializes count to 0 and you lose value. This isn't the best example, but in more complex code you would want certain things initialized there. Here is some code you can run to see the what goes on:

public class Solution {
    static{
        System.out.println("*****Static block reached*******");
    }

    public static void main(String[] args) throws Exception {
        System.out.println("Creating new Solution 1");
        Solution sol1 = new Solution();
        System.out.println("Creating new Solution 2");
        Solution sol2 = new Solution();
        System.out.println("Creating new Solution 3");
        Solution sol3 = new Solution();
    }


    public Solution(){
        System.out.println("Constructor reached");
    }
}

If you initialize it in main instead wouln't that also run only once?