FYI I don't believe you are supposed to ask discussion-based questions like this on SO... But the simplistic answer is this: don't limit your thinking to the logic of the code. We all know there are ten thousand ways to accomplish the same thing. You can probably rewrite a bunch of your code to avoid encapsulation. However, data encapsulation provides a few benefits when you start working on larger projects or with larger teams that go beyond just writing functional code:
(1) organization by concept: if you're coding a bike you would code a class for the wheel, a class for the frame, a class for the handlebars, etc., and you'd know where to go to resolve an issue, perhaps even after months of time away from the code;
(2) separation of implementation and interface: you can tell the world about your public interface and handle the actual implementation privately, so people don't have to know how things work in your code, they just know that it works, like a black box; and if later you have to change your private implementation you can do so freely as long as the public interface still works;
(3) simplification for humans: remember, humans read your code, so would you like to slam them with every bit of data and logic in your project? that would just make for a bunch of angry programmers.
So that's a gentle introduction to encapsulation.
