I have to make a project for my c++ course, they let us choose, and I made (am making) a mahjongg-game. Its comming along nicely, all I need is some graphics, and a little code concernig the highscore, nothing difficult. BUT I made it as folows : I have a global object called GAME, this contains other objects called HIHSCORE, MENU, BOARD. Depending on the state of the game (playing, in menu,..) GAME calls the correct function from the other objects (like draw()....) Is this so wrong?? He now wants us to us paradigm or something like that to make our classes, and insists that we completely seperate the users interface from the game. I understand what he means but I don''t think it is neccesary to do this here. I would just like to here your comments on this, either convince me my teacher is right, or give me some arguments against him please. Best part is that he is not my c++-teacher, just some guy who thinks he knows it all and who gave a really crappy OOP-course in the first semester Any comments are welcome

I don''t think you''re wrong, if your code looks nice, it''s a good way to go, if the project is small. But, you''ll have to ask yourself if you''re able to convert the game into oop. It might be good to test oop on such a small project, instead of using it "when you think it''s neccesary".
I say you should do what he thinks. It''s good practice.

A polar bear is a rectangular bear after a coordinate transform.

Yeah but I''m using OOP right?
just not the way he sees it.