what is difference between cold start vs early rater problem in collaborative filtering-recommender system?

cold start-: system requires huge amt of current user data to make accurate predictions

early rater-: new user hasn't rated many items to make predictions.

both same? isn't it?