On the recent proof of the Willmore problem (I)