Thank you for two very relevant questions! We will develop an algorithm that will compare "debate topics" by analyzing their "initiators", text content, and "evidences". This algorithm will assign a correlation factor to all possible one-to-one comparisons from domain of "debate topics". Any two "debate topics" with factor exceeding a threshold may be integrated together.

Required minimum word counts for all responses, post flagging system at user-end, and algorithms searching for abusive language will all be used to prevent all-out brawls from breaking out.

These are among important parameters that will be under observation starting in the development stage. We will learn from our experience and improve our algorithms.