Review comments may contain decisions that are useful later. They may also contain internal discussion that should not appear in the public page.
What would you test before publishing real work? Would you keep a private review record connected to the finished article?