In recent years, the blog has become the most typical social media for citizens to share their opinions. In addition, a large number of blogs reflect current social trends or major issues. Especially, more than thousand articles and more than 10,000 responding messages (comments) are registered on a well-known blog in a day. It is hard to search and explore useful messages on blogs since most blog systems show articles and their comments in a form of a sequential list. Also, there can be many unrelated comments, such as ad messages, and these spam comments hinder the user from locating the helpful comments. To overcome these shortcomings, we have designed and implemented TRIB (Telescope for Responding comments for Internet Blogs) for visualizing blog articles including the replying comments for them. TRIB considers the semantic weight between the subject article and corresponding comments using user-defined dictionaries, which provide various personalized views for a large set of comments on blogs or Internet discussion forums. To show the usefulness of TRIB, we conducted some experiments with articles that have more than 1,000 comments.