Mining for hidden text in published data is analyzed. It focuses on automating and scaling the hidden text effect. The antiword does not reveal usernames, pathnames, and other metadata and rarely finds finds hiddentext. The string-matching method returns deleted words or lines containing a deleted or altered word, word-matching yields some information about Word documents internal structure and deleted text.