Information overload is considered one of the fundamental human-computer interaction problems today. The Internet provides access to millions of documents. Closer to home, average desktop users may have thousands of files and documents stored on their systems. Users may deal with budgetary information distributed amongst hundreds of categories or access databases of enormous size. The computer can, in a few seconds, retrieve information that humans would take years to digest. Often, much of this information is irrelevant. In an effort to cope with this problem, more and more users are turning to information visualization. This paper surveys the area of Information Visualization. It discusses visualization designs and techniques from a prospective based on Shneiderman's abstract user-task taxonomy for information visualization. It then presents some design guidelines that should be considered when  designing information seeking applications based on visualization techniques. It also observes that little scientific study has been performed on the usability of information visualization and calls for study in realistic work situations.