An "open source search" of the Internet involves a search for any documents in the public domain that pertain to matters of interest to intelligence analysts. In open-source searching for material deemed relevant to the work of intelligence analysts, a search technology should facilitate specific queries by accessing all related information and connecting the data into credible threat profiles. Search technology that accommodates a "semantic-based" search does this best. This article uses the three-layer cake analogy to explain how a semantic-based search is conducted. The first layer of the search would explain how objects found over the Internet relate to each other, i.e., terms synonymous with the search term would be automatically included in the search. The second layer of semantic-based search is taxonomy, which places objects in categories. This brings order to the millions of objects it retrieves and indexes over the Internet. Currently, semantic-based search technology has almost five million taxonomies, and the number continues to grow. The third layer of semantic-based search is ontology, which pertains to the meaning of what may appear to be random. This involves linking all the information retrieved in a query by connecting search terms in ways the searcher did not envision when beginning the search. This occurs as the search technology returns results that are related to the search term in a variety of ways, grouped in facets that are distinct and related information categories. This ensures more efficient and productive results. The article illustrates how a semantic-based search works with a query with the word "anthrax."