From the beginning of the twentieth century on, the use of the World Wide Web has become a current theme among corpus linguists. As the Web has grown both in size and in popularity many linguists have look at it as a potential source where they can carry out linguistic researches. The Web has quickly become the playground of corpus linguists. The Web allows corpus linguists to use hundreds of millions of text and billions of running words for conducting linguistic analyses. However, the Web as corpus is a very different kind of corpus in many respects, For instance, we do not know its precise size and that what sorts of texts it contains. The aim of this paper is to summarize those possibilities the Web can offer as a linguistic corpus.