On the Web, documents are identified using Uniform Resource Locators, or URLs [62]. The fields of a URL describe the protocol, called scheme in URL terminology, used to retrieve the document, in combination with a protocol specific part. For most protocols, the specific part denote the host on which the document may be found, an optional network port to connect to, and a path identifying the document.
What follows is a couple of sample URLs: The above URL indicates a server on the host www.whitehouse.gov, listening on port 80, and communicating using HTTP. The document is further identified with the path WH/Welcome.html, indicating that this is a hypertext document. According to [32], port 80 is the default for HTTP, so the :80 could have been omitted. This URL identifies the directory pub/Linux/ on the server sunsite.unc.edu, accessed using FTP [33]. URLs need not necessarily reference existing documents, as illustrated by the above example. The mailto-scheme specifies an electronic mail address. Most browsers will pop up a window in which the user may compose an E-mail to the address given.