As mentioned above, the browser is responsible for fetching and displaying documents from servers, as requested by users. Most browsers have built-in support for communicating with various types of information providers in addition to HTTP-talking Web servers. The built-in support typically includes FTP (general file transfer [33]), NNTP (news [63]), SMTP (mail [34]) and Gopher (a distributed document search and retrieval protocol [64]).
Before displaying a document, the browser needs to decide the contents, to choose an appropriate way of viewing. Documents are typically categorized using MIME-like descriptors. MIME [51] is an Internet standard for identifying the contents of mail, using a classification scheme built on types and subtypes. As an example, an MPEG video stream would be classified as video/mpeg.
When transferring a document using HTTP, the HTTP header may, depending on the server, include a Content-Type field giving the type and subtype of the document in question. If no Content-Type field is present, or if another protocol than HTTP is used to transfer the document, the browser will have to do a qualified guess on the contents. This guessing is typically done by examining parts of the URL used to reference the document, in particular by looking at the extension of the filename part of the URL. Browsers may be configured to associate certain extensions with MIME-types.
Once the type of the document is identified, the browser decides how to display the contents. Widely used formats, such as HTML, plain text, GIF and JPEG, are normally handled by the browser itself. Other formats may be sent to external applications or plug-ins, as described below. Handling of initially unsupported MIME-types, may normally be specified by the user.