HTTP

HTTP stands for the HyperText Transfer Protocol. It is the protocol used for browsing web sites.
HTTP is a stateless protocol, which means each request between a client (such as a web browser) and a server is independent of all others. This is in direct contrast to FTP, SSH or Telnet, which are connection-oriented. The stateless nature of HTTP makes it an ideal platform for building massively scalable applications since all you need to add capacity (in theory) is to add more web servers.
Application state (eg. a shopping cart) is built on HTTP by means of some tracking mechanism that allows the application to identify a client's recent activity. Popular tracking mechanisms include cookies, URL "mangling" and invisible form elements.
This article assumes the reader is familiar with related concepts like DNS, URIs, and TCP/IP. Also, the term "client" will refer to any program or script that communicates using the HTTP protocol, such as a web browser, the curl program or a Perl script using the LWP library. The term "application" will refer to programs that run on top of clients and may not use direct HTTP communications, such as SquirrelMail, Gmail or Facebook applications.

HTTP version

There are several versions of HTTP, the most popular being 1.1.

HTTP Request/Response

HTTP follows the request->response model. The request and response have a few required elements (all others being optional.) For purposes of demonstration, all examples will assume HTTP version 1.1.
A minimal request will include the relative URI, the HTTP protocol version being used and the Host header.

Headers

Headers are an element in the HTTP protocol that allows clients and servers to exchange information about the request and response. They do not contain