Uniform Resource Locators (URLs) are a subset of Uniform Resource Identifiers (URIs) that consist of two components that identify how to access a resource on a TCP/IP internetwork. These two components are the location of the resource, and the method to be used to access it. These two pieces of information, taken together, allow a user with the appropriate software to obtain, read or otherwise work with many different kinds of resources such as files, objects, programs and much more.

The most general form of syntax for a URL contains only two elements, which correspond to the two pieces of information just described:

<scheme>:<scheme-specific-part>

The term scheme refers to a type of access method, which describes the way that the resource is to be used; it usually refers to either an application protocol, such as “http” or “ftp”, or a resource type such as “file”. A scheme name must contain only letters, plus signs (“+”), periods (“.”) and hyphens (“-”). In practice, they usually contain only letters. Schemes are case-insensitive but usually expressed in lower case.

The rest of the URL after the scheme (and the required colon separator) is scheme-specific. This is necessary because various protocols and access methods require different types and quantities of information to identify a particular resource. When a URL is read, the scheme name tells the program parsing it how to interpret the syntax of the rest of the URL.