of real-time video to set top boxes sometimes known as IPTV. There is another approach known as application layer multicast which uses peer-to-peer technology, which does not require any changes at the IP layer. This will be discussed in Chapter 15.

1.3 Internet Names

Internet addresses, covered in Chapter 2, are used to route individual datagrams over the Internet. However, they are not very friendly for humans to use. IPv4 addresses can be as long as 12 digits while IPv6 addresses can be as long as 32 hexadecimal digits long. A given Internet host with only one IP address may have many identities. Also, some Internet identities are that of the human user, not the actual host. For example, an e-mail address identifies a user, not a particular host on the Internet. The user may utilize multiple Internet hosts to access e-mail.

Internet names began with RFC 822 [13] which defined the user@host format that is so familiar today with email addresses. These text based names were defined to enable a piece of software known as a parser to be able to extract the various parts of the address and any parameters. In addition, some of the first Internet applications such as e-mail used a text-based way of encoding protocol messages. The method of encoding both Internet names and messages was defined in RFC 822 as Backus Naur Format (BNF). BNF was developed by John Backus to define the early programming language ALGOL. Today, many Internet protocols, including SIP, are defined using Augmented Backus Naur Form (ABNF) [14] which is based on BNF. The appendix has an introduction to ABNF.

1.4 URLs, URIs, and URNs

Uniform Resource Locators (URLs) are an addressing scheme developed for the World Wide Web (WWW). It is defined in RFC 1738 [15], and is a syntax for representing a resource available on the Internet. The general form is:

scheme:scheme-specific-part

for example, consider:

http://www.artechhouse.com/Default.aspx

The token http identifies the scheme or protocol to be used, in this case HTTP. The specifier follows the “:” and contains a domain name (www. artechhouse.com), which can be resolved into an IP address and a file name

(/Default.aspx). URLs can also contain additional parameters or qualifiers relating to transport, but they can never contain spaces. For example, telnet://host. company.com:24 indicates that the Telnet Protocol should be used to access host. company.com using port 24. New schemes for URLs for new protocols are easily constructed, and dozens have been defined, such as mailto, tel, and https. The sip and sips schemes will be introduced in Section 4.2.

Most protocols reference URLs, but with SIP we mainly reference Uniform Resource Indicators (URIs). This is due to the mobility aspects of SIP, which means that a particular address (URI) is not tied to a single physical device but instead is a logical entity that may move around and change its location in the Internet. However, the terms URL and URI are often used almost interchangeably in other contexts.

Some other examples include:

http://www.ese.wustl.edu

sip:barney@fwd.rubble.com

mailto:help@example.com?Subject=Help!

The initial set of URL schemes defined in RFC 1738 are in Table 1.1. Common SIP and Internet Communications URL and URI schemes are

listed in Table 1.2. The details of SIP and SIPS URIs are covered in Section 4.2.

Uniform Resource Names or URNs are defined by [16]. A URN provides a standard name for a resource but does not provide any information for how to access the resource. An example URN namespace is book ISBNs (International Standard Book Numbers):