|< Day Day Up >|| |
What problem precipitated this constitutional moment? What great issue animated these global negotiations? The object of the controversy was control of a seemingly obscure set of technical functions related to naming and addressing computers on the Internet. Data communication on the Internet takes place by breaking messages into smaller units called packets and routing them from network to network. In order to know where to go, each packet must carry a numerical address, known as an Internet Protocol (IP) address. Every computer connected to the Internet must have a unique IP address. To supplement these numerical addresses, the computers, routers, and other resources connected to the network can be given user-friendly names like www.yahoo.com, known as domain names.
Many vital activities on the Internet, such as email or the World Wide Web, use domain names rather than IP numbers as addresses. But for packets to flow across the network, the user-friendly names must be translated into IP addresses. Both kinds of addresses-domain names and IP numbers-are valuable resources, a kind of virtual real estate that can be bought and sold.
It was name and address management that created the controversies that led to the IFWP. The specific set of functions at issue can be summarized as
The authority to set policy for and to manage the allocation and assignment of Internet Protocol addresses
The authority to add new names to the top level of the Internet domain name hierarchy
The responsibility for operating root servers that distribute authoritative information about the content of the top level of the domain name space
These functions are defined more precisely and discussed in greater detail in chapters 2 and 3. Although they may sound uninteresting, they are the technical underpinnings of what the Internet is all about. We tend to speak of the Internet as if it were a thing, but in reality the Internet is entirely virtual; it consists of nothing but a software protocol suite known as TCP/IP. [5 ]The software enables any computer in the world to exchange information with any other computer, regardless of the particular physical networks to which they are attached or the hardware they use. It does this largely by giving computers addresses and names, and providing instructions about how to use them. Consistent and scalable naming and addressing protocols are at the core of TCP/IP's design. The functions enumerated previously are needed to ensure that the names and addresses will be unique. Throughout this book, I refer to that cluster of functions as 'the root.' [6 ]
The root is the point of centralization in the Internet's otherwise thoroughly decentralized architecture. The root stands at the top of the hierarchical distribution of responsibility that makes the Internet work. It is the beginning point in a long chain of contracts and cooperation governing how Internet service providers and end users acquire and utilize the addresses and names that make it possible for data packets to find their destinations.
Addresses and names must be globally unique. Ensuring uniqueness in an open, rapidly growing network with millions of users is a coordination problem of some magnitude. The root is the Internet's answer to the problem of coordinating unique identifiers.
The security and stability of the root server system is critical to the viability of any service or function that relies on the Internet. No one disputes the operational significance of the root, and hence no one disputes the need for the formation of permanent, stable organizational arrangements to control-to govern-those functions. But the word governance has wider implications.
[5 ]TCP stands for Transport Control Protocol; IP stands for Internet Protocol. Both work together to guide the movement of packets across networks.
[6 ]Internet purists may object to this label. Usually, root refers only to the unnamed space at the top of the domain name hierarchy and does not include the IP address space. I have chosen to use root as a generic term that applies to both, for three reasons. First, even though the IP address and domain name spaces are distinct entities technically, there are structural similarities that are important institutionally and economically. IP address blocks are delegated in a hierarchical fashion just as domain names are, and the question of who controls the initial delegation- the top of the hierarchy-poses many of the same institutional issues in either case. Second, domain name root management and IP address management are technically interrelated in important ways (such as in the in-addr.arpa domain). Although this does not necessarily mean that responsibility for both should be combined in the same organization, historically they have been, and ICANN continues this practice. Indeed, the White Paper explicitly rejected appeals from some members of the technical community to place the two functions in separate organizations (NTIA 1998b, 31744). So in that respect it makes sense to speak of a generalized root that embraces both. Third, if the two previous points are valid, for reasons of readability and style it make sense to use a single word to refer to both in many instances. When the term is used in that fashion, it means the centrally coordinated naming and addressing functions required to ensure universal connectivity on the Internet. When there is cause to discuss the root of the domain name system specifically, I refer to the 'DNS root' or the 'name space.' When there is a need to refer exclusively to IP numbering, I refer to 'IP address space' or 'IP number allocation.'
|< Day Day Up >|| |