Today the Internet is omni-present. The unprecedented exponential growth of the “networks of networks” is the result of a vision of limitless connectivity originating from a small group of technology enthusiasts in California combined with a system of ingeniously simple communication protocols.
What is the Internet? — Four Subsystems:
Interconnection and Traffic
Data travels through the communication links of telecom operators and internet service providers. The traffic routing system, using the Border Getaway Protocol (BGP), finds a path from users to a data source. When a connection fails, the system is designed to self-reconfigure and find an alternative path. Today’s users’ often see only “last mile” wireless networks, but these are only the entry points to this vast and complex system.
For data to travel from one place to another there has to be addresses. Every computer, every point on the Internet has its own numerical identifier, its Internet Protocol (IP) address. Because the Internet is a global network, all addresses have to be globally unique.
Numerical IP addresses proved not to be convenient for humans to use, so a system of alphanumerical addresses was created. These identifiers are part of the Domain Name System (DNS) — the address book of the Internet – and they also have to be globally unique.
Services and Hosting
Services can be system services (the “wiring” of the Internet — DNS, BGP4, …), or user services such as e-Mail or Web Wide Web. The data that internet users access is located on computers called servers or hosts. Web sites use server space and resources leased from hosting providers to enable users to access that content.
Content and Platforms
Content is why we use the Internet – from simple Web sites, through various chat apps, to complex platforms like Facebook and Tik-Tok. From a technical point of view, social networks and other platforms are just services that offer content – they are something existing over the top of the Internet (OTT). The internet provides the access and transport network for these services.
The flexibility of Internet architecture enables unlimited creativity, thus there is always something new. However, for a system as complex as the Internet to function, there has to be a standard way to communicate – that’s why we have internet protocols.
The fundamental set of protocols is known as TCP/IP. It was developed in the 1970s with the aim of interconnecting different networks. The use of standard internet protocols is free, and there are no limitations. This liberty has led to numerous innovations and the rapid development of the Internet. Users’ need for content and services has driven the interoperability imperative – the requirement that systems be compatible with standards.
Internet standards are developed mainly by the Internet Society (ISOC) and its Internet Engineering Task Force (IETF), through documents called Requests for Information (RFC – rfc-editor.org) The motto of the IETF: “We reject kings, presidents and voting. We believe in rough consensus and running code” reflects the liberal standardisation process adopted by the IETF.
Internet protocols were envisioned to interconnect freely, so initially, designers didn’t consider control, oversight, different traffic categories or distance-related charges. These protocols created the Internet as a globally connected network, but did not guarantee access speeds, service quality, security and protection from abuse. Modern standards development focuses on security and user privacy and protection, primarily using cryptography.
Internet identifiers must be globally unique. The Domain Names System is hierarchical with levels that are separated by a dot. In a domain name, (for example icann.org), the first part “icann” is a second-level label, and “org” is a top-level label (TLD). Domain names can be country-code or generic. Country-code TLDs are traditionally two-letter codes (assigned to countries and territories on the ISO 3166-1 alpha-2 list). There are more than 1500 top-level domain names, some of them in non-English alphabets. A domain can be an internet address, or a more complex address and can have a service or host name as a prefix, such as “www” in www.icann.org.
For a user’s computer to map a domain address into a numerical IP address and access the server that has the desired content, there has to be a list of top-level domain names. This list is at the top of the DNS system hierarchy, the Root Zone. The IP addresses of DNS servers containing the Root Zone data are publicly known. A DNS query for .org (for example) first returns IP addresses of DNS servers for the .org top-level domain zone and then the IP address for the server where the icann.org content is located.
The Internet’s Main Address
The Internet data traffic system is de-centralised and designed to dynamically re-route connections that have failed, so no particular part is critical. The DNS system is the only centralised internet system, and it is essential for the functioning of the whole Internet. The Internet’s continuous availability is made possible by the Root Server System (RSS – root-servers.org) that contains the addresses of the DNS servers for internet top-level domains. The Global RSS consists of 13 independent systems of root servers, managed by different Root Server Operators. There are more than 1000 root servers of various size, answering millions of queries every second, and providing free service to internet users globally.
IANA and ICANN – Who governs?
The assignment of top-level domains began with John Postel, one of the pioneers of the Internet. While working at the Information Sciences Institute at the University of Southern California (ISI-USC) he started the Internet Assigned Numbers Authority (IANA – www.iana.org) to govern the Internet’s unique identifiers.
IANA manages the DNS system root zone, coordinates the global IP address system and the routing Autonomous System (AS) numbers and is the central repository for Internet protocol names and numbers.
The Internet Corporation for Assigned Names and Numbers (ICANN – icann.org) was formed in 1998 to take over the IANA. ICANN is the central organisation in global Technical Internet Governance and is recognised as the global steward of unique internet identifiers. ICANN operates in the global public interest through a system of contracts, and, most importantly, through acceptance by internet users worldwide.
The structure of ICANN and the way decisions are reached are as unique as the Internet itself. ICANN policies are created by the ICANN multi-stakeholder community through a bottom-up method. The Policy Development Process (PDP) includes interested stakeholders, such as domain registries and registrars, business and non-commercial organisations, internet users, non-governmental organisations and government representatives (currently from 178 governments and inter-governmental organisations).
Registries of generic top-level domains (gTLD) are obliged by a contract to abide by all ICANN policies, resulting from a consensus, bottom-up Policy Development Process. Registries of country-code domains (ccTLD) are free to create their own policies while observing RFC standards and best practices.
Why are the addresses important?
Many equate Facebook with the Internet, but Facebook is just a website, one of many platforms that function on top of the Internet. When we use Facebook, it’s easy to forget the complexities of the Internet and all the changes that the Internet has brought to our personal and working lives. When we are connected, the whole global Internet is within our reach. The hundreds of millions of locations we can browse, what content we will see and what services we will use is made possible by the system of internet identifiers.
Domain names are actually just alphanumeric labels, but they can be couched as terms and names with meaning, which makes them a valuable, easy-to-use resource. Some domain names have been sold for millions of dollars. This raises questions of registration priority, trademarks, intellectual property protection and possible abuse. The importance of domain names crosses state lines, cultural boundaries and laws and regulatory regimes. For the future of the Internet it is imperative that the system of unique internet identifiers remains globally unified, otherwise fragmented Internet might stop being the driving force of development.
Discussions around the governance of the Internet are more critical and dynamic than ever. United Nations Internet Governance Forums (IGF – intgovforum.org) and other bodies are searching for answers to questions on privacy, personal data protection, freedom of expression, intellectual property, surveillance and legal oversight, abuse prevention, security, stability and resilience of the Internet and the relationship between the global Internet society versus local laws and customs.