The domain name system (DNS) represents the “telephone book” of the internet. Similar to using the telephone book to look up a name in order to find a phone number, you look up addresses in DNS to find an IP address.
We all know that a telephone call is not conducted through the telephone book itself. In the same way, Internet traffic does not pass through DNS.
IP addresses and domain names are affiliated
It is the services provided on the technical infrastructure that represent value to internet users. The best-known services are websites and email, but it is also possible to use the internet for activities such as connecting telephone conversations, downloading files and logging on to various databases.
So how do we get access to these services? All computers linked to the internet have their own IP address, which consists of a long series of numbers. Using this address makes it possible to connect directly to a computer. However, to save users from having to remember long strings of numbers, DNS attaches a unique domain name to the IP address. When you look up a domain name, a search for an IP address is initiated in order to contact the server hosting the service you are looking for.
Domain names may or may not offer services
Norwegian domain names are commonly written in the form mydomainname.no, where mydomainname is the domain name you choose. If the domain name offers services such as websites and email, the web address may be www.mydomainname.no, and a typical email address within the domain may be written in the form email@example.com or firstname.lastname@example.org.
It is important to note that it is possible to register and hold a domain without it offering services to users. It is required that the domain name has to be set up technically, but it does not need to have a “content” in the meaning web pages or email addresses.
A domain name is always unique. regjeringen.no and regjeringa.no (Norwegian government websites) represent two different domains because they are spelled differently. This rule applies even if the two names share the same meaning and the same domain holder, and point to the same IP address.
A global hierarchy
The domain name system is structured as a hierarchy and can be compared with a plant’s root system. The uppermost level is often called the DNS root zone, or just the “root”. The so-called “top level domains” represent the uppermost level just below the root. There are two types of top level domains. The first are the country codes, such as .no (representing Norway) and .se (Sweden), and regulations governing these are drawn up at the national level. The second are the generic top level domains such as .com, .org and .net, and .statoil, .cloud and .paris – in pricipal .whatever. Regulations governing these domains are stipulated at the global level.
Below the top level domains we find what we most commonly associate with the term “domain names”, (second level domains) such as uio.no for the University of Oslo. Some top level domains also have their own so-called “second level category domains” at this level. These domains are set up for specific groups, such as dep.no, which is a category encompassing
all the Norwegian government ministries, or kommune.no, which is a category for the counties in Norway. At the next level down we find the so-called “sub-domains” such as minestudier.uio.no, and the names of computers such as stream-prod02.uio.no.
Responsibility applies per “rootlet”
The root structure also reflects the levels of responsibility. Separate organizations have responsibility for “rootlets” at respective levels.
A variety of registries administer the regulations and operate the central database for each individual top level domain. Norid assumes this role for the Norwegian country code .no. A user who has registered a Norwegian domain name (the holder) is entitled to create whatever sub domains he wants. Rights and obligations pursuant to the regulations governing the Norwegian top level domain rest at all times with the holder, regardless of whether he has given a third party permission to establish a sub domain.
A person or an organization can only influence the rootlet immediately below its own level in the hierarchy. This means that Norid cannot influence a domain such as .com or domain names registered under this top level domain.
What happens when a domain name is looked up?
Each domain name is linked to a series of computers which respond to enquiries regarding addresses registered within the domain name, for example the IP address for www.norid.no. These computers are called name servers. For the most part, the user is unaware of communication with these computers.
An example of a typical enquiry:
You want to look up a specific event posted on the University of Oslo’s website. You know that the university’s address is www.uio.no, so you enter this in the address field of your browser.
- A small software application in your computer contacts a separate computer – a so-called “recursive resolver” – which has been set up to process enquiries made to the domain name system. This computer is usually located on the premises of the Internet Service Provider (ISP).
- The job of the recursive resolver is to find the IP address for www.uio.no. It sends the enquiry onwards to one of the name servers for the root in the domain name system. The root name servers recognize only the level below them in the hierarchy and so send back a list of the name servers for .no.
- The resolver then re-sends the enquiry to one of the name servers for .no. These also recognize only the level immediately below them, and so send back a list of the name servers for uio.no.
- The resolver repeats the enquiry to one of the name servers for uio.no, which responds with the IP address for www.uio.no.
- The resolver then sends the IP address to your computer. When your browser receives the address, it can then contact the university’s web server and download the website you want.