What was the ISC Domain Survey?
Internet Systems Consortium’s Internet Domain Survey, originally started by SRI International, is the longest-running survey of the number of computers connected to the Internet. This data provides an in-depth look at the active domains on the Internet. Despite its name, the ISC Domain Survey is not a survey of domains; it is a survey of IPv4 addresses that have a domain name attached to them. The name “Domain Survey” derives from an early algorithm used to locate addresses in use.
How many people use the Internet?
There is no way to determine how many users are on the net, besides making guesses and estimates. Other organizations and companies may do surveys to estimate the number of users on the Internet. You might find some useful information at The Netcraft Web Server Survey, The Center for Applied Internet Data Analysis, or The Internet Society.
How big is the Internet?
We consider the numbers presented in the Domain Survey to be fairly good estimates of the minimum size of the Internet. Obviously, we cannot tell if there are hosts or domains we could not locate. Whatever those numbers measure, they measure the same thing every time, so it is meaningful to make comparisons to determine growth.
It is not possible to determine the exact size of the Internet, where hosts are located, or how many users there are.
How do I figure out the breakdown of hosts per country from your data?
You can’t. There is not necessarily any correlation between a host’s domain name and where it is actually located. A host with a .NL domain name could easily be located in the US or any other country. In addition, hosts under domains .edu/.org/.net/.com/.int could be located anywhere. There is no way to determine where a host is without asking its administrator. Many countries sell names in their national domain as a source of revenue; for example, the tiny island nation of Tuvalu (population 11,000) has some 80,000 names defined, because companies are willing to pay good money for a “.TV” domain name. None of them are located in Tuvalu.
How did you compute the adjusted host count figures?
We start with the figure we published for each old domain survey (on the distribution by domain name charts), in the column named “Percent Domains Missed.” If for example it says “20%” domains missed, this means we got 80% of the domains, or 0.80. We then take the total host count and divide it by this number. For example if the hostcount was 800 and we missed 20% of the domains, we would divide 800 by 0.80 to arrive at an adjusted host count of 1000.
What is a host?
A host used to be a single machine on the net. However, the definition of a host has changed in recent years due to virtual hosting, where a single machine acts like multiple systems (and has multiple domain names and IP addresses). Ideally, a virtual host will act and look exactly like a regular host, so we count them equally.
What is the relation between a host and a network number?
There is not necessarily any correlation between a network number and a domain name. A single network number could span many countries, and a single domain may have hosts on multiple network numbers.
Can I have permission to reproduce your data or charts?
You have permission to reproduce our data provided that you mention the source as “Source: Internet Systems Consortium, Inc. (https://www.isc.org/ ).” For derivative works based on our data, you must say your data or charts are “Based on data from Internet Systems Consortium, Inc. (https://www.isc.org/ ).”
Are the raw data available?
We have some of the raw data available. The data for each survey occupies several gigabytes. In the early days of the Domain Survey it was very expensive to store that much data. Contact firstname.lastname@example.org to see if we might have what you want. We make no promises.
What terms are used in the Domain Survey?
These are definitions of terms used by the Internet Domain Survey. Not everyone uses these terms with the same meanings.
domain name: A domain name is any name representing any record that exists within the Domain Name System, i.e. com, example.com, www.example.com.
domain: A domain is a domain name that has name server (NS) records associated with it. In other words, there may be subdomains or hosts under it, i.e. com, example.com.
domain server: A domain server is a computer system that holds all the records associated with a particular domain, and answers queries about those names. Often called a domain name server or just a name server.
domain (top-level): A top-level domain name, or TLD, is either an ISO country code or one of the generic domains (com/org/net, etc.). It should be noted that there is not necessarily any correlation between a country code and where a host is actually located. Domains named after ISO country codes are often referred to as “ccTLDs”, and generic domains are often referred to as “gTLDs”.
duplicate host: A duplicate host is a single host name that maps to multiple IP addresses. When a host like this is found in the survey, it is only counted once.
host: A host is a domain name that has an IP address (A) record associated with it. This would be any computer system connected to the Internet (via full or part-time, direct or dialup connections), i.e. example.com, www.example.com.
host name (or firstname): A host name is the first part (before the first dot) of a host’s domain name, i.e. www.
network numbers (class a/b/c): A network number is the first part of an IP address, which identifies what network the hosts numbered in the rest of the address are connected to. The class a/b/c system is no longer used and the data are presented that way only for historical reasons. It is no longer possible to tell how many network numbers are in use by looking at IP addresses, since the netmask can not be determined from the number. Also, note that domains do not map directly to particular network numbers and no correlation between them can be inferred.
zone transfer: A zone transfer is the process of downloading all the records associated with part of a domain from a domain server.