What do the numbers stand for in an ip address?

Answer 1

(I am assuming the question refers to IPv4 addressing. For IPv6 addressing the idea is the same, but the specifics of implementation differ.)

An IP address is a numerical representation of a unique computer in the relative "address space". What that means is - in the particular network using this address space, that specific address uniquely identifies a specific computer (as long as it has that address), or alternatively - a network-capable device other than a personal computer (a router or a network printer are two good examples).

Each time someone connects to a web page, for example, they are connecting to another computer somewhere elsewhere in the world - and that computer must have its IP address too.

An "address space" is a specific range of addresses "routed to" (and assigned) in a network segment (a distinct group of computers). Simply put, an "address space" is a bunch of addresses that can be assigned to computers within a network, and which would (when assigned) enable these computers to communicate with each other directly without the need to use special means (routing) to reach each other. The internet - often called a "network of networks" - assigns these addresses (in this case known as the "global pool") to subnets, so that they can "see" each other, but that is a yet another matter.

The four numbers separated by dots in an IPv4 address are called "octets", because each of the numbers can be represented by 8 bits. The lowest number recorded so is 0 (binary 00000000), the highest - 255 (binary 11111111). These four numbers together uniquely identify one specific "destination" (explanation of routing and/or address masking - or NAT - are beyond the scope of this answer).

It may be easier to understand this by an analogy: when you write down someone's address, it consists of a name, a street name and number, a city, and a country. So someone's address could easily be written as such:

Country . City . Street No . Name

The networks are, of course, not divided into such "zones" as Cities or Streets. Something like "USA . New York" is valid, but in this method "New York . USA" is not, since there is no country named "New York". Computer IP addresses are not "zoned" this way, so "128.255" is as valid as "255.128", just points somewhere else.

The values (0-255) also define how many computers are "addressable" in such way that they can "see" each other and communicate directly. Think "how many countries" by "how many cities" by "how many streets" and so on. So, if the first computer that can be addresses is 1.0.0.1, and the last 255.255.255.254 (taking into account the requirements of the protocol), there are possibly 254 * 255 * 255 * 255 - 1 addresses available in any given pool. That is some 4 211 669 249 unique addresses!

Of course, the global address pool is divided into sub-networks and "ranges" which are then assigned to countries to further assign (mostly sell) to their people. That makes the global pool somewhat "chopped up", and as a result, not all addresses end up being available. IPv6 is designed specifically with the availability problem in mind, extending the address' length so that potentially "every ant on the planet could have an IPv6 address".