IP Addressing:


It's a struggle to keep this simple, so I'll probably start out that way and then make it as complicated as it can be (just kidding - I think...).

In order for PC's to communicate using IP addresses, they have to be on the same range of network addresses. A given range of addresses (called a network or subnetwork) is determined by a combination of two things:
  • IP Address
  • Subnet Mask
To start out, IP addresses consist of 4 groups of numbers separated by dots (e.g. 192.168.1.100). These are actually decimal representations of binary numbers. Each of these is referred to as an OCTET, because in binary form they each consist of eight digits.

Furthermore, part of this address represents a "network" address (which all machines on the given network will share in common), and the rest of it represents the unique PC address.

I know this is confusing if you don't know binary, but bear with me - I'm just trying to communicate the geek language to you so things will make sense as we go along...


BINARY

You're probably well aware that computers use a digital language, and generally that language is binary. In binary, you only have 2 numbers. Actually, you don't really have numbers at all, though they're represented by 1 (one) and 0 (zero). It will make more sense for you to think of it as two states (i.e. on/off or even checked/unchecked).

When you count in binary, you count exponentially. If you were to list a group of 1's like this "11111111" each digit would have the following decimal values (counting from right to left no less!):  "1,2,4,8,16,32,64,128."
Once you've digested this, read on ...

You would then add all the values together to get the decimal (NORMAL/UNDERSTANDABLE/HUMAN LANGUAGE) equivalent: 255.

If, on the other hand, one of these binary "1's" was a zero, you'd simply subtract it's value from the total as in the below examples:
  • binary "11111110" = decimal "128,64,32,16,8,4,2,0" for a decimal total of 254.
  • binary "11111101" = decimal "128,64,32,16,8,4,0,1" for a decimal total of 253
  • binary "11111011" = decimal "128,64,32,16,8,0,2,1" for a decimal total of 251
  • binary "10111101" = decimal "128,0,32,16,8,4,0,1" for a decimal total of 189
  • binary "00111101" = decimal "0,0,32,16,8,4,0,1" for a decimal total of 61
You might have to study that for a minute to grasp it. It's important to understand this if you need to manage your network schemes. Mostly, it comes into play when you combine it with your subnet mask.

However, you might not need all this. In fact, most people probably don't. I said earlier that part of the address represented the network, and part of it was the unique PC address. The subnet mask tells you which is which. Consider the below address and mask:

ADDRESS:
192.168.1.28


SUBNET MASK:
255.255.255.0



If you were to line up the bits on these one above the other, you'd have:

11000000.10101000.00000001.00011100  (IP address)
11111111.11111111.11111111.00000000  (Subnet mask)
Imagine that the subnet mask is printed on a transparency, and you can lay it directly on top of the IP address. All the "1's" of the subnet mask correspond to (or "mask out") the network address (in yellow). All the remaining bits of the IP address equate to the PC's unique ID (in gray).

Let me simplify.  There are three classes of IP address schemes that are common:
  • Class A: has a subnet mask of 255.0.0.0
  • Class B: has a subnet mask of 255.255.0.0
  • Class C: has a subnet mask of 255.255.255.0
If you aren't concerned with internet access, you can configure your network anyway you want using one of these classes.  But my guess is that you probably are interested in accessing the Internet.  So here's how you deal with your Internet Service Provider (ISP) ...

next...