Introduction

A fixed number of data elements that have the same type can be conveniently stored in an array.  However, if you have a variable number of data elements, then an array can be inconvenient.  Imagine the difference between a text field which might allow a fixed number of characters, and how this text field might be used to develop a word processor.  How many characters should your word processor allow?  If you don't set the figure high enough, your word processor will not be able to handle large documents.  If you set the figure too high, you will waste a large amount of disk space to store small documents.  Word processors need a variable amount of storage.

When using a word processor, you should also notice that new characters are not always added at the end -- they are inserted after the cursor.  The cursor is like a "current pointer" into a (doubly) linked list of characters.  The cursor can be moved forwards and backwards along the list, and insertion and deletion of characters can only be done at the cursor.  Notice that unlike arrays, linked lists do not have "random access" -- you have sequential access, so you cannot access a random element directly.  Thus, to understand linked lists, you have to think in the perspective and limitations of your computer -- you have to learn to think like a computer.
 

Resources

• To see the functions of a linked list, try this example.  Notice that the head pointer is not shown, and that the current pointer is shown in green.  The commands "Insert First" and "Delete First" use the head pointer, and the commands "Insert" and "Delete" use the current pointer.  The current pointer can only move forwards in the linked list (using "Next") or return to the front (using "First").
• Now that you've seen what a linked list can do, you might want to try this code-based simulation of a linked list to see how it is done.  This example graphically animates the functions of a linked list and steps you through the underlying code at the same time.  Although rather slow, it can help you see exactly how pointers work.
• To see things happen at a faster pace, you can also look at these demos for a singly linked list and a doubly linked list.  With data structures, it is generally more important to understand the required algorithms than it is to understand the underlying code.  When you follow these examples, try to imagine the small code fragments that it would take to implement each step -- you should see that if you know what you want to do, it is relatively easy then to program it.