bravemouse

To understand how computers get information over the internet, you first need to understand what happens when you open a document which is saved on your computer.

When you open a document using the Finder or Windows Explorer you are usually do it by clicking on little pictures of folders. But what is a folder and why does this process help your computer find documents?

As you know, a folder is a category into which you can sort the documents. In fact, a folder acts like a document's "address".   To find a document , we need to know what folder it has been saved in, what folder that folder is in, and again what folder that folder is in…this is called the "path" to that document.  The "path" starts with the place where all folders are saved: the hard drive of your computer. The hard drive is referred to by a letter – usually C:

Imagine we are looking for a document called letter.doc.  If you wanted to find a document that was saved directly on the hard drive and had not been put in a folder, its folder path would be written like this:

C:\letter.doc

Now, there might be any number of folders on the hard drive. And each of these folders can contain many folders, each of which contain many folders. If letter.doc was saved in a folder called "homework" which is in a folder called "school" which is in a folder called "My Documents" which is in a folder called "Documents and Settings" the folder path would look like this:

C:\ Documents and Settings\My Documents\school\homework\letter.doc

But what if you want to get a document from another computer via a network?

All computers on a network have their own individual addresses – this is called an IP address. Networked computers use the IP addresses to specify which computer a message is going to, and to check where it is coming from.

Finding a document over a computer network also requires you specify that document's path. This time, instead of the first letter in the path specifying the hard drive of your own computer it would specify the name of the computer that you want to contact over the network. Something like this:

Mandy's Computer\Documents and Settings\My Documents\school\homework\letter.doc

Of course, the name of the computer would not be "Mandy's Computer" but I'm using that for simplicities sake.

Back to HTML & CSS Day One

Tags: No Comments.

Before you can see a web page displayed on your computer's screen, it must first be copied from the machine on which it is saved on your own hard drive.  Retrieving a web page over the internet is essentially just like getting a document over a local network.  The basic steps are:

• You need to specify which computer, and where in the world, the page you want is saved.
• You need to specify where on that computer the page is saved – in which folder.
• The page must then be copied to your computer

But how does the message from your machine get to the computer on which the web page is saved?  If the internet is really thousands of computers all joined up together, how on earth can anything be found?

What are servers for?

A small network is relatively easy to understand. Ten or twenty computers can all be connected to one another without too much difficulty. But it would be impossible to have a direct connection between every computer linked to the Internet.  To cope with this complexity, there are two types of computers.  "Client Computers" which are your personal machine that you use at home, work or at your school.  "Servers" ( or "Routers") are a backbone of computers that communicate with one another, relaying messages along the shortest route.

The diagram above shows a very simplified view of the client computers and servers on the Internet.

1. When the client computer 'A' wants to see a web page that is saved on the server 'D', it cannot send a message directly to that server – the server might be in another city or country. Instead it sends a message to its local server – server 'B' in the diagram.
2. The server checks the message to see where in the world it is headed, and sends it on to the server closest to server 'D'.
3. That server does the same, and the message gets passed on until it reaches the correct server.
4. Server 'D' then sends a copy of the requested web page back, from server to server until it reaches the original client computer.

In the first step the client computer must specify which server the web page is saved on.  This is done by you typing the address of the web page in the web browser's "address" field.  A web site's address is a special kind of path called a Uniform Resource Locator or URL. The URL specifies where in the world the server is, and where on that server the document has been saved.

Now you might be thinking: when you type an address into a browser, you don't usually type in an IP address or a full path listing the folders in which a particular web page is saved. You usually type in something much shorter, like www.amazon.com or www.dogstuff.co.za/dogs.  So what's the deal?

This page is based in part on an article from howstuffworks.com – you can read the whole article at How Internet Infrastructure Works

Back to HTML & CSS Day One

Tags: No Comments.

If we wanted to retrieve a document saved on a server on the other side of the world, the address of that document can be quite long and complicated.
The address would specify in which country to find the computer that the document is on. Maybe which institution in that country. Which network in that institution. Which computer on that network. Which folder in which folder in which folder in which folder…… on the hard drive of that computer.

A long and complicated address.

Instead, we have worked out a system of "nicknames" called domains. When you store a website in a  folder on a server, you register that folder as a unique domain. Your domain might be called www.bestflowers.com. or www.shirts.co.za. The actual full address of your site is saved on a special server called a Domain Name Server (DNS) which keeps a list of all the domains and their corresponding addresses.

Next time somebody types in www.bestflowers.com into their browser's address field, a message is sent to the DNS to check what the real address is. The domain is matched to the address, and the request is sent off.

Back to HTML & CSS Day One

Tags: No Comments.