| How does the Internet work? | How does the World Wide Web work? | What science made the Internet possible? | What science made the World Wide Web possible? | The future |

The Internet functions a bit like an ultra fast mail service. When information is transferred along the wires of this electronic superhighway, it must have an address to go to, a return address, and a message. Just as a letter being sent from Paris to Manchester would go through several post offices along its journey, so too does information traveling the Internet. The letter from Paris would first go to the local French post office where the UK code would lead to the letter being forwarded across the Channel. Once in the UK, the British post office would use the Manchester post-code to further forward on the letter. Information on the Internet travels in a very similar way, but rather than going through offices, it goes through computers that direct it to a final destination.

However, information on the Internet does not travel in a single piece as a letter in the postal system does. Internet material is broken into fragments as it is sent, and reconstructed after its journey. Imagine writing a letter to a friend on a series of post cards, labeling them from beginning to end with numbers, and then mailing them. The cards would not necessarily arrive in the order needed for them to be read, but because you numbered them in sequence your friend would be able put the message back together once all the cards have arrived. This system of sending information in small and manageable pieces is called "packet switching" and while the system may not seem very sensible to those of us used to traditional mail, it is what makes it possible for the Internet to reliably send files of all sizes around the globe.

Transferring information in packets very quickly found many uses on the Internet. Each usage needed its own rules of behaviour, called a protocol. One of the first uses was to send large files from one computer to another. This came to be known as file transfer and is still heavily used today. Every time software is downloaded from the Internet, the 'file transfer protocol', or ftp for short, is used.

Email also uses the system of sending packets of information. However, unlike ftp, email is related to a specific person and not just a computer. It is for this reason that e-mail addresses are of the form John.Dupont@Hoola.nl where the part before the famous "@" sign is the person's name and the rest is the name of his/her mail computer. E-mail uses a protocol called smtp (simple mail transfer protocol).

The Web is just another clever way of using packets to transfer information. Consider a computer with some information on it that you want to make available via the Web. To get this material on the Web, the computer would need to be hooked up to the Internet and be running a program called a server. This just sits on the computer listening to requests coming in. Whenever it gets one, it sends the material requested. However, more often than not, there is nothing specific in the request, and the server sends a standard page (known today as a home page) to welcome the person to the site.

Home pages are often covered with lots of suggestions that can be clicked upon to get more information. These click-able suggestions are hyper-links. They are the basis for the Web and make it possible for computers communicating across the Internet to link to one another so easily. Without hyper-links there would be individual pages that you could access by entering in addresses like http://www.cern.ch, but there would not be a way to jump between pages without having to constantly enter new addresses. Surfing would simply not exist without them. Like ftp, the Web has its own protocol, called "hyper text transfer protocol" or http.

In the 1960s the world was in the midst of the Cold War. Fears that a nuclear strike would hit a central section of the US telephone system and wipe out all communication across the country led President Eisenhower to support the development of a de-centralized network that could survive nuclear attacks. Independent research in the US and England led to the concept of packet switching and sending messages without having any central exchange became a reality.

The ideas were first tested in the academic world, by linking up large computers at top universities across the USA: the ARPANET. The Americans were not alone though, France and the UK were making similar efforts and packet switching networks were soon common in the academic world. It soon became clear that all these networks would have to be able to talk to each other, and by 1980 researchers had created the Internet standard known as TCP/IP (Transmission Control Protocol/Internet Protocol) to make this possible.

The Web, oddly enough, developed out of particle physics research at one of the world's first international science organizations, the European Organization for Nuclear Research (CERN). In the 1980s CERN was bringing together some of the most creative and intelligent people from all over the planet to work on the big questions surrounding particle physics. CERN was also putting to use hundreds of separate computers to monitor particle collisions in its accelerators. While the computers were good at looking at collisions and the people were very good at thinking about physics, neither was particularly good at communicating their findings in any sort of organized fashion.

In 1989 British physicist Tim Berners-Lee tackled this challenge. He took the concept of the hyper-link, which he had been using as a method of toggling between menus on the accelerator computers, and used it to establish a system that allowed information within the organization to flow much more freely than ever before. The Web was born.

The Future of the Internet and Web... Grid Technology

Today the Web allows users across the globe to travel the Internet and access information electronically. And fantastic as the Web is, it is merely the tip of the iceberg, there is much more to come.

At the moment researchers working in projects that requires vast amounts of computer power to analyse data have to either buy the computers necessary to do the work or physically bring the data from the project to an outside source of analysis. Neither option is particulary efficiente, but the second one, that of bringing data to an outside source could be efficient if the external resources could be access electronically. And that is where the Grid technology comes in. It will allow computers resources to be shared.

How Will Grid Technology Work

Grid computing will make use of systems connected to one another through the Internet and pool all of their individual resources into a very powerful, single, virtual system. In a sense the idea of the Grid is a bit similar to that of the Web. Both, the Web and the Grid use the Internet to access outside computers, but unlike the Web (which only looks at hypertext documents) the Grid actually use the computer power of the computers it accesses.

Putting Grid Technology To The Test

Researchers working to solve many of the most difficult scientific problems have long understood the potential of shared super computing systems and in Europe the Grid is becoming a reality in the form of the European DataGrid Project.

The project is coordinated by CERN, the European Laboratory for Particle Physics and is intended to create a Grid resource that scientists all over Europe will be able to plug into and use. A major force that has driven the development of the DataGrid to be specifically at CERN is the laboratory's developing particle collider, the Large Hadron Collider (LHC), which is expected to generate several million gigabytes of data per year. This is far more data that any single research centre could ever possible analyse, making the computer resource sharing offered by Grid technology essential.

More info on the GRID technology.

By Matt Kaplan, science writer

and Robert Cailliau (CERN)