Nonfiction

It’s a World Wide Web Out There

25 February 2016

There’s a piffling argument about when to celebrate the birthday of the World Wide Web.

Many mark 12 March 1989 as its day of creation. It was on this date that Tim Berners­Lee submitted a proposal for a new Information Management System to his boss at the European Organisation for Nuclear Research (CERN).

Almost exactly nine months after his proposal, Berners­Lee physically brought the first web page, web browser and server software into the world. This was on 20 December 1990, an alternative date some celebrate instead.

Perhaps the day you choose depends on whether you view the web as a conceptual or physical invention.

We tend to use the words internet and web interchangeably, thereby blurring crucial distinctions between infrastructure, software and abstract ideas about information sharing. What many people might not know, is that whilst Berners­Lee was in the midst of refining his project, the internet had already existed for some time.

Back in 1965, two people had managed to connect a computer in Massachusetts to one in California using a dial­up­ telephone line, impressively creating the first wide­ area network. The problem with telephone lines however, is that when you speak, there is only one line to transfer your conversation to the other end, meaning that when the line is in use no one else can make a call.

Imagine if every computer in the world had to be rigged up to all the others with separate phone lines? How slow! And who would pay for all the wiring?

The solution was packet switching. Say you want to send a friend an IKEA coffee table for their birthday, sending such a big thing via post is unwieldy and expensive. So you break it into smaller parts, give the pieces to other friends and ask them to drop those pieces off when they next visit the lucky friend’s place. The table can be assembled on arrival.

This is a thoughtless manner by which to send gifts but a great way of sending data through networks. In packet switching, data is broken down into packets and each one has the address (IP) of the computer it wants to get to. When there are lots of smaller packets, they don’t have to follow a direct route down one wire in the way that your telephone conversation does.

Data can break up and go all over the place (sometimes to different countries) before it arrives where we want it. Routers help the data decide which direction is the most efficient at any given moment. Once all the data has been delivered, Transmission Control Protocol (TCP) tells the computer how to reassemble the data in the right order, just like an IKEA instruction manual, but probably easier to understand.

During the Cold War, the US military was interested in protecting sensitive information being conveyed to and from distant installations. They were convinced enough of the potential of packet switching to see through its development. On 30 August 1969, four computers were connected to each other in what was called ARPANET, the world’s first internet.

So if computers can send data to one another with ease, what do we need the web for? It all comes down to user friendliness.

The idea Berners­Lee proposed was actually intended to solve internal problems at CERN. CERN is that place in Switzerland where they smash tiny particles together (ostensibly at the world’s peril). It’s a hub of extremely intelligent, creative physicists who work on large, brow­furrowing projects.

Back then, they had trouble sharing and keeping track of the all the information they produced. Berners­Lee noted: “…all the information could be written in a big book. [But] keeping a book up to date becomes impractical, and the structure of the book needs to be constantly revised.”

His solution was to do away with a classical hierarchical and fixed system of information sharing in favour of “a ‘web’ of notes with links (like references) between them”.

In his proposal, Berners­Lee referenced a term that was coined way back in the ‘60s by a guy named Ted Nelson – Hypertext. Nelson describes Hypertext as “non­sequential writing”. It allows users to follow ideas, breaking from a text or media and moving to another at their will.

Nelson noted that speech and books have to be sequential for the sake of convenience, but that “the structure of ideas are not sequential. They tie together every which way”. Hypertext thus gives power to the user, allowing them to seek information in a way that mirrors their brain’s own functions much more closely than say, your Mum’s old Encyclopedia Britannica set.

CERN scientists came from all over the world and so they used many different kinds of computers such as old Unix, Macs and PCs. If different computers are going to talk to each other efficiently they need a common language, a set of rules. So Berners­Lee developed Hypertext Transfer Protocol (HTTP), a computer language that enabled these computers to request and retrieve information from each other using hypertext links.

He also invented the first web server, Nexus. A web server is a computer that waits for a request to come from another computer – a browser. When you click a link in your browser, it builds a request in the computer’s common language (HTTP). The server reads it and sends back the requested files.

These languages, processes and wires were all weaved together by Berners­Lee to support the world’s first webpage, which still exists today – Google ‘World’s First Website’ for a geeky and underwhelming experience).

Following the success of ARPANET, commercial internet service providers showed up in the late ‘80s. Finally in 1991, CERN very kindly bequeathed the World Wide Web to the world – another possible birthdate, if you view the web as something that should be accessible to everyone.


Leave a Reply

Your email address will not be published. Required fields are marked *