C is for client-server. What on earth is that, cried the computer industry when the term was first used about 10 years ago. Pure jargon, said most people. No, no, said a few: it's the Next Big Thing in information technology. (Everyone who works with computers calls their line of work "information technology", or IT.)

As events have turned out, the few were right. Client-server is now the truly big thing in computing, or IT. You almost certainly use it yourself every day without realising. So what on earth is it?

The concept is simple enough: you, sitting at your computer, are the "client"; a remote computer, with lots of information that you want to examine, is the "server". Your computer contacts the remote one via a network, and asks for the data you want. The remote machine chunters around its file stores (which may be enormous, or may even require it to contact another machine entirely to do some processing on some raw data) and sends back the data you want.

For companies, that means that you only require one copy of the data essential to your company, rather than having copies on every machine where people might want to access it.

Where is client-server used? Almost everywhere, now. The hole-in-the- wall cash machine is a client that contacts your bank's central server to check your PIN number when you want cash or an account balance. The World Wide Web is a client-server system: your PC is the client, and contacts file servers that may be located anywhere in the world.

The intriguing thing about client-server is that the really important element is not the computers. The servers are as fast and as stuffed with data as the mainframes of old; the clients are similar to their old "dumb" terminals. The key is fast networks. Once you have those, you can spread your data and processing power around, and not have to worry about where it is. On a really fast network, everywhere is the same place.