Way back in 1937, a diligent group of BBC researchers began a project to catalogue everything the corporation had created.

Already creating a vast range of programmes, the BBC was in danger of losing track of its own output.

Someone needed to make a searchable record of every broadcast, including notes on who appeared in it, what they said, when it was first shown and repeated, and so on.

Once set in motion, this enormous task never stopped. Sometime in the 1970s or 1980s, the first steps were made to computerise this huge database.

Thankfully the BBC's archivists had been extremely clever at organising their data sensibly.

Over the decades they had stuck to the same categories, the same date formats, and used the right names for the right people.

Once it was digital, this information was a fantastic website, waiting to be made.

In 2005, members of the BBC's new media team rediscovered the database in the BBC archives. Stunned by its detail, they watched as one of the library staff flicked through pages and pages of data on an ancient 80s-era computer.

Their task was straightforward: make the website a reality.

The first version of the site has now been launched. You can see it at open.bbc.co.uk/catalogue/infax

Note the prominent disclaimers on the front page: this is a work in progress, and still very much a prototype. It doesn't necessarily include everything, and certainly contains mistakes. But so what?

There's so much to browse, so much to discover. It's addictive.

Take, for example, the archived comments about an episode of science fiction classic Blake's 7: "Dreadful shot of two attacking spacecraft. They look cheap and nasty ... More shots of cheap, tacky spaceships."

These archivists didn't mince their words.

The database is presented as a hyperlinked "cloud" of information.

You can search for just about anything you like, including the names of programmes and people. Once inside, it's easy to jump around from one subject to another by using the tags assigned to each broadcast.

You can also navigate by date, which means you can see all the shows broadcast on the day you were born, or on any other specific date.

The archive goes right back to that earliest archive entry in 1937 (poetry by WB Yeats on the wireless).

Despite dealing with very old archived data, the modern website was built using some of the most cutting edge internet technology.

It was built using a web programming language called Ruby on Rails, and includes all the latest web gizmos. There are RSS feeds all over the place, and all the data is available as RDF, a structured language for presenting data online.

The project is just one part of a much wider BBC initiative to open itself up to licence fee payers.

Soon, the BBC website will undergo a facelift and, a bit like MySpace.com in the US, start offering users the chance to create their own content.

There will also be the launch of the new BBC iPlayer, a web-based system for viewing TV shows up to a week old. This is an extension of the existing Radio Player (bbc.co.uk/radio) that offers the same service for radio.

As well as launching the online catalogue, the Beeb has announced a competition for anyone with an eye for design. Called "Reboot", the idea is to throw out everything on the existing BBC home page and start from scratch.

Anyone can enter; all you need to do is read the rules at open.bbc.co.uk/reboot, and submit your entry (which could be a working prototype, or just a sketch). The winner gets a top-of-the-range Mac laptop, runners up get swish MP3 players. Worth a go.

Another online project from the BBC is Backstage (backstage.bbc.co.uk) which offers web developers (professional and amateur) the chance to integrate BBC data with their own (non-commercial) offerings.

For example, one chap has created a mobile phone application that fetches traffic information about the M4 from all kinds of different news sources, and wraps it up into one convenient package.

See timalmond.com/fetch-m4 to find out more.

* Giles Turnbull has a web site at gilest.org.