Amid all of yesterday’s gloomy news from the BBC was hidden one cause for hoorays. The project I’ve been involved with for the past couple of months, BBC Programmes, launched. No, not much consolation to people worried about losing their jobs, but definitely a very good thing for the BBC.
The site looks pretty simple and you might wonder why it’s such a big deal. Its aim is to provide a permanent page for every episode of every TV and radio programme broadcast by the BBC. From every episode of EastEnders through to every game of Shinty. This seems like a simple, and even obvious, challenge but from the first idea it’s taken around three years, and I don’t think any other TV or radio network does something similar (OK, that’s a wild guess; please correct me).
This is all important because, as Tom Coates said in 2005, when he was working on a precursor of the project, on the internet you need to be able to point at things. If I want to tell someone about a particularly good episode of Screen Wipe then I need to be able to send them a single URL to that episode, so they can read about it and see when else it is broadcast. Which highlights one of three main differences between BBC Programmes and something like BBC What’s On:
- A page per episode, not per broadcast. This makes it much, much easier to tell when a particular episode will be on again, and means you can point at the “thing” itself, rather than one of the times the “thing” was being shown.
- Permanent URLs. Even when Charlie Brooker is running the BBC (I live in hope) and Screen Wipe is a distant memory then, with a bit of luck, each episode of the show will still be visible at the same addresses. All easily indexable and findable through Google et al.
- Simple, short URLs. http://www.bbc.co.uk/programmes/b0084kgm is an episode of Strictly Come Dancing.
OK, “b0084kgm” doesn’t roll off the tongue. I remember when Tom wrote about unique identifiers like this, way back when, I thought they were silly. No one can guess the URL of a particular show’s episode. That’s true, but the alternative is to base the URL on the show’s title which can cause other problems. Should the URL include the brand name (“Silent Witness”), and the series name (“Series 11”), and any sub-series name (“Suffer the Children”) and the episode name (“Suffer the Children: Part 2”)? You could just use series and episode numbers but you still have the brand name to deal with, which can be very long, can change and isn’t necessarily guessable (I forgot it was “Charlie Brooker’s Screen Wipe” and thought it was just “Screenwipe”). Unique identifiers, like the ISBN numbers of books, or the codes used for TinyURLs, are short and guaranteed never to change.
But I’m straying from the point: excitement. This is a big thing! There are a lot of programmes broadcast by the BBC that currently have no presence on the web at all. Many are small programmes with tiny budgets that now, automatically, have a home online. Others are big… try Googling for “Friday Night with Jonathan Ross” and you currently get nothing from the BBC except press releases and a “What’s On” 404 error. Now there’s a page for the show and it’s easy to see when it’s next on and repeated (the number of times I’ve tried to find out when, or if, a programme will be repeated on the rest of bbc.co.uk…).
If you’ve had a look at the site you’ll notice it’s far from perfect. For one thing the data is pretty ropey. Many shows are listed a number of times (because their episodes haven’t been aggregated together under a single brand name). Some episodes are listed several times for the same broadcast slot (because national BBC stations — Scotland, Wales, N Ireland — haven’t been grouped together properly yet). Some of the data is just plain weird; a show about Donny Osmond categorised under “cricket”?
This will all improve — hell, it can’t stay like that. The idea is to get it out there, as a stake in the ground, something to build on. I’d say this was part of the “ship early, ship often” school of software development, but given “early” in this case has taken three years that might be pushing it. Still, “ship often” should be true. Tom Scott talks about the site and lists some future URLs that suggest things coming up: closer integration with radio networks, proper daily schedules, tagging, RSS and Atom feeds… Many of these are well underway already, and coupled with the existing URL structure and the site’s current support for Microformats this will hopefully provide a good basis for development by third party developers, and other in-house development, quite aside from being useful to your average BBC licence fee payer.
As you can tell, I’m unusually excited about this. This is something I knew the BBC badly needed and I was very honoured to have the chance to work on it. I don’t want to take any credit for it though — they’d have got along quite happily if someone else had jumped in at the end of the lengthy process to do their HTML and CSS. And so a big “Congratulations!” to Anoushka, Dan, Duncan, Jamie, Matt, Michael, Paul, Richard, Sophie, Suzy and all the other people I didn’t directly work with. I’m sad to say goodbye (back to college next week) but they’re off to a wonderful start.