Open data for everyday life

8 September 2011

TfL Countdown board at bus stop

I’ve noticed two main motivations in the campaign for government open data.

Some people want to change the world. They want to reform government and create a new era of accountability through transparent public bodies.

And some people just want to find out when the next bus is due.

The two of these aren’t mutually exclusive and there’s a point at which they meet, but today I’m going to look at how open data can improve people’s everyday lives.

My interest in local government stems from my core motivations as a designer. We can put people on the Moon and build complex and elegant skyscrapers but why do so many people struggle through their daily lives? To paraphrase William Gibson, at what point does the future become evenly distributed? When do the rest of us Earthbound ground dwellers get a slice of the techno-pie?

Local councils touch so many aspects of our everyday activities. We can go for months without having any direct contact with a central government department. We can avoid pretty much any company we don’t like, though perhaps not all of them at once. But we can barely go a day without using a service provided by our local councils.

Councils collect our bins. They maintain our streets. They educate our children. They run our parks and public libraries and leisure centres. And once a month they send us the bill. Whatever you think about your own council or local government in general, they’re unavoidable. Making councils work better directly and immediately makes people’s lives better. Designers and technologists should be all over them.

But in as much as I think that councils should always be striving to provide better services and to provide better information and access to those services, I know that they can’t do it all by themselves. People’s needs are too diverse. Technology changes too quickly. Aspirations always outstrip the ability of councils to meet them. Try their very best with the budgets and resources available and it’s impossible to keep up let alone get ahead.

Open data offers councils the opportunity to tackle one part of the problem.

So when’s the next bus due?

If you live in London this is now a much easier question to answer.

TfL have just launched their Countdown website where you can look up bus arrival times. It’s the same data that appears on the red screens at some bus stops.

TfL Countdown mobile website screenshot

In fact, there are three Countdown websites. One for computers with “normal sized” screens, your typical PC or laptop. One for smartphones and other mobile devices. And one “accessible” site for people who have difficulty using conventional screens.

These three websites tell a story in themselves. A story of diverse needs and aspirations. About wanting to provide for everyone in every context and on every device while still trying to deliver the best user experience possible in each situation.

And these sites are good. TfL has put the effort in and succeeded.

But better still, TfL has built its websites in a smart way. All three sites draw data from the same source so there’s no need to access the underlying database directly in every app. Just build an API and pull the data from there.

That API is available for anyone to use (unofficially, at least). Just request the current bus arrivals data for any stop and you get the answer in a format you can use in your own apps.

So I built something that TfL probably never considered and would almost certainly never have got approved had they thought of it: a Unix command line utility for Countdown.

Why? Because I’m in a niche group. My aspirations (if not exactly my needs) are well outside the mainstream. Like a few hundred people in London but unlike just about everyone else, I spend nearly all day every day sitting at a computer with a Unix terminal window open.

So rather than visiting a website and clicking my way around, it’s much easier for me to type:

countdown

and get a response like this:

Stonecot Hill outside Metalcraft (50799)
----------------------------------------

93    Putney Bridge        5 min
413   Morden               12 min

Ash Road (50823)
----------------

93    North Cheam          3 min
413   Sutton Garage        9 min

A staggering 33 people have downloaded this software since I published it four days ago. That’s nearly nine people a day.

This vanishingly small audience represents a technological and organisational triumph. This isn’t one size fits all. It’s one size fits one if it needs to. It’s what Clay Shirky calls situated software – extreme customisation for the most niche tastes. It’s the information you need, when you need it, in the way that suits you best. It’s a tiny public service in a place where government wouldn’t dream to reach.

If the Unix command line is your thing you’re welcome to become the 34th user.

So those three websites that TfL created for their Countdown service aren’t overkill. They aren’t even covering all the bases. There are hundreds of different ways in which this data could be used in every conceivable context and through any imaginable technology, past, present or future. You could build it into other websites and apps. Put it on screens in shops or hospital waiting rooms. Print it out on till receipts. Run a flag up a flagpole when a bus is around the corner. Whether it’s a labour of love, a business opportunity or a way to improve something that already exists, you can make it. TfL has built the API and the mainstream web apps. The rest is up to you.

So that’s London buses. Some other UK towns and cities have got similar websites though I’m not sure whether they’ve got APIs yet. They need them.

What’s next?

Socitm has recently published its plans for its annual survey of council websites, Better Connected.

Better Connected 2012 will focus on top tasks, the most common things that people visit their council websites to do.

Socitm suggests that the top tasks are:

Socitm will be looking to see how well councils provide this information and these services on their own websites.

Councils should work to make all these things as easy as possible to do on their own websites.

But you shouldn’t need to visit your council’s website to do any of them if something else would suit you better.

An iPhone app. An app for a tablet computer that got withdrawn after two months on sale. A nicely formatted table on a leaflet.  A desktop or web widget. A little light flashing on your bin on the day it needs to be taken out. And not forgetting the Unix command line for the three dozen people that might use it.

As councils polish up the top tasks on their websites they should be building APIs for those services too, starting with the most popular and working their way down.

There’s more than one way to do it and there’s nothing better than doing it your own way. We need open data and APIs for everyday life. Councils have the ability to make it happen.