How we built a new tool for journalists in 12 hours
studies science journalism at City University
Tech-minded journalists met journalism-minded techies at City University, London, for a BBC-sponsored hackathon. The participants were challenged to develop a digital product using something called the Juicer – a program developed by BBC News Labs to tag news content.
We asked Kata Karáth, a member of the winning team at the hackathon, to tell us what happened.
The Juicer is a news aggregation system. It tags articles from the BBC and other news sites based on their content. The tags are grouped in four categories: People, Places, Organisations and Things (which is everything that doesn’t fit in the other three).
In the right hands, the gathered information can uncover entirely new types of stories. However, if you’re not the most tech-minded journalist you will need a “front end” to use the Juicer – a user-friendly interface that lets you use its capabilities without bothering you with what’s going on behind the scenes.
This was our challenge. We had to come up with a new front-end for the Juicer using different approaches to its data to target different areas of journalism.
Twelve hours over two days was our time limit. We formed three teams with a diverse mix of members from journalism, data and creative design courses. And we came up with our own Apprentice-style team names – the Hack-Athenas, the Trend-setters and the Neutrons.
So what did we do? Well, the Hack-Athenas tried to tackle some of the problems that journalists always face: finding a good and reliable source, preferably as quickly as possible. Their main idea was to use social networks to contact potential sources.
The Trend-setters wanted to help journalists to predict the future - by creating an app to track trends around certain topics and try to figure out what is likely to make big news in the near future.
My team, the Neutrons, set their sights on science journalism. Covering science news might be intimidating for some journalists. But wait: there are press releases that explain everything in simple terms and even have quotes from the scientists. However, copying press releases and using pre-packaged quotes can hardly be called journalism.
We created a prototype of a tool that would let science journalists see how many news sites were using the same quotes when covering a particular science story. The prototype would also be able to identify the quoted people and see their relationships with each other.
The idea evolved from something very different: I started by imagining a map that would show how science news coverage spread – in both space and time - after a press release is published. However, the reality of coding gave us many problems.
As we came up against these obstacles, Neil Maiden, professor of digital creativity at City, said to us "I know you want to change the world, but you can’t do it until Monday”.
One of our team members, Igoris Gajosinskas, was an experienced UI (user interface) designer. Together with Sylvia Tippmann and Lei He from BBC News Labs, Igoris helped us to understand the basic concepts of digital product designing.
Trimming down our wild ideas, we were persuaded that we needed to come up with something buildable. In the end our prototype, also called Neutron (below), is very much possible with the current version of the Juicer.
However, we won’t stop there. At our final presentation, we sketched up for BBC News Labs how we plan to realise the original idea. I’m glad to say that our team won the hackathon. The prize is a visit to BBC News Labs. You can be sure I’ll be looking for a chance to ask them about any possibilities to develop Neutron.
This was my first hackathon. I wasn’t even sure what it was when I signed up. Creating something was exciting and what amazed me most was the incredible things that were born just out of 12 hours of collaboration.
I hope Neutron, if it is finally developed, will help journalists track the origins of science news coverage and start to ask more questions.