Please, write less

And other tales of developers and journalists

Analía Plaza
Interviews on digital culture

--

Stijn Debrouwere is an OpenNews 2013 Fellow, a network by Mozilla Foundation and Knight Journalism Centre whose fellowships “bring some of the best minds on the internet into newsrooms for 10 months of open-source experimentation”. Last year he worked at The Guardian. He also blogs about computer code and the future of journalism on topics like Information Architecture for News Websites and Disruptive Innovation in the News Industry.

Even though I think journalism is — and has always been — what it has to be (going to the street, talking to people, picking up the phone, fact checking, understanding numbers and telling so people understand, too) I find interesting how traditional and new media are finding their path in our digital lives. After working in a printed magazine that thought of its website as a repository of the print instead of as a different product, I still wonder how media should be on web and mobile.

I talked about this to Stijn. We sat at The Guardian Coffee in London, a recently opened cafeteria that somehow collects all the trends of nowadays journalism: a data visualization screen, some iPads on its tables and even a long form stories printed newspaper.

An edited transcript of our conversation follows.

You ‘help people build websites’. What do they need?

If you look at how most news websites get made, you always have a bunch of developers, coders and designers, but there’s usually no one outside of that. No one who really thinks about why do you this or what you wanna build. Most of the work I do has to do with everything that regular developers or designers get to do.

For the last year I have been working on analytics to measure whether the stuff on the website was actually visited or not. Before that, in Iowa, I worked for a local newspaper and TV station. They were interested in doing more community journalism, where you not only write about the community but you also interact with it. I helped out with how can you actually do that. Because it’s a really big concept.

Last year you worked at The Guardian. What kind of problems do they face right now?

I think their biggest problem right now is that they have a lot of stuff going on: really good journalists, great coders. But it is such a big organization and everyone does their own thing. So, for example, when you look at movie reviews, The Guardian would have reviewed the same movie by 5 journalists, 5 different times. Sometimes on purpose, but sometimes because nobody talks to each other. If you have over 500 journalists, not everyone reads what everyone else does. So you get into situations like that.

“Of 150 articles, maybe 3 or 4 get 90% of the page views”

How did you try to solve it?

I don’t think I solved it. In my mind, the problem is that The Guardian is so big and even people who works there do not know how it actually works. A lot of what I did was analytics and figuring out how many reviews, how many culture stories, how many sport stories, we published. When you ask any journalist or editor how much do they think they publish on any given day, I think 90% of them will not have any idea. And the first step towards solving that, and try to come up with solutions, has to be with content analysis.

For the election of Pope Francis we (they) wrote about 150 articles. It is really interesting to see what gets read and what does not. You’ll find that of those 150 articles, maybe 3 or 4 get 90% of the pages.

“There is potential to focus on fewer articles that are better, sort of authoritative”

Is it worth publishing 150?

I think, again, part of it is that things are not always as coordinated as they should. And that we don’t know beforehand which articles are gonna get traffic.

If you look the way Buzzfeed works — a good example, because traffic is all they care about — , they publish hundreds of articles because they know it’s almost random and that you never know what people are gonna like (even thought they know a little bit, and that’s why they post so many cats). But they post as much as possible to see what works.

To a certain extent, that applies to The Guardian as well. There is potential to focus on fewer articles that are better, sort of authoritatively, that people feel like: ok, if you wanna know something about this topic, you have to read it at The Guardian.

“Journalists or news organizations often assume that you read everyday”

I read you saying that ‘news are difficult to understand’. I think, because of the information overload, I might have the same problem. Do you still have it?

Exactly! It’s also the way it’s written. I’m originally from Belgium, but I don’t follow the news from Belgium anymore because I don’t live there. But every once in a while I go to a belgian news website and it’s like ‘OH! This is happening. Wait, but… What is this about?’ Journalists or news organizations often assume that read everyday, and if you don’t…

“Most people are doing interactive stories as a form of marketing”

There is a debate on ‘Snowfall storytelling’. What are your thoughts on it?

Those kind of stories can work, but if you look at the numbers a lot of people don’t scroll down until the end.

Did you finish reading Snowfall?

I read about a quarter and then I figured that I would really wanted it on my Kindle. The main thing to keep in mind with interactive experiences is that they take a team of 2 to 20 people that will be working on that from a week to a couple of months. Let’s imagine that the pages they are 10 or 20 times as good as the average article. Well, but if you put 100 times the effort in it turns then is still 10 times as bad in terms of the return on investment.

Most people are doing this kind of interactive stories as a form of marketing, of branding, because it’s impossible to make money with them. But as a brand you can say: hey, I am The Guardian or I am The New York Times. And this is the kind of journalism we do.

You also said that some journalists were asking you to do interactive stuff. Are we kind of obsessed with interaction and superb forms of presenting?

Ever since the web got a widespread option, people have been wondering what journalism should look like on the web. And you see it with every trend, not just interactive stories but also… When the iPad came out. People were asking: is iPad the future of journalism? With every news innovation, people just jump on it and pretend it’s gonna be the future.

What’s the latest — useful — innovation that you have seen?

In terms of how news website works? To be honest, I don’t think they’ve gotten better at all. Since ten years ago. If you look at it they’ve gotten a little bit prettier, nicer, but in terms of how they work almost every news website is very clutter and when you got to a front page there’s twenty stories that get your attention — that’s the same thing.

“When I was at The Guardian, people in the tech department were sort of pleading journalists and editors: please write less, but write more in depth stuff”

Could you mention any example that we should look at?

In most newspapers, the front page is about 50.000 pixels down? It’s scroll and scroll. And you could probably cut out the bottom half and it wouldn’t make a difference. Instead, it would make your site load faster and that does make a difference, because the slower the site is the faster people are gonna click away. The trouble, when you’re The Guardian or The New York Times and you publish four hundreds stories a day they have to go somewhere, right?

The ironic thing for me, when I was at The Guardian, is that people in the tech department were sort of pleading journalists and editors: PLEASE WRITE LESS, but write more in depth stuff. And things like topic pages, overviews or the bigger picture, so if someone has not been following a piece for a week could get something.

“When you look at daily news, we present people episodic bits of information”

“We’re in the information of business”. Should we be information centric instead of story centric?

I think so.

So what should I do with this interview, for example?

I don’t know if I would do anything in particular with this interview. It’s more when you look at things at a global perspective. For example, if you interview a hundred people about this topics. Then you could start to categorize, find certain trends. It applies to some kinds of journalism. I still read a lot long form journalism, which is exactly the opposite.

When you look at daily news, we present people these episodic bits of information. If you look at how well informed people actually are when they read a newspaper, turns out that they now a little bit more about what’s going on than people who do not read the newspaper, but still they know very well.

If you’re following a murder trial and the process spreads out over hundred different news articles, with a more information centric approach maybe you would still write all of these independent articles but structure things in a way that says: this is when this happened, and this is when that happened. So people could get an overview. It would have things like topic pages, so you can tell people that has never heard about a topic: first read this, then maybe you wanna read that.

Politifact is interesting. They have a big amount of structured information. A politician says something, so who is the politician, when did she or he said so, what was it exactly that he said, is it true? They factcheck it, they list of all their sources separately, then they list whether it was made for which election, if it had anything to do in an election… If you have all that information, it’s easy for people to browse through it. They got a Pulitzer.

“As part of the responsive movement, you will see smart movements towards simpler websites, bigger forms and less interruption”

Everyone is now talking about 2014 trends. What do you think we are going to see?

I am a bit of a cynical because what I keep seeing year after year and to some extent I contribute myself is people saying ‘we’re gonna do this and we’re gonna do this’ and then nothing ever happens. I don’t know if there’s gonna be anything special. More websites are going to be responsive and, as part of the responsive movement, you will see smart movements towards simpler websites, with bigger forms and less interruption. If you look at Medium’s design or Wired redesign, they go in that direction.

I’ve heard a lot about data journalism and infographics.

I think there would be a movement in terms of standardizing that work. Right now, if you want to create a simple map you have options that require very little programming. But for a lot of other visualizations it takes a team of programmers to build them. People are starting to realize that you need some of that for branding, where in most cases it’s just so wasteful. So I think that what we have seen over the last years, and what we are going to see next year, is people creating more tools. So you can put out visualizations quickly in different kind of stories.

It affects what journalists do. If you have those tools, you have more power to create instead of needing someone else to do it.

“There is a entire world of information that would be available to journalists, but you need a way to extract it, to make sense of it. And that’s coding”

You’ve blogged about why everyone should learn to code. Why do journalists should?

A lot of people have been saying that journalists should learn to code, but I think it’s bigger than that. Just think of any kind of job! Even what a doctor does: they have to track so many files; sometimes they are interested in see patterns, but most doctors do not have enough statistical or programming skills to do anything on that. A lot of business people enter manual information in spreadsheets, whereas there are automatic ways on doing that work.

In Iowa, there were certain court cases that the newspaper was interested in. The information was available online but in a format that was weird to follow, so you never knew what was going on unless you had a list with everyone you wanted to follow. I wrote a little script that took those names automatically and did all of those searches. That takes a bit of time, but court reporting is not gonna go away, so if you write the script once you can keep using it for years.

If you think of the work that journalists do, a lot of times is about trends. If you know a bit about statistics, you can get those numbers and search for stories there. There is a entire world of information that would be available to journalists, but you need a way to extract it, to make sense of it. And that’s coding. And you’ll find that it’s like that in almost every job.

--

--