Al Jazeera’s interactive team AJ Labs have a mantra: “human driven data journalism”. In a guest post for OJB Hanna Duggal speaks to the team’s lead Mohammed Haddad on what this means and how he tackles big data, including a recent story commemorating the Arab Spring. 

Mohammed Haddad joined Al Jazeera just as the Egyptian revolution began to unfold in 2011. Since then he has been behind some of Al Jazeera’s most prolific data stories, covering everything from UN General Assembly voting to mapping India and China’s disputed borders.

And, while many of the issues Al Jazeera covers are deeply complex, AJ Labs often help to explain such narratives using data journalism.

The key is to avoid overcomplicating the charts, says Mohammed.

“You should be able to just look at a chart and very quickly grasp what it is trying to communicate to you.”

Working with big data: The Arab Spring 10 years on

In 2011 revolutions engulfed the Middle East with pro democracy uprisings in Tunisia, Egypt, Bahrain, Libya, Yemen and Syria.

To commemorate the 10 year anniversary AJ Labs recaptured how the Arab Spring unfolded on social media, specifically on Al Jazeera’s own Twitter account at the time. Mohammed, who was covering the uprisings in 2011, says:

“Al Jazeera played a significant role in the spreading of information during that time. So, we wanted to recreate those events for our audience.”

Using tweets from the time, along with maps and charts, AJ Labs mirrored a Twitter feed to construct a play-by-play of each uprising. 

“From a design decision we wanted it to feel like you’re scrolling through this endless feed of information.

“There was definitely a side to it that was experimental, but fundamentally it was a story.” 

One of the challenges with the story was the sheer size of the Twitter dataset – at 250 gigabytes, it took Mohammed several days to download the data.

“The Arab Spring Retweeted was a good exercise in what big data really means,” he says.

“The whole point is that these stories help us to push the boundaries forward for the next story.”

Unique data challenges

At NICAR this year data journalists working in the global south highlighted some of the obstacles they face when reporting – from accessing data to dealing with formats that aren’t machine readable. 

For AJ Labs, some of these challenges involve credibility, says Mohammed. 

“We face the same problems as other news organisations, but we have additional challenges.

“Access to credible data has always been a problem.”

Although data is available, and more datasets are being created by people understanding the importance of data, newsrooms in the region still face unique challenges such as issues with language.

“We’ve got to be very practical about the tool sets and the training available, especially in the Arabic language. You don’t have as much available as you do sources,” explains Mohammed, who also runs data journalism training sessions in the global south.

Groundbreaking projects: Palestine Remix

Although data journalism is still a fairly new approach in Middle Eastern newsrooms, AJ Labs employed technology quite early in their work.

In 2014 filmmaker Rawan Damen approached Mohammed with hundreds of hours of footage on the Israeli Palestinian conflict. Palestine Remix was born as a result of their collaboration. The platform allows users to cut and remix documentary footage on the conflict. 

“The remix culture was very much the thing people were involved in then,” says Mohammed. “This was a time when the creation of content was really encouraged and popular.”

It took three months to launch the platform. The project won the Drum Online Media Award and Webby Award in 2015.

“We ended up getting the most phenomenal feedback from students to academics to our own audience.”

After several years it is still one of Al Jazeera’s most popular stories on Israel and Palestine and has outlived the usual news cycle. 

“For 2014, nobody had really been doing this yet. For us, it was an opportunity to show the world that we can also use leading edge technologies on a story that is so important to us and our readers.”

Human driven data journalism 

Data can be the source of a story or it can be the tool used to tell the story itself. AJ Labs has covered a range of topics using different approaches, from immersive scrollytelling to simple maps accompanying news stories. 

“Yes, these are data stories,” says Mohammed. “But they are so much more. These are human stories first and foremost.

“Every single story we do, that’s our mantra at AJ Labs: human driven data journalism.”

Data journalism has the ability to abstract complicated datasets and break down complex narratives using infographics.

However, this doesn’t necessarily mean data journalism is the best way to tell a story. 

“Sometimes speaking to a source is enough to drive a narrative 100% and other times it will be represented by data because you don’t always have the same level of detail,” says Mohammed. 

A good example of this is the Broken Homes project published in 2016 – one of AJ Labs’ first big data stories, which looked at homes being demolished by Israeli forces in occupied east Jerusalem. 

“We worked with the UN and every month we would get the dataset to show how many homes were being demolished and how many individuals were left homeless or displaced.”

After sending a reporter to the ground to get interviews, the data, images and personal accounts were overlaid onto a map. 

“It’s a really telling piece. It was a kind of ‘aha’ moment for most of our newsroom, who understood that data is not just the numbers that are in a spreadsheet.”

As newsrooms adjust to the ebb and flow of data journalism it’s worthwhile noting what role data has in journalism compared to other fields.

“I think it’s really useful to go through the tools, and we could have huge conversations about programming languages and technical stuff,” says Mohammed.

“But, when it comes to representing to the world what we’re trying to achieve, we’ve always got to be story-first.”