In January 2010, a US dollar traded against 1,233 Francs Bu in the official market. Eight years later, it is worth 1,785. As for the price of beer, national drink par excellence, it continues to grow, from doubling to double in ten years. If we look at green coffee production in Burundi, it has been on a roller coaster since the beginning of the last decade and despite optimistic forecasts for 2018, they are far from the records recorded in the mid-2000s. Penitentiary population continues to increase: + 48% compared to 2018. All this information, handled by the press group Iwacu, one of the last independent media operating in Burundi since the destruction of several media in 2015, have a point common: they result from a data analysis. If the data are raw facts, they only make sense from the moment they are the subject of journalistic input.
At Iwacu Press Group, we are engaged with data driven journalism. But we also expose the facts as they are: all data collected by the editors and journalists are made freely available on a unique open data portal. IWACU Open Data, initiated and developed by Iwacu Press, aims to provide open and reusable data about Burundi, and to centralize all valid data about Burundi. With giving journalistic insights about the new or updated data, it permits to follow the situation of the country trough data. Data journalists are working closely with the newsroom, which can also reuse the data visualizations as well as the data to enrich their articles published on the portal iwacu-burundi.org The data are a working tool for journalists (fact-checking and data-journalism projects), but it is also a tool for foreign journalists, citizens, researchers, NGOs and decision-makers. The team is composed of a data manager and a data journalist, who give the impulse to data driven journalism approaches within the publications of IWACU.
What makes this project innovative?
As journalists, we need to correct, accurate and reliable data to document our articles. There is no open database specifically dedicated to Burundi , hence our idea of centralizing data, but also to add ours , those we receive in our daily work , whether for the investigation purpose or for projects of data journalism. Our goals are not only journalistic. We encourage the reutilization of data since they provide information on Burundi. While there is a lot of data on Burundi, it is disseminated on the web pages of open data portals or NGOs. Sometimes you have to dig hard to find them. Before the launching of this project, editors and journalists already worked with data, but they did not do anything with the collected data after their article was published. The main idea was to centralize those data in one point, and to share it as we consider that information is a public good. Since the destruction of independent media in 2015, IWACUPress Group remains the only independent media still active in Burundi, where transparency and good governance are not a political priority or preoccupation. Indeed, before this project, there were no open database specifically dedicated to Burundi. News information about Burundi is often manipulated and fake news are a reality within a very difficult political context. That is why IWACU Open Data contributes to document the country with facts as well as to fight against disinformation at an age where numbers can be manipulated or reinterpreted. This unique project in the area also permits to open windows on Burundi, this poor and small country often forgotten, its challenges and its difficulties. With no big means, this project proves that working with data and to open it are possible even it is remain a daily challenge.
What was the impact of your project? How did you measure it?
Website audience (from March 6th to April 6th): users 54,330 New users 40076 Seen pages 525 642 social media analytics Twitter (Summary over 28 days): 38,4 k followers Tweets 167 Profile visits 19.6k 162.8% Mentions 752 251,4 % Facebook : 47K followers Most popular datasets Total change in external debt stocks (current US$) 2637 views Children out of school, primary, male and female 2579 views Repeaters, primary, male and female (% of male & female enrollment) 2516 views Aquaculture production (metric tons) 2188 views
Source and methodology
Standardization is an essential element given that the objective is a broad sharing and data dissemination. It starts with a strong semantics, standard-based IPTC, for controlled vocabulary for the media. Taxonomy and proposed definitions are free to use (Creative Commons 4.0 license): Our project is open, so we use open tools. We have adapted this to portal uses in our own taxonomy. It is connected to our platform. You can access the data sets in alphabetical order via the “Tags” tab. You can check the organization of these keywords on this page, developed with the open source Tematres vocabulary server. This is the first strong semantic layer to our datasets but there are others (Dublin Core standards and Schema.org).
We prefer open formats like CSV *, they must both be readable by humans and machines to encourage all forms of reutilization. A Uniform Resource Locator (URL) to assign a unique identification data. If this is possible, we provide related information (linked data ) . Our philosophy is that of total open, respecting the W3C recommendations for open data, web standards body. A CSV file opens with any spreadsheet (Excel, Open Office …)
1. Elvis Mugisha 2. Pierre Claver Banyankiye 3. Laurence Dierickx