 Welcome to my lightning talk on the R-Package Data Rebellion. My name is Benjamin Witzenberger. I work as a data journalist for Süddeutsche Zeitung in Munich, Germany, and I'm the author of the package. Data journalism has sparked a lot of interest during the last year, with visualizations, dashboards and models being everywhere around the news and on social media. And COVID has kind of proven the value of journalists that have a sense for sometimes a bit old park side of numbers, their uncertainties, and their need for an explanation. In combination with the ability to explain and visualize data to a broader audience, data journalism really made a difference during the pandemic for many news corporations. As you can see, we have a broad range of visualizations that came out during this time. As data journalists, we use our lot in our everyday work. For example, to scrape data, to tidy it, or to analyze it. But we rarely use the visualization tools for publication. Take ggplot for instance. It needs a lot of tweaking to arrive at a style guide acceptable theme. And for mobile and desktop screens, you will need at least two variants of each graphic with different dimensions. And this is where and why dynamic graphics come into play. This light lists some options that you have for those. You could of course create those charts yourself by using libraries like D3 or high charts, which have really powerful R packages. But this requires at least some maybe a lot of knowledge about their respective frameworks and probably a self-hosting instance. There are publication-ready software as a service offering like Plotly, Instagram or Flourish. They come with a prepared set of kind of the standard toolbox of charts. Some even have special WIS types, Flourish for example offers a bar chart raise out of the box. And some of those services already have a good R implementation, for example Plotly. Others at least have an API that you can send calls to. And those all include hosting options. One of those ready for publication services is DataRapper, which I'm going to talk about a bit more. They are a small startup located in Berlin and they were originally founded by journalists and visualization experts. Right now they've grown to, they've grown a lot during the last couple of years and also their product has grown a lot. And up to this day most of the news companies around the world use DataRapper for creating visualizations for their reporting, as well as some financial and governmental institutions. And I think one reason why DataRapper might be this popular, they have a very comfortable free tar, which covers nearly all the futures apart from branding. And it's very easy to use and embed in webpages, but still creates powerful and even slightly interactive charts. I created DataRapper, the R API library to the tool in the end of 2019 as a leisure time project. Right on time for COVID. Until now it is only available via GitHub. It offers access to the most used functions from within R, creating, updating data and the elements of a chart and publishing it. DataRapper became an important feature for automating COVID charts last year. We used it to update our charts and annotations twice an hour, modified maps and tooltips, and even created small multiples with an export function that allows image exports from the tool with our own branding. The next major release will bring increased API handling by using retries for the most important functions. It will also handle non-CSV formats better and a couple of arguments should work smarter. This next release is version 1.2, should be up within the next couple of weeks. I will try to make the library accessible via Chrome this year to simplify the download process. If you have any further questions, don't hesitate to contact me. I am Munich Rocker on Twitter and GitHub or you can send me an email at info at benedict-witzenberger.de Thanks for watching and enjoy the user conference.