Project description

“Is there specific political inclination on media toward elections?”: We grabbed daily news reports in Taiwan and used keywords to extract specific politicians and observe the changes in quantity. And pass the content to Google’s artificial intelligence analysis to see whether the media reports are neutral or not. This project continues until the next presidential and legislative elections in 2020.

“Digitalization of political donations version 2.0”: According to Taiwan’s laws, candidates need to declare the sources of political donations and publicize the data for public examination after election. However, those data are much less public than “collected” in the Control Yuan. We need to go to the Control Yuan in person to review the data, and get permission to copy all the data on paper, which can’t be used for further analysis. Therefore, we have decided to digitalized the data, and visualized the data to make the tangled connections of politicians and businessmen easy to read.

“Call to action : collection of election billboards of 2018”: We designed a “billboard check-in” system to invite readers to take pictures of the candidates’ billboards on the street and upload it. Those photos are assembled into a billboard map, recording the election by the cooperation of the civil community.

“The secret of the claw machine trend”: In the year of 2016, nearly 2000 arcade claw machines that had been set up from main road sides to small alleys in Taiwan, changing the landscape of the cities gradually. It was not the first time for Taiwan to experience that kind of craze. The streets of Taiwan once filled with arcade slot machines, however, with the law enforcement by the government, slot machines had been wiped out. The arcade machine manufacturing industry had switched westward to China.

“The hereditary phenomenon of county and city councilmen in Taiwan”: In a democratic society, support from voters is the essential key to success for politicians. Therefore, “connections” becomes important assets that can be passed to the next generation. Hereditary political succession thus becomes the most powerful rival for political novices. How ”tight” the connections behind the candidates are ? How strong the power behind the candidates is ? We searched massive information online, examining family tress of councilmen then visualized the connections. READr published this feature before 2018 national election, hoping the feature can be a reference for our readers to vote.

“fakebook: fake news and where to find them”: fake news is a worldwide headache. In Taiwan, fake news has influences on election, national security or even a human’s life. Taiwan government wants to introduce new laws against misinformation, but critics worry that freedom of expression is under attack. READr has studied various kinds of data, analyzing fake news issue from different angles ranging from the government, online platform operators, fake news manufacturers and citizens. We hope to offer our readers more information and encourage more discussion before fake news regulation being legalized.

“Graphic explainer: the deadly 43-minutes chat record of Puyuma Express derailment.”: Taiwan Railway 6432 Puyuma train was derailed in front of Suao Xinma Station at 4:50 pm on October 21, 2018, causing serious casualties, 18 people were killed and 190 injured. On October 25th, the communication records of the dispatchers and drivers of the Taiwan Railways were exposed. We used the map and interactive dialogues to bring readers back to the situation on the day.

2018 is the second year of this team, we tried some different methods to collect data, use the data, and make report. Although there are no resources as large teams, we have also made a lot of attempts.

What makes this project innovative?

READr is not just a data newsroom, it is also a digital innovation team. Therefore, we always hope to make breakthroughs in every topic, not just using traditional news methods to tell stories.

The difference between our team and the general data newsroom is that the founder has a background in engineering and photography. Therefore, the various roles of the team members (including journalists, designers, engineers, community managers, product managers, etc.), everyone's opinions are equally important to the reports. Each member can make the best contribution based on his own experience and professionalism to complete the methodology of the storytelling and the user experience of the interactive. Without the framework of traditional thinking, the team can make the report more creative and also keep the news professional by presenting stories in a true and complete way.

What was the impact of your project? How did you measure it?

The development of data journalism in Taiwan media is still not perfect at present. Although the READr is only a small data journalism newsroom, it is almost one of the few relatively well-established teams in Taiwan. Therefore, many of the topics produced by READr have an indicative impact on the development of data journalism in Taiwan.

For example, in the project“PoliticalContributions”, we have attracted hundreds of thousands of readers to understand the political contributions of the Taiwan Legislative Council elections and the problems of the current system. Soon after the report released, the public began to pay attention to the fact that the political contribution information is not accessible, and finally force the government to pass a new bill. Now in Taiwan, all the record of political contributions must be uploaded online, and open to the public.

The bias and false news of the media is one of the most serious problems encountered in Taiwan. Through objective data, people can know exactly how serious the problem is. And after we released the report, many citizen technology groups are willing to cooperate with us, ready to collect more information and carry out more relevant reports.

Source and methodology

There is no fixed way for us to produce projects. Methods and data differ depends on contents, which we always strive to find the most suitable one. Some of the materials used in the project are from the government's public information, and some are from our own. We digitized materials by manual typing, web crawler on websites, sometimes from the supporters' contributions.

We designed a variety of ways for readers to contribute, such as the pattern of Google reCAPTURA, which cuts the file into small pieces and lets readers help identify it, just like playing a typing game. Or design an uploading system for the project “ElectionBillboard”, readers can take photos and upload it when passing by the billboard, just like everyone will "checkin" in the restaurant. The way led us to successfully collect the data of candidate billboards for the 2018 election.

We also reduce readers' reading thresholds through different web forms. For example, in the project of“Aboriginalpursuit of traditional areas”, we tell the story by couples of card news with stunning photos, making the content more lively.

In addition, due to Taiwan's thriving citizen technology community, we often work with them to obtain government-related information. We always choose the most suitable way to make the reports perfect.

Technologies Used

We usually use Perl and Python to complete the crawler, in order to get the data. And use Mongo DB and Google Cloud SQL to store data. As for data analysis, we will use Perl, Python, OpenRefine, and R, etc., to have different tools according to different needs.

To render web pages, we use HTML, CSS, Javascript and JS's framework - Vue.js. And use HiChart, Flourish, D3.js, HTML5 Canvas for data visualization. Some pages that require more complex operations and generate APIs will include Python and Golang. Visual design generally uses Adobe Illustrator, Adobe Photoshop, Sketch.

In the deploy section of the digital report, Kubernates is used to execute the Docker container and deploy on the Google Cloud Platform. At the same time, we also seek the best technical support according to the needs of the topic, including Google NLP (Natual Language Processing) API, Google Map API, Google Earth API, Google Docs API.

Project members

Lee Yu Ju, Chien Hsin-Chan, Hsu Ling-Wei, Tan Hsueh-Yung, Chiang Kai-Chih, Hsiung Kai Wen, Chen Yi-Chian, Hsu I-Chiao, Chen Yen-Yu, Li Chao-Yen, Chen Tzy-Tyng, Liu Tzu-Wei

Link

Additional links

Followers

Click Follow to keep up with the evolution of this project:
you will receive a notification anytime the project leader updates the project page.