Sieve - Data Privacy Made Simple

If a mobile device user is interested in an app’s privacy policy, he would have to search for it online, just to find out, that it is a very long and pretty much unreadable document. Therefore, a lot of people do not get the information they were looking for, even though it would be very important to know, e.g. how a company handles the user’s data. The data privacy company Sieve wants to empower its users by teaching them about privacy policy in a way that is easy to understand. To achieve this, Sieve wants to present the app policies in a condensed form and rate the different apps based on how much data they collect. This should allow users to make an informed decision on what apps to install on their phone.

For the client Sieve, a React Native MobileApp, that shows data privacy information of apps was developed. To help a user choose what to install on his mobile phone, the Sieve App can sort apps into categories and shows a unified rating for each of them. Furthermore, the Sieve App also allows its users to stay up to date on privacy policy changes of their favourite apps and get general news on app privacy. All of the information inside of the Sieve app is entered manually by Sieve using a React web application called AdminWebsite. Between these two components lies a Node.js WebServer that manages the data and is reachable through a REST API. Unstructured files like images are saved in an object storage and the structured data is on a MongoDB Data Store. These two storages as well as the AdminWebsite and the WebServer are hosted in the cloud.

Our prototype contains the three core components MobileApp, AdminWebsite and WebServer, that Sieve needs to inform their clients on app privacy. The three parts of the application each offer basic functionality and can easily be expanded upon, because of the use of well-known web technologies. The modular design and the use of the flexible REST API allows to add different data producers and consumers to the Sieve landscape in the future. A possible expansion that Sieve is considering would be a Natural Language Processing unit, that automatically creates key points from a verbose privacy policy. Such an addition could easily be integrated into the existing system by using the REST API.