Analysis and Prototyping of Personal DataStore Solutions and Applications
Master thesis
2019Users have no control over data they uploaded or that was collected about them on the internet as nearly all of the data is stored in databases of different platforms. The companies behind these platforms often hide the details on how the data is used behind long terms of use agreements. This way, users often do not know that their data is used to create extensive profiles about them. These profiles allow third-party companies to target advertisements to specific user groups. The profiles can also be used to manipulate and spy on users as well as to spread fake news. The potential abuses and the problems concerning the lock-in on platforms-which forces users to stay on existing platforms-are a problem and need to be addressed.
Before presenting a solution, the thesis provides insights into Personal Information Management. The reasons why information management is important are discussed. Various types of personal data are explained to showcase the amount of data that is relevant to users. Afterwards, different tools, their problems, as well as shortcomings, are presented.
The next chapter explains the idea of Personal Data Stores (PDS), starting with their historical development and the definition of essential terminology. Reasons, why PDS could be a solution to the problems mentioned earlier, are presented. Afterwards, a short market analysis of current PDS solutions is done. Next, the thesis reviews the necessary key technologies in the areas of semantics, as well as authorization and authentication. They are followed by a compilation concerning the regulatory framework of PDS solutions.
Finally, an application that uses an existing PDS solution as the backend is developed. During the development, several problems with the existing PDS solution occurred. The encountered problems are analyzed and are solved in a prototype PDS solution that gets designed and implemented.