Marie-Charlotte Matthes:

Qualitative Untersuchung der Bot-Richtlinien in der englisch- und deutschsprachigen Wikipedia

Master of Science (M.Sc.)


In the German and English language version of the online encyclopedia Wikipedia, human and algorithm-based users (so called bots) work together to accumulate knowledge and make it freely available to all people with access to the Internet. Bots hereby support the human users in performing repetitive tasks and thus extend the functions of the socio-technical environment. In both language versions, the communities have developed local policies that coordinate the use of bots. The development of policies in Wikipedia is typically surrounded by negotiation between users. Talk-pages are directly linked to the policy -pages and give users a space to address relevant problems and jointly work on solutions. The policies and their previous ver- sions are the result of a negotiation process that took place focally on the associated discussion pages, so the basic assumption. The cooperation of human and technical agents gains relevance in many areas of everyday life. It is therefore important to develop a deep understanding of how such cooperation can be shaped, in particular through behavioural governance mechanisms. In this thesis, therefore, all contributions to the discussion pages mentioned above have been examined in a qualitative study and connections and interrelations between the identified problem-categories have been explored. The discussions have accompanied the development of the policies from the day of their creation up until now and therefore provide a suitable starting point for the analysis. The context resulting from the analysis provides explanatory approaches for the development of the policies in the two language versions. In a concluding comparison of the results, it is also roughly explained in which points the development in the two communities differed. The research design is based on the threestep coding method according to Glaser and Strauss. In the data analysis, five main categories were formed, which are linked by a key category. Three categories represent the problem areas that result in the broadest sense from Wikipedia’s technical environment or are rooted in the technical functionality of the bots. The problems that are subsumed under the remaining two main categories have a more social nature. The key category refers to discussions that have arisen around coordinating functions such as processes or definitions and stands connecting between the five main categories.