User:Filll/Quantitative Taxonomy of Wikipedia Editors

Quantitative Taxonomy of Wikipedia Editors edit

Abstract edit

Some quantitative methods for categorizing Wikipedia editors and their editing styles are described.

Introduction edit

Wikipedia is one of the most popular websites on the internet. It is composed of several million articles written in over 250 different languages. Wikipedia is created by thousands of volunteer editors, who are guided by a set of policies, principles, guidelines, conventions and traditions. Different editors are involved in a wide variety of tasks, and exhibit different editing styles.

There are many reasons to categorize editing styles on Wikipedia. For example, not all editing styles contribute equally to the production of Wikipedia, and it is valuable to understand which styles are more useful to Wikipedia and in what circumstances. Recognizing unproductive or inefficient behavior and measuring it's characteristics can be useful in the creation of evidence-based management systems, which hopefully will replace the current intuition-based management systems. For example, when selecting editors for promotion to administrative positions, or other positions of responsibility, current methods can skew the selection process towards those with limited forms of editing experience. Gauging editing styles and experiences can allow a conscious decision to avoid certain kinds of editors, to create a wider diversity of candidates, or to otherwise assist in the selection process. Also, measuring the results of certain administrative actions or imposition of policy changes can be valuable information for evaluating the appropriateness of these actions and changes.

Quantitative methods of analysis are often desirable. Quantitative methods are amenable to automation and computer implementation, which is appropriate for a computer-based document production project. Numerically-based procedures can reveal features of situations that are not readily apparent to less exacting and rigorous methods.