The dynamic nature of conflict in Wikipedia

TL;DR

This paper introduces an agent-based model to simulate Wikipedia editing conflicts, revealing how editor tolerance influences conflict levels and demonstrating the impact of automated bots on editing dynamics.

Contribution

The study develops a simple, data-driven agent-based model that reproduces key features of Wikipedia edit wars and introduces a new conflict measurement parameter based on mutual reverts.

Findings

Conflict level depends on editor tolerance parameter.

Distribution of inter-peace periods is heavy-tailed.

Wiki-robots influence conflict levels and editing patterns.

Abstract

The voluntary process of Wikipedia edition provides an environment where the outcome is clearly a collective product of interactions involving a large number of people. We propose a simple agent-based model, developed from real data, to reproduce the collaborative process of Wikipedia edition. With a small number of simple ingredients, our model mimics several interesting features of real human behaviour, namely in the context of edit wars. We show that the level of conflict is determined by a tolerance parameter, which measures the editors' capability to accept different opinions and to change their own opinion. We propose to measure conflict with a parameter based on mutual reverts, which increases only in contentious situations. Using this parameter, we find a distribution for the inter-peace periods that is heavy-tailed. The effects of wiki-robots in the conflict levels and in the edition patterns are also studied. Our findings are compared with previous parameters used to measure conflicts in edit wars.