Social Annotation: Emergent Text Signals through Self-Organization

Abstract

In this paper we discuss a type of social software that might be called social annotation. Such software enables individuals to add annotations underlines and highlights, for example while reading a text. Additionally, some aggregation of those individual annotations is then added to the text in the form of typographical text signals (e.g., bold typeface). A study of 40 undergraduates using a social annotation application, called CoREAD, suggests that the software enables increasingly stable typographical text signals to emerge over time. Stability indicates that a group consensus had been reached about the most relevant sections of the text. Additionally, these social text signals were shown to match quite well with text signals that had been added to the text by its author. This indicates that group consensus can generate social text signals that adequately capture the important concepts in a text.