Understanding collaborative work practices represents a critical factor in and a necessary fundament for the development
Background and MotivationUnderstanding the way a group of workers interacts with each other represents a major challenge and a prerequisite for the design and development of computer systems that aim to support collaborative work. Today, collaborative work increasingly takes place in a distributed fashion, with actors being dispersed across geographic locations. In the past, comprehensive research has been conducted to analyze roles and relations among a network of actors. On one hand, this has led to the emergence of the field of Social Network Analysis (SNA) and the development of supporting software tools [23,22,31] to visualize and analyze especially distributed social networks. Evidence for the relevance of SNA in the context of designing Computer Supported Collaborative Work (CSCW) systems can be found, for example, in the initiation and organization of the workshop on social networks during CSCW 2004 ([34, 32, 7] in Chicago. However, utilizing social network models as a basis for the development of CSCW systems has turned out to be difficult because of some challenges: First, SNA approaches that rely on analysis of available data (such as [7] on the basis of e-mail logs) struggle to provide appropriate visualizations that can present the huge amount of resulting data in a way that aid analysts in understanding the networks [34]. Also, because SNA predominately focuses on social aspects, other aspects that are relevant in the context of system design, such as organizational or technological factors, are not covered adequately.On the other hand, knowledge management (KM) as a research domain emerged during the last decade to investigate and manage knowledge as the most critical resource of organizations [17]. To a certain extent, knowledge management is concerned with understanding social relations within communities [25], but also introduces complementary modeling dimensions, such as knowledge-, organizational and technological aspects [29]. Within the knowledge management domain, knowledge work analysis can be regarded a research branch of its own which is concerned with the descriptive modeling of socio-technological real world systems (or "object systems" [5]). Key issues in this domain include the multi-dimensional identification and visualization of knowledge work practices. However, approaches in this domain typically are labor-intensive and rely on a co-location of analysts and investigated actors (such as B-KIDE [29,30] or ARIS(KM) [2]) and therefore are not well equipped for being applied in distributed settings.So while social network analysis on one hand provides suitable concepts for conducting analysis among a set of distributed actors (through e.g. automated log analysis), it lacks consideration of necessary modeling dimensions and appropriate mechanisms to handle the huge amount of data to successfully aid the process of developing CSCW systems. On the other ha...