homeresearch › webcommunities

Web communities: Dynamics of online collaboration

One year of group activity in the flickrsphere. Each dot corresponds to a Flickr group. Details on membership/content growth for highlighted groups are displayed on the right.

The study of the growth dynamics of collaborative Web communities has attracted considerable attention in the research literature. Online peer production systems typically die of inactivity for an insufficient number of valuable contributions or, conversely, whenever quality assessment becomes unmanageable due to content explosion or ineffective measures against spam or vandalism. The governance of such communities has been based for decades on best practices. A growing body of research, however, has started studying the impact of policies, social and technical constraints on how these communities evolve over time. Measuring the performance and growth over time of these communities, and identifying the effects of disruptive events such as the drop-off of very active contributors, explosions in content or disruptive types of interaction can shed a light on how to better grow these communities.

Can we identify factors that determine the life and death of content-based online communities? How can communities in mutual competition secure their performance and the quality of their output? What is the relation between governance, content quality and participation in online communities?

The aim of this project is to study factors affecting the sustainability of content-based online communities and help develop tools to achieve or restore desired goals in content and population dynamics. Case studies for this project include social media services like Flickr and wiki-based communities.




Web services

I designed and developed two Web services as part of this project (read more):

A prototype web service providing platform-independent analytics and comparative growth statistics for wiki-based communities.

Year: 2008-2009 (not publicly released)
Languages: PHP, Javascript, Flash, XML, XSD, SQL, CSS

Flickr Group Trackr
A web service to track and analyse demographic and activity metrics for public Flickr groups.

Year: 2007-2008
Languages: PHP, Javascript, SQL, CSS
Featured on: Lifehacker, Programmable Web ("Best new mashup"), Yahoo! Gallery ("Editor's picks"), Computational Aesthetics, Flickrbits.

This research was supported by two grants from the European Commission: FP6 project PATRES (NEST-043268) and FP7 project QLECTIVES (FP7-ICT-231200). Work published since 2011 was partly supported by the Wikimedia Foundation.

Qlectives logo PATRES logo
Valid XHTML 1.0 Transitional :: Valid CSS :: Powered by WikkaWiki