Authors: Stelios Lelis, Petros Kavassalis, Mahmoud Rafea, Seif Harid
Society for Computational Economics, Modeling, Computing, and Mastering Complexity 2003.
From its early days, the World Wide Web space has demonstrated strong agglomeration trends with a very small number of web sites capturing the larger part of the Internet population. At a firstFrom its early days, the World Wide Web space has demonstrated strong agglomeration trends with a very small number of web sites capturing the larger part of the Internet population. At a first glance, agglomeration over the virtual space sounds as a paradox. Web sites are numerous and highly diversified and can be easily reached from everywhere and anybody, with no particular transportation or search cost. However, Internet users use only a small number of sites for searching for information and products, interacting with others and socialize, thus producing dense concentrations and locational patterns similar to those observed in the physical space where few cities and industrial clusters host the huge majority of population and the entire industrial activity. Is that depending on the attractiveness of the popular web sites or are there agglomeration economies providing incentives to users to be in a location which have been visited by other users or pointed-in by other sites? This paper provides a sound basis for the dynamics of population concentration in the Web and put forward an explanation to web sites' growth by developing an agent-based computational model, with behavioural and economic variables, where the aggregate outcome emerges from the interaction of individual decisions. The model reproduces the empirically observed power law distribution of Internet users across web sites, demonstrating that a plausible explanation of web agglomeration phenomena can lie on the assumption of increasing returns and the percolation-like diffusion of the information over the Internet.
