Mapping an Audience Centric World Wide Web: A Departure From Hyperlink Analysis

📝 Abstract

This paper argues that maps of the Web’s structure based solely on technical infrastructure such as hyperlinks may bear little resemblance to maps based on Web usage, as cultural factors drive the latter to a larger extent. To test this thesis, the study constructs two network maps of 1000 globally most popular Web Domains, one based on hyperlinks and the other using an “audience centric” approach with ties based on shared audience traffic between these domains. Analyses of the two networks reveal that unlike the centralized structure of the hyperlinks network with few dominant “core” websites, the audience network is more decentralized and clustered to a larger extent along geo-linguistic lines.

💡 Analysis

This paper argues that maps of the Web’s structure based solely on technical infrastructure such as hyperlinks may bear little resemblance to maps based on Web usage, as cultural factors drive the latter to a larger extent. To test this thesis, the study constructs two network maps of 1000 globally most popular Web Domains, one based on hyperlinks and the other using an “audience centric” approach with ties based on shared audience traffic between these domains. Analyses of the two networks reveal that unlike the centralized structure of the hyperlinks network with few dominant “core” websites, the audience network is more decentralized and clustered to a larger extent along geo-linguistic lines.

📄 Content

Mapping an audience centric World Wide Web: A departure from hyperlink analysis Harsh Taneja Missouri School of Journalism, University of Missouri

Suggested Citation: Taneja, H. (in press). Mapping an audience centric World Wide Web: A departure from hyperlink analysis. New Media and Society. Forthcoming This Pre-Print Version May Differ from Final Publication Abstract

This paper argues that maps of the Web’s structure based solely on technical infrastructure such as hyperlinks may bear little resemblance to maps based on Web usage, as cultural factors drive the latter to a larger extent. To test this thesis, the study constructs two network maps of 1000 globally most popular Web Domains, one based on hyperlinks and the other using an “audience centric” approach with ties based on shared audience traffic between these domains. Analyses of the two networks reveal that unlike the centralized structure of the hyperlinks network with few dominant “core” websites, the audience network is more decentralized and clustered to a larger extent along geo-linguistic lines Keywords: Hyperlinks, Web Usage, Media Globalization, Cultural Proximity, Audience Duplication, Network Analysis

Harsh Taneja (PhD, Northwestern University) is an Assistant Professor in the School of Journalism at University of Missouri. His research focuses on media audiences, and explores connections between patterns of media use and the accompanying institutional and social structures.
Address: 181-C Gannett Hall, Missouri School of Journalism, Columbia, MO, USA −65211- 1200.
Email: harsh.taneja@gmail.com

Mapping an audience centric World Wide Web: A departure from hyperlink analysis The Internet and specifically the World Wide Web (henceforth, Web) has emerged as a platform with a massive capacity for global information exchange with social, economic, political and cultural consequences. Hence, scholars in disciplines ranging from physics and information sciences to geography and communication studies are interested in mapping its structure. All such maps attempt at providing a factual reality of the Internet and include “architectural plans, engineering blueprints, anatomical drawings, and statistical graphics” (Dodge, 2008: 352).
A common approach in the social sciences to mapping the Internet is to create networks of the structure of the Web based on hyperlinks between Websites. Hyperlinks reflect the behavior of webmasters and any imagery of the Internet based on technical infrastructure alone is insufficient to map the heterogeneous contours of actual global Web usage, especially as Internet access has both broadened and deepened globally. Therefore, maps of the Web, based solely on hyperlinks, provide a partial representation of the medium’s structure. Recognizing this limitation, recent studies have mapped the Web using audience centric approaches, either based on shared audiences between websites (e.g., Taneja and Webster, 2016, Taneja and Wu, 2014) or user clickstreams (e.g., Wu and Ackland, 2015). This paper empirically compares audience centric structure of the Web with its’ hyperlink structure.

Specifically, this study maps the Web using two simultaneously obtained datasets on the world’s 1000 most popular websites, one an “audience map”, based on shared audience traffic between these websites and the other on hyperlinks between them. A network analysis of both these maps suggests that cultural factors, such as linguistic and geographic similarity between websites, explain the audience map to a larger extent than they explain the “hyperlinks map”. These findings have implications for popular and scholarly imaginations of the global Internet. The global Internet structure Hyperlink analysis Maps of hyperlinks have contributed to popular and scholarly imaginations of the global Internet for about two decades. Based on counting hyperlinks between pages, these maps conceptualize the Web as a network with Web pages as nodes, where any two pages are tied to one another if they have hyperlinks between them. These nodes can be individual Web pages, or their higher-level aggregations such as Websites or “top level domains” (e.g.“.com”, “.de”). The earliest such studies, found that the Web has a centralized “bow-tie” structure with a strongly connected “core” component and several peripheral weakly connected components (e.g., Barabasi and Albert, 1999 Broder et al., 2000;). As the Web has grown enormously in size, recent studies also find hyperlink networks to exhibit similar properties irrespective of whether one analyzes individual Web pages (Meusel et al., 2014) or aggregations of pages such as Websites, or “pay level domains” (the latter refers to sub domains of top level domains that webmasters generally pay for, see Lehmberg et al., 2014). Hyperlink analysis at this scale has consistently suggested that the Web has a “core/ periphery” structure. In other words,

This content is AI-processed based on ArXiv data.