Free Riding on Gnutella

By Eytan Adar, Bernardo A. Huberman. First Monday, 2000

1,831 citations recorded by the source corpus.

An extensive analysis of user traffic on Gnutella shows a significant amount of free riding in the system. By sampling messages on the Gnutella network over a 24-hour period, we established that almost 70% of Gnutella users share no files, and nearly 50% of all responses are returned by the top 1% of sharing hosts. Furthermore, we found out that free riding is distributed evenly between domains, so that no one group contributes significantly more than others, and that peers that volunteer to share files are not necessarily those who have desirable ones. We argue that free riding leads to degradation of the system performance and adds vulnerability to the system. If this trend continues copyright issues might become moot compared to the possible collapse of such systems.

Subfields: Artificial Intelligence, Computer Science Applications, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/94558e330577b429abdc358791cda57e715bdd13

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Science Applications

In Sociology and Political Science

Citations in the corpus

Efficient search in peer to peer networks (2004, methodology): “Keywords: Peer-to-peer, distributed data, search, performance modeling, evaluation Technical Area(s): Databases, performance modeling and evaluation”
Crowdsourcing, attention and productivity (2008, positive): “The dilemma ensues when every individual can reason this way and free ride on the efforts of others, making everyone worse off [1, 3, 7, 5, 10].”
Mapping the Gnutella Network: Properties of Large-Scale Peer-to-Peer Systems and Implications for System Design (2002, methodology): “Others have used this data to analyze Gnutella users’ behavior [2], to analyze search protocols for power-law networks [6], and to forecast network growth through simulations [15].”
KARMA : A Secure Economic Framework for Peer-to-Peer Resource Sharing (2003, background): “…implicitly assumed that peers will altruistically contribute resources to the global pool and assist others, recent empirical studies have shown that a large fraction of the participants engage in freeloading: 20 to 40% of Napster and almost 70% of Gnutella peers share little or no files [1] [2].”
Psychological research online: report of Board of Scientific Affairs' Advisory Group on the Conduct of Research on the Internet. (2004, background): “106 February/March 2004 ● American Psychologist cial loafing and altruism ( Adar & Huberman, 2000 ), and friendship patterns (Adamic & Adar, 2001), among other topics.”
Search and replication in unstructured peer-to-peer networks (2002, background): “For example, the freeloader phenomenon is examined in [2], and the topology and query distribution are studied in [21, 14].”
Should internet service providers fear peer-assisted content distribution? (2005, background): “Apart from BitTorrent, several measurement studies have addressed the issues of availability [2, 3, 8], integrity [29], flash-crowds [4][14], and download performance [1][26][25][3] in other P2P systems.”
Cooperative peer groups in NICE (2003, background): “while the vast majority of users use the services offered by this generous minority [4].”
A reputation-based approach for choosing reliable resources in peer-to-peer networks (2002, background): “6%, lies between the value of 70% reported in [3] and the 25% value reported in [19].”
Free Riding in BitTorrent is Cheap (2006, background): “In 2000, Adar and Huberman [1] noticed the existence of a large fraction of free riders in the file sharing network Gnutella.”
Chainsaw: Eliminating Trees from Overlay Multicast (2005, methodology): “Our system does not rely on any specific topology, however we could use other membership protocols like in BitTorrent [7] or Gnutella [1] For the remainder of this paper, we assume that the application is similar to live streaming.”
Harvesting ambient geospatial information from social media feeds (2011, background): “This behavior is conﬁrming observed blogosphere characteristics (e.g. Shi et al. 2007) and comparable behavioral patterns observed in online forums (see Zhang et al. 2007) or ﬁle sharing sites (e.g. Adar and Huberman 2000).”
Distributed algorithmic mechanism design: recent results and future directions (2002, background): “However, initial studies show that there is a serious “free rider problem” [2].”
Improving search in peer-to-peer networks (2002, background): “References [2] and [15] measure user characteristics through passive observation of messages.”
Analyzing peer-to-peer traffic across large networks (2002, methodology): “Previous research [1], [2], [3], [4] has focused almost exclusively on P2P signaling traffic, and on the public-domain Gnutella and Napster systems.”
A reputation system for peer-to-peer networks (2003, neutral): “Getting content from a P2P network involves two phases: 1) content search and 2) content download.”