Under the Skin of Networks: How Concentration Affects Social Practices in Web 2.0 Environments

/

92132371

Let’s go straight to the point: to what extent are Web 2.0 services – blogs, social media, communities, and all the environments we usually refer to as participative and free – really democratic? Whether it is due to structural or accidental factors, it seems evident that new hegemonies are taking place inside the horizontal pattern of the internet, to the point that nobody can still imagine that this pattern will eventually lead to a horizontal reconfiguration of society at large. On the contrary, we could question if the web, as such, is a participatory, egalitarian, and democratic structure, by analyzing some new foundings coming from both historical evidence and empirical research.

In the last few years, in fact, the discovery of the power law nature of the web¹ has opened up a new field of research, revealing that in the real networks, such as the World Wide Web, a majority of sites with very few links coexist with a few highly linked nodes, or ‘hubs’. In other words, the power law structure is the mathematical demonstration of inequality, intended to be the main rule in the web’s development.

This new field of research is now focused on the concentration mechanisms by which the most valuable resources of the web cluster around a small number of nodes: most of the traffic, an increasing number of documents, and, above all, the number of incoming links (and, consequently, the hierarchical ranking as determined by Google’s algorithm). The most delicate aspect of the matter is, however, not so much the power law distribution of links, which has already been widely investigated,² as the impact of this structural inequality – the so-called ‘predicable imbalance‘ of the Pareto Principle – on web surfing behavior, which, a little less predictably, appears to be subjected to increasing levels of concentration.³ Being that the internet is affected by a huge concentration of resources, the way people use the web is revealing to be unequal and non-democratic in its turn.

Let’s consider the case of Twitter, the well-known microblogging platform, first released in 2006, and now very popular. A recent study conducted on 260 million tweets showed that 0.05% of the user population attracts over 50% of all attention,⁴ following a statistical distribution that, with alarming symmetry, reflects the general consolidation of web traffic around a few major hubs. In the case of Twitter, however, this result is not so surprising: Twitter is in fact a micro-broadcasting platform where, as other studies show, ‘the top influentials are retweeted or mentioned disproportionately more times than the majority of users’,⁵ following a power law distribution of information.⁶ It is instead more interesting to examine what happens in other web domains, which are supposed to be characterized by more democratic and horizontal modes of participation. For this purpose, I am here going to consider about 40 studies related to three web environments, in ascending order of size: open source communities, where people cooperate in order to create and distribute free software applications; Wikipedia; and finally, the blog system.

Open Source Communities

Let’s start with the open source communities: made up of free software activists, they are the privileged spaces for a vibrant participatory democracy, whether inspired by ideological motives, as traditionally pointed out, or by a more down-to-earth competitive advantage brought about by human cooperation, as stated in more recent studies.⁷ It is only by engaging in voluntary acts of cooperation and peer production that it is actually possible to break free of the selfishness myth, claims Yochai Benkler⁸, well-expressing the widespread tendency to overestimate a phenomenon that – apart from the successful case of Linux and a few others – has generally produced poor, if not disastrous, results.⁹ This faith in the open source alternative, and in its capacity to bring out ‘virtuous behaviours’ from the grey and dull background of social relations¹⁰ was already challenged by an old study showing that, among the 400 members of the Apache open source community, only 15 were responsible for over 90% of the changes to the code.¹¹ If such a concentration is also found in the development of free software, can we therefore conclude that there is no substantial difference between the open source organization and other production systems?


Of course, we can’t. There are still many differences, especially with regard to the participants’ motives, which are not always ascribable to the classical forms of legitimation of economic activities: the fleeting pleasure of intrinsic satisfaction;¹² the virtuous connection between effectiveness, ideological tenets, and mutual trust;¹³ but also, and more realistically – as emerges from the interviews with about 80 participants in open source activities – the singular combination of an ideal ambition to increase one’s ‘human capital’ and the concrete investment in the perspective of ‘future monetary rewards’.¹⁴ These differences are very significant, at least according to a widespread rhetoric, so much so that they constitute the revolutionary premises for a new social contract.¹⁵ Nevertheless – and this is the point – they are still far from being translated into a really alternative working condition, free from hierarchical and authoritative structures. If we broaden the scope of the analysis to a wider community – considering not only the developers, but also the participants in the Apache discussion forum, in the period from 1996 to 1999 – the result is the same: the flow of information is ‘relatively concentrated’, because 50% of the answers are provided by the 100 most prolific providers, who only account for 2% of all providers.¹⁶ The same result is also obtained by a similar study on the Mozilla development community, where 50% of the technical problems turn out to be fixed by a small minority of 113 users.¹⁷ Further confirmation comes from an analogous study on the development of the Debian operating system, based on the Linux open source platform, which was originally inspired by radical participatory ideals and then, as in the worst-case scenario, ended up being subjected to a rigid centralization of management.¹⁸ Open source software development does not lead to a ‘self-organizing system’ based on peer production processes, as another study points out, because projects are organized ‘from above by developers and maintainers whose control and authority is important to the quality of the outcome’.¹⁹ Although it is not a traditional authoritative power, adds Steve Weber²⁰ referring to the case of Linux, there is definitely a ‘pyramidal flow’ of information, a clear division of labor, and an organization that ‘sounds very much like a hierarchical decision structure’.

Free software development may therefore be ‘distributed’, as the results of a more extensive survey confirm, but it is most certainly ‘very top heavy’: of more than 12,700 authors considered, only 10% accounted for more than 72% of the total code base, and the top 10 authors alone (0.08% of the total) were credited for about 20% of the overall result.²¹ Commons-based peer production, observe Benkler and Nissenbaum,²² emerges in environments driven by collaborative efforts and results from the meeting of free individuals allergic to ‘managerial hierarchies’; but, as often happens with human things, the shattering of old hierarchies ends up producing new ones, as blatantly revealed by the statistical measures of online activities and by their compliance with the ‘80/20 rule’ of power law distributions. However, this conclusion does not come as a surprise: open source communities are actually built around specific technical objectives; it is therefore inevitable that, over time, the leadership of the most expert individuals emerges, as many studies point out.²³ It is a meritocratic authority based on technical competence and therefore regarded as legitimate. However, is this the only reason or is there something more elusive, structural, and deep-rooted than that? To verify that, it is necessary to shift our attention to other web environments, where access skills become more widely available.

Wikipedia

Everybody knows Wikipedia and, what is more, everybody agrees on the fact that Wikipedia is the most successful collaborative project on the web. What is less well known, however, is that the distribution of updates per page follows a power law, fragmenting the encyclopedia’s texture into a core of highly-linked pages, which stand out from an immense periphery of less visible content.²⁴ Although on a smaller scale, this brings us back to the initial question: if there is no democracy among the pages of Wikipedia, does it exist, at least, among the members of its community?

A friend of mine once told me of his interest in writing a Wikipedia article about a particular (and almost unknown) musical genre (a very narrow sub-genre, close to micro-music). He sent the proposal, which was eventually turned down, for the music he was interested in was supposed to be too elitist for Wikipedia’s standard. Therefore, the answer not only revealed to me that some gatekeeping practices are affecting the political economy of the web, but at the same time showed how many selection criteria, that are decisive in this process, still depend upon traditional values, such as the supposed public interest of an item, the dictatorship of the active ones, and so forth. And so, what is really new in Wikipedia?

To better examine this issue, we should focus on the concept of ‘edit wars’: heated discussions about the correct version of a page, conflicts arising from consecutive revisions of contents, which are not confined to controversial topics, but are an integral part of Wikipedia’s activities,²⁵ so much so as to spark a wider debate on the scope and inclusiveness of the project.²⁶ This is not, however, a conflict between equals: among the authors of Wikipedia there is actually a small number of ‘strong’ contributors, who actively work on the creation and revision of contents, and a greater number of ‘weak’ authors, who contribute only occasionally, usually on specific topics of interest – defined respectively as ‘Zealots’ and ‘Good Samaritans’.²⁷ More prosaically, Bryant, Forte and Bruckman call them ‘experts’ and ‘newcomers’, insisting on the fact that the transparency of Wikipedia’s structure makes it easier to keep revisions under control, relegating ‘novices’ to ‘peripheral participation’ with activities that ‘carry low risk’ to the community as a whole, so as to stabilize the internal hierarchies of the system.²⁸

As for the above-mentioned edit wars, according to the partial statistics available, the frenetic growth of these disputes is not accompanied by a proportional increase in the number of revisions. The average number of edits per article is actually fairly low (roughly 11), at least in the English edition, and the number of unique users responsible for the edits is even smaller.²⁹ Another study, conducted on a small empirical data set consisting of seven articles, points out a further complication: in 70% of the cases, edits are made by a small number of repeat contributors, or habitual editors, with a great number of edits made by the initial article creators themselves.³⁰ Most significantly, over the five-month period examined – from February to July 2007 – a considerable amount of the initial content is left unchanged: between 76 and 98% of the original text, depending on the articles, undergoes no modification. Such a marked departure from the collaborative nature of the wiki format, both in terms of substance and figures, requires some explanations:

One pattern we call first-mover advantage. The initial text of a page tends to survive longer and tends to suffer fewer modifications than later contributions to the same page. Our hypothesis is that the first person to create a page generally sets the tone of the article.³¹

Now, the first-mover advantage – the benefits coming from the early occupation of a space, just like in the first pioneers’ race toward the West – is a golden rule of the political economy of the Web,³² so it is not surprising that it also applies to our case. The question is another: if the increasing collective participation in Wikipedia does not result in a more active work on the contents – which seems quite limited and shows no sign of increasing – where does it go?

Here we find a paradoxical trend. While social participation in Wikipedia continues to grow, the amount of direct work leading to new article contents proportionally tends to decline: the percentage of edits made to article pages has actually decreased over the years from 90% of all edits in 2001 to 70% in 2006, and the percentage of edits resulting in the creation of new pages has decreased to less than 10%.³³ In contrast, the amount of indirect work – which does not go into creating new topics and articles, but is spent on activities such as ‘conflict resolution’, ‘consensus building’, or ‘community management’ – has greatly increased.³⁴ Through these seemingly secondary and invisible communication channels, argues Benkler, Wikipedia allows its members to forge stronger connections and defines the boundaries of its own identity.³⁵ In other words, what Wikipedia is doing is laying down its own rules.

The quality and effectiveness of these rules are still to be established. On the one hand, interviews with members of the community seem to suggest an increasing decentralization;³⁶ on the other hand, there is relatively little empirical evidence to support this widely held view. Of the 29,000 registered users examined in a study, for example, 143 were administrators and only seven had developer rights and decision-making autonomy³⁷: in other words, there are only a few arbitrators vested with the authority to ban other individuals from the game.³⁸ Wikipedia’s organization chart, observes Christian Fuchs, can be described as a pyramid, with ‘anonymous users’ at the bottom, and above them in ascending order, ‘registered users, ambassadors, mediators, administrators, arbitrators, developers and bureaucrats’ (with Jimmy Wales at the top, acting as a sort of metaphysical ‘benevolent dictator’), of which only the last four, at various levels, share decision-making responsibilities and power.³⁹ However, rising in the ranks is not difficult, continues Fuchs, and depends on the progressive acquisition of skills: an assumption that recalls the principle of ‘meritocratic authority’ of open source communities, which is however questioned by some critics, who report acts of sabotage, exclusion, and censorships carried out by administrators with the aim of preserving the ‘status quo’ and keeping newcomers at bay.⁴⁰ Such a hierarchical structure is regarded as quite open and reversible by some, and rigid and inaccessible by others. So, have we really discovered something new in our exploration of the wiki world?

The vast number of policies in Wikipedia […] indicate that governance is a thriving aspect of this community. […] In fact, Wikipedia’s formalised processes […] seem to share several of the design principles found […] in offline, self-governed communities around the world.⁴¹

There is a ‘hidden’ order, according to Viégas, Wattenberg and McKeon; an invisible Foucault-like ‘tyranny of structurelessness’, adds Vasilis Kostakis,⁴² here referring to the well-known paradigm of ‘networks of power’ first proposed by the French philosopher Michel Foucault. And this close relationship between network and power, here, means that Wikipedia, just like the whole Web 2.0, has not taken social experience out of the shadow of rules and dominion, but has merely submitted them to a different regime, with its own division of administrative roles, edit wars, conflicts triggered by revision processes, hidden mechanisms, and quibbles of new bureaucratic practices. If we look at the way Wikipedia actually works, conclude Emigh and Herring,⁴³ we can see that in order to establish itself as a repository of encyclopedic knowledge, it has been forced to set out some particularly strict rules, which were quite unprecedented in the wiki collaborative environments, to discover the function of procedures, codes, and network architectures. After all, as Lawrence Lessig puts it, on the web ‘code is law’.⁴⁴

The Blogosphere

Finally, there is the blogosphere, which, in some respects, represents the most extensive and accessible system, and therefore the one that was expected to constitute a really free and equal space of socialization. However, since its explosive growth in 2001, the picture has been less rosy, as a study conducted on about 25,000 blogs shows.⁴⁵ Right from the start, the increasing number of blogs has followed a curiously abrupt and discontinuous evolution, both on a diachronic level – because the quantitative growth is concentrated in short periods of extraordinary proliferation, or ‘bursts’, instead of being uniformly distributed over time – and on a topological level – because the distribution of links in the blog system follows the power law pattern, with a handful of sites accounting for almost all the blogroll links. This ‘bursty’ evolution has been confirmed by more recent studies, so much so that it has become an area of research in itself.⁴⁶ However, for space reasons, I will only focus on the second aspect: namely, the topological dimension of the relationships established within the blogosphere.

If the power law is the statistical measure of the connections established on the web – with a few nodes attracting the majority of links, acting as extremely powerful poles of attraction – the blogosphere is not an exception: even here, the in-links connecting one blog to the other are distributed unevenly, substantially following the 80/20 ratio of the Pareto Principle. In spite of its liveliness, the paradox of the Italian blogosphere, as Fabio Metitieri observes,⁴⁷ is the schizophrenic separation between hundreds of thousands of unknown bloggers and a very small number of blogs – no more than 200, starting with the striking case of Beppe Grillo – which actually draw the attention of the public. Apart from recalling some peculiarities of the Italian national history, this situation reflects a more general, and structural, trend: namely, the imbalance between the so-called ‘A-list’ – a small number of hyper-successful blogs, accounting for almost all the incoming links – and the vast majority of blogs collecting very few in-links, active but in fact invisible, lost in the remote peripheries of the web.⁴⁸

The power law nature of the blogosphere has been widely demonstrated by a number of studies, all based on the quantitative analysis of links through automatic calculation programs called ‘crawlers’. In the analysis conducted by Clay Shirky on a network of 433 blogs, for example, the top two sites accounted for 5% of the total inbound links, and the top 50 blogs (12% of the total) accounted for 50% of such links, clearly following a power law distribution pattern.⁴⁹ The same indications come from a study by Jason Kottke⁵⁰ on the top 100 blogs ranked by Technorati; by a larger study conducted by Cameron Marlow⁵¹ on a network of 27,000 blogs; by a survey on a sample of 45,000 blogs and more than 2 million blog posts⁵²; and by a study conducted by Farrell and Drezner on 4,500 political blogs, where the authors point out the presence of ‘especially “attractive” focal points that will allow individuals to coordinate more easily’,⁵³ commonly referred to by the literature as ‘hubs’.

In order to have a clear idea of the phenomenon, we can consider the case of Beppe Grillo, the former Italian comedian that finally founded the political movement Cinque Stelle (Five Stars), basically inspired by ecologist and anti-corruption instances, and by the rhetoric of web-based direct democracy as well. Even though his project is presented as a horizontal, non-bureaucratic and participative challenge to traditional vertical hierarchies and political parties, in fact, there is no such thing as a democratic process inside the movement. Discussions are limited to comments on topics submitted by Grillo himself in his blog, and usually don’t affect the very profound levels of strategy and decision-making, while the role played by Beppe Grillo seems to reproduce a charismatic, old-style and populist form of power, resurfacing through the apparently egalitarian structure of the blogosphere. Such an old form of power, taking place in the new digital environment: is it an exception, in the way the Web 2.0 works, or is it the demonstration of an emerging process – and has the horizontal structure of the internet little or nothing to do with its real democracy?

Even other aspects of the blogosphere, then, seem to follow a similar statistical trend: both the frequency with which authors update their sites⁵⁴ and the number of posts produced,⁵⁵ for example, fall in the same power law distribution, showing an enormous gap between a vast majority of peripheral actors and a small number of particularly active nodes. On the whole, the power law has therefore been confirmed as the general model for the evolution of the web, exerting its influence over its most diverse segments. However, if the most crucial element in the process of ‘self-organization’ of the web is the incoming and outgoing link distribution,⁵⁶ it is probably on this aspect that we need focus in order to draw the most relevant conclusions.

The Pareto Principle, instead of a normal distribution, imposes a law of imbalance on the web, exactly as it helps to understand the inequality of income distribution, by showing that 20% of population usually collects 80% of revenue. In this regard, Clay Shirky observes that the power law is not exactly a discontinuous trend: apart from the top most-linked nodes – incomparably more connected than the others – all the other nodes have a similar number of inbound links, because in a power law distribution, the gap between the first and second position is larger than the gap between second and third, and so on.⁵⁷ From this point of view, the blogosphere ends up taking on the paradoxical shape of a pyramid, with a very large base extremely distant from the apex: the reason is that the most successful blogs serve the purpose of selecting the topics of discussion for all the others – observe Farrell and Drezner⁵⁸ – which in turn link to the most important ones in hopes of indirectly gaining some sort of visibility. This contrast between a handful of successful nodes and an endless host of subjects is more than just a statistical measure: it also reflects the difference of credibility and authority between a small number of individuals, organizations or institutions, and the vast majority of others, in a sort of paradoxical return to a charismatic form of power, argues Mathieu O’Neil,⁵⁹ which unexpectedly emerges from the folds of innovation. A confirmation of the charismatic nature of these hegemonic positions comes from the above-mentioned study by Cameron Marlow, and by his distinction between blogroll links, which point to the front page of a site, and the so-called ‘permalinks’, which instead point to the deep content of the archives, where a specific topic is developed. The distribution of permalinks, elsewhere defined as ‘dynamic links’, is actually much less uneven than that of the ‘static’ blogroll links: this is probably a sign that, whereas the discussion of a topic can follow a wide range of different paths, the links pointing to the front page of a blog are instead influenced by the aura of its author, and therefore show a higher concentration.⁶⁰

Power on the Web

Now, how can these indications be generalized, and to what extent do they really tell us something about the configurations of power on the web? First of all – insisting on the different qualitative and quantitative dimensions of open source communities, Wikipedia, and of the wider blogspace – a number of studies seem to confirm the scale-free nature of the web: that is, its fractal property of falling in the same aggregation law regardless of the network size.⁶¹ With regard to the initial hypothesis, these studies seem to confirm the suspicion that the tendency of the web to cluster around a few ‘hubs’ has an impact on its moral economy: ‘the pattern doesn’t apply just to goods’, argues Shirky,⁶² with their power law curve and ‘long tail’ effects, but to ‘social interactions’ as well. However, in terms of social theory, this issue seems to remain quite elusive because these statistics cannot be turned into a general theory on the web experience, and do not provide an exhaustive mapping of the various fields of observation (open source, Wikipedia, blogs), of which we are still far from having a complete understanding. Even if we limit the analysis to the case of Wikipedia, just to mention one, an empirical study shows that the ‘power distance’ – that is, the distance between the most central members and the less influential ones – varies across different national contexts, at least between France, Germany, Holland, and Japan⁶³; if the power distance index even fluctuates among the local domains of the same community, thinking about a general theory of web hierarchies appear quite unrealistic.

Across the web we seem to be witnessing the emergence of different trends: the ‘meritocratic’ authority of open source groups, which brings the progressive and audacious ideology of the early internet communities into the age of Web 2.0; the ‘procedural’ tyranny of Wikipedia, which operates under the surface of our perceptions, through a hidden intertwining of rules, protocols, and roles; and finally, the charismatic power of the blogosphere, where words still have the face of those who utter them, and their emergence gives a sudden twist to the development and patterns of legitimation of the web ecosystem. These three different trends however share a common feature: none of them suggests a decentralized, democratic, or equal evolution of the web; on the contrary, all of them, each in their own way, imply more or less new levels of dissymmetry and power, of inequality. It is therefore from here that the imaginative effort of the theory should begin.

References

Anthony, Denise, Sean W. Smith and Tim Williamson. ‘The Quality of Open Source Production: Zealots and Good Samaritans in the Case of Wikipedia’, Dartmouth College Computer Science Technical Report Series, September 27, http://www.cs.dartmouth.edu/reports/abstracts/TR2007-606.

Bakshy, Eytan, Jake M. Hofman, Winter A. Mason and Duncan J. Watts. ‘Identifying “Influencers” on Twitter’, 2001, http://research.yahoo.com/files/bakshy_wsdm.pdf.

Barabási, Albert-László. ‘The Physics of the Web’, Physics World, July 2001, pp. 33-38.

______. Linked: The New Science of Networks, Cambridge, MA: Perseus, 2002.

______. Bursts: The Hidden Pattern Behind Everything We Do, New York: Dutton Books, 2010.

Barabási, Albert-László and Réka Albert. ‘Emergence of Scaling in Random Networks’, Science 286 (15 October 1999): 509-512.

Barabási, Albert-László, Réka Albert and Hawoong Jeong. ‘Scale-Free Characteristics of Random Networks: The Topology of the World-Wide Web’, Physica A 281.1-4 (June 2005): 69-77.

Benkler, Yochai. The Penguin and the Leviathan: How Cooperation Triumphs over Self-Interest, New York, Crown Business, 2011.

Benkler, Yochai and Helen Nissenbaum. ‘Common-based Peer Production and Virtue’, The Journal of Political Philosophy 14.4 (2006): 394-419.

Bryant, Susan. L., Andrea Forte and Amy Bruckman. ‘Becoming Wikipedian: Transformation of Participation in a Collaborative Online Encyclopedia’, 2005, http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.62.5337.

Buriol, Luciana S., Carlos Castillo, Debora Donato, Stefano Leopardi and Stefano Milozzi. ‘Temporal Evolution of the Wikigraph’, Proceedings of Web Intelligence, IEEE Press, Hong Kong, 2006, http://research.yahoo.com/pub/1662.

Cha, Meeyoung, Hamed Haddadi, Fabrício Benvenuto and Krishna P. Gummadi. ‘Measuring User Influence in Twitter: The Million Follower Fallacy’, Association for the Advancement of the Artificial Intelligence, 2010, http://an.kaist.ac.kr/~mycha/docs/icwsm2010_cha.pdf.

Ciffolilli, Andrea. ‘Phantom Authority, Self-Selective Recruitment and Retention of Members in Virtual Communities: The Case of Wikipedia’, First Monday 8.12 (2003), http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/1108/1028.

Duguid, Paul. ‘Limits of Self-Organization’, First Monday 11.10 (2006), http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/1405/1323.

Ehmann, Katherine, Andrew Large and Jamshid Behesti. ‘Collaboration in Context: Comparing Article Evolution among Subject Disciplines in Wikipedia’, First Monday 13.10 (2008), http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2217/2034.

Emigh, William and Susan C. Herring. ‘Collaborative Authoring on the Web: A Genre Analysis of Online Encyclopedias’, Proceedings of the 38th Hawaii International Conference on System Sciences, 2005, http://www.ufrgs.br/limc/escritacoletiva/pdf/online_encyclopedias.pdf.

Faloutsos, Michalis, Petros Faloutsos and Christos Faloutsos. ‘On Power-Law Relationship of the Internet Topology’, Proceedings of ACM Special Interest Group on Data Communication’, 1999, pp. 251-262.

Farrell, Henry and Daniel W. Drezner. ‘The Power and Politics of Blogs’, Public Choice 134.1-2 (January 2008): 15-30.

Forte, Andrea and Amy Bruckman. ‘Scaling Consensus: Increasing Decentralization in Wikipedia Governance’, 2008, http://dlc.dlib.indiana.edu/dlc/bitstream/handle/10535/5638/ForteBruckmanScalingConsensus.pdf.

Fuchs, Christian. Internet and Society: Social Theory in the Information Age, New York: Routledge, 2008.

Ghosh, Rishab Aiyer and Vipul Ved Prakash. ‘The Orbiten Free Software Survey’, First Monday 5.7 (2000), firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/769/678.

Gruhl, Daniel, David Liben-Novell, R. Guha and Andrew Tomkins. ‘Information Diffusion Through Blogspace’, Proceedings of the 13th International World Wide Web Conference, 2004, http://people.csail.mit.edu/dln/papers/blogs/idib.pdf.

Hars, Alexander and Shaosong Ou. ’Working for Free? Motivations for Participating in Open-Source Projects’, Proceedings of the 34th Hawaii International Conference on System Sciences, 2001, http://csdl2.computer.org/comp/proceedings/hicss/2001/0981/07/09817014.pdf.

Hindman, Matthew, Kostas Tsioutsioulikis and Judy A. Johnson. ‘“Googlearchy”: How a Few Heavily-Linked Sites Dominate Politics on the Web’, Annual Meeting of The Midwest Political Science Association, 2003, http://www.cs.princeton.edu/~kt/mpsa03.pdf.

Kittur, Aniket, Bongwon Suh, Bryan A. Pendleton and Ed H. Chi. ‘He Says, She Says: Conflict and Coordination in Wikipedia’, 2007, http://dl.acm.org/citation.cfm?id=1240698.

Kostakis, Vasilis. ‘Peer Governance and Wikipedia: Identifying and Understanding the Problems of Wikipedia’s Governance’, First Monday 15.3 (2010), http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/2613/2479.

Kottke, Jason. ‘Weblogs and Power Laws’, Kotke.org, 9 February 2003, www.kottke.org/03/02/weblogs-and-powerlaws.

Krafft, Martin. The Debian System, San Francisco: No Starch Press, 2005.

Kumar, Ravi, Jasmine Novak, Prabhakar Raghavan and Andrew Tomkins. ‘On the Bursty Evolution of Blogspace’, Proceedings of the World Wide Web Conference, 2003, http://www2003.org/cdrom/papers/refereed/p477/p477-kumar/p477-kumar.htm.

______. ‘Structure and Evolution of Blogspace’, Communications of the ACM 47 12 (2004): 35-39.

Lakhani, Karim R. and Eric von Hippel. ’How Open Source Software Works: “Free” User-to-User Assistance’, Research Policy 32.6 (2003): 923-943.

Lerner, Josh and Jean Tirole. ‘Some Simple Economics of Open Source’, NBER Working Paper Series, 2002, http://www.nber.org/papers/w7600.

Leskovec, Jure, Mary McGlohon, Christos Faloutsos, Natalie Glance and Matthew Hurst. ‘Cascading Behavior in Large Blog Graphs’, 2007, http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.117.9104.

Lessig, Lawrence. Code Version 2.0, New York: Basic Books, 2006.

Lih, Andrew. ‘Wikipedia as Participatory Journalism: Reliable Sources? Metrics for Evaluating Collaborative Media as a News Resource’, Paper for the 5th International Symposium on Online Journalism, 2004, http://bsunytimes.pbworks.com/f/Wiki_News.pdf.

Marlow, Cameron. ‘Audience, Structure and Authority in the Weblog Community’, International Communication Association Conference, 2004, http://web.media.mit.edu/~cameron/cv/pubs/04-01.htm.

______. ‘Investment and Attention in the Weblog Community’, American Association for Artificial Intelligence, 2006, http://alumni.media.mit.edu.

Metitieri, Fabio. Il grande inganno del Web 2.0, Roma-Bari: Laterza, 2009.

Mockus, Audris, Roy T. Fielding and James Herbsleb. ‘A Case Study of Open Source Software Development: The Apache Server’, Proceedings of the Twenty-Second International Conference on Software Engineering, Limerick: ACM Press, 2000, pp. 263-272.

______. ‘Two Case Studies of Open Source Software Development: Apache and Mozilla’, ACM Transactions on Software Engineering and Methodology, 11.3 (2002): 309-346.

O’Mahony, Siobhán and Fabrizio Ferraro. ‘The Emergence of Governance in an Open Source Community’, Academy of Management Journal 50.5 (2007): 1079-1106.

O’Neil, Mathieu. ‘Weblogs and Authority’, Blogtalk Downunder Meeting, Technology University of Sidney, 2006, http://incsub.org/blogtalk/?page_id=107.

______. Cyberchiefs: Autonomy and Authority in Online Tribes, London: Pluto Press, 2009.

Peters, Parker. ‘Lesson #2: Procedure vs Content, or “You Didn’t Genuflect Deeply Enough”’, Parker Peters Livejournal, 18 January 2007, parkerpeters.livejournal.com/1195.html.

Pfeil, Ulrike, Panayiotis Zaphiris and Chee Siang Ang. ‘Cultural Differences in Collaborative Authoring of Wikipedia’, Journal of Computer-Mediated Communication 12.1 (2006), http://jcmc.indiana.edu/vol12/issue1/pfeil.html.

Shirky, Clay. ‘Power Laws, Weblogs, and Inequality’, posting to Networks, Economics and Culture mailing list, 8 February 2003, http://www.shirky.com/writings/powerlaw_weblog.html.

______. Here Comes Everybody: The Power of Organizing Without Organizations, New York: Penguin Press, 2008.

______. Cognitive Surplus: Creativity and Generosity in a Connected Age, London: Allen Lane, 2010.

Stewart, Katherine J. and Sanjay Gosain. ‘The Impact of Ideology in Effectiveness in Open Source Software Development Teams’, MIS Quarterly 20.2 (2006): 291-314.

Tapscott, Don and Anthony D. Williams. Wikinomics: How Mass Collaboration Changes Everything, New York: Portfolio, 2006.

Viégas, Fernanda B., Martin Wattenberg and Dave Kushal. ‘Studying Cooperation and Conflict between Authors with History Flow Visualizations’, CHI Paper 6.1 (2004): 575-582.

Viégas, Fernanda B., Martin Wattenberg and Matthew M. McKeon. ‘The Hidden Order of Wikipedia’, Proceedings from Online Communities and Social Computing: Second International Conference, Olympia: Springer, 2007, pp. 445-454.

Weber, Steven. The Success of Open Source, Cambridge: Harvard University Press, 2004.

Wu, Shaomei, Winter A. Mason, Jake M. Hofman and Duncan J. Watts. ‘Who Says What to Whom on Twitter’, 2011, http://research.yahoo.com/pub/3386.