The Free Encyclopaedia that Anyone can Edit: The Shifting Values of Wikipedia Editors

Wikipedia is often held up as an example of the potential of the internet to foster open, free and non-commercial collaboration. However such discourses often con-flate these values without recognising how they play out in reality in a peer-production community. As Wikipedia is evolving, it is an ideal time to examine these discourses and the tensions that exist between its initial ideals and the reality of commercial activity in the encyclopaedia. Through an analysis of three failed proposals to ban paid advocacy editing in the English language Wikipedia , this paper highlights the shift in values from the early editorial community that forked encyclopaedic content over the threat of commercialisation, to one that today values the freedom that allows anyone to edit the encyclopaedia.


Introduction
Wikipedia is an encyclopaedia in transition. Its core values are being called into question as an increasing number of users are paid to contribute to the encyclopaedia. How then is the open editorial community of this free encyclopaedia responding to the increasing presence of commercial interests and paid editors? Through an analysis of three failed proposals by the community to impose bans or limits on paid editing, this study reveals how the values of the English language Wikipedia editorial community are in transition and how these shifts reflect wider changes in assumptions about commerciality in digital media.
Throughout its history Wikipedia's status as a non-commercial, non-profit, top web property among commercial counterparts has often seen it being praised for holding all the promises of an open and democratic web. In this discourse, debates about freedom, openness and non-commercialism often get conflated or neglected in favour of celebratory accounts of collaboration. However, it is important to recognise that there are different logics at work in each of these narratives and this paper aims to untangle the threads of freedom, neutrality and commercialism to investigate how ideals around the collaborative production of knowledge online are changing and how within Wikipedia there has been a shuffling of the community's values.
Once the threat of commercial activity in Wikipedia and the ability to derive a profit from the unpaid labour of others prompted a volunteer walk-out, known in Wikipedia folklore as the Spanish Fork. In response to suggestions in 2002 that Wikipedia may take advertising, Spanish language volunteers forked their content to other servers and started a new encyclopaedia (Lih 2009;Tkacz 2011). Now the presence of paid advocates -those editors who gain financial benefits from editing Wikipedia articles on another party's behalf -has drawn a public response from the Wikimedia Foundation, its then Executive Director Sue Gardner and founder Jimmy Wales (who has always been a vocal opponent of PR involvement in the encyclopaedia) (Wikipedia 2012;Roth 2013;Wikimedia Foundation 2013b). However, the community response has been divided. It is interesting to analyse these divisions, along with the involvement of different actors and groups in Wikipedia to see how paid advocacy has been constructed and how it reflects a separation of the values of openness and freedom and a shift away from the ideals of earlier contributors to the encyclopaedia. Indeed, as Wikipedia is reconfiguring its norms and values, this analysis reveals important truths about how the boundaries between the commercial and the non-profit in the context of peer production are sometimes fuzzy, overlapping and far from clearly defined. earlier encyclopaedic efforts based on Enlightenment ideals is the desire to make available the totality of knowledge. In Wikipedia, this ideal is expressed as providing access to the 'sum of all knowledge,' and this similarity between Wikipedia and earlier efforts has contributed to the experiment becoming an extremely successful global encyclopaedia (Wikipedia 2013d). Indeed as Benjamin Mako Hill found in his study of failed encyclopaedias, one of Wikipedia's strengths is that despite being online, it still largely resembles a traditional encyclopaedia (Garber 2011).
However these encyclopaedias of the past have been of a momentary nature, taking 'snapshots' of information (Yeo 2001) at different points in time. On the other hand, Wikipedia, which is popularly criticised (Sanger 2006) for its information being unstable and transient, is perhaps the only encyclopaedia to aggregate these 'snapshots' to construct a history of a particular subject over time. As each edit is logged and timestamped, Haider and Sundin (2010) note that in Wikipedia, 'permanence has reached a new height…Everything is constantly changing at the same time as it is always being saved and stable, archived.' Indeed, encyclopaedias are important in exemplifying the ideals of a period in history, of capturing intellectual consensus and establishing the knowledge of the time. These 'snapshots' provide an insight into the current ideals around free and open access to knowledge, and in Wikipedia's case the potential of the web to be a forum for this knowledge. Ideals concerned with the greater social good are a historical feature of encyclopaedias as debates about property and copyright have played out since the early 18 th century, just as they play out about commercialisation in Wikipedia today (Yeo 2001;Loveland & Reagle 2013). Analysing how Wikipedia has responded to paid editing can therefore reflect wider popular feeling about commercial activities on the web.

Untangling the Threads: Peer Production and Collaborative Knowledge Ideals in Reality
Events in the platform's history, such as the Spanish Fork, suggest that Wikipedia has long been opposed to commercial involvement and values its place as a neutral non-profit. As Wikimedia Foundation spokesperson Jay Walsh notes in relation to paid editing in the community, 'there's a historical resistance towards it from early days within the project' (Mullin 2014). Now however, the encyclopaedia is negotiating how to maintain its ideals in a web environment where commercial players inevitably want to be involved in producing content for a top six website (Alexa 2013). In order to examine what and how things are changing, we must first look back at the ideals in question -freedom, neutrality, and commercialism -and how they have been conflated in imaginings of Wikipedia in utopian discourses of peer production.
The popular discourses (Benkler 2006;Leadbeater 2006;Tapscott & Williams 2006;Bruns 2008;Shirky 2008) around peer production, collaboration, prosumption and produsage normally invoke Wikipedia as a separate entity from market forces and portray its users as contributing due to a commitment to free and open knowledge. Attributing these motivations and ideals 'fits neatly with the longstanding rhetoric about the democratizing potential of the internet, and with the more recent enthusiasm for user-generated content (UGC) [and] amateur expertise' (Gillespie 2010: 352). Indeed Wikipedia is often situated as part of a gift, or sharing economy that operates differently to traditional market forces (Benkler 2006;Lessig 2008).
These narratives also suggest that one of the key aspects of peer production and co-creation is collaboration, where amateurs and/or volunteers work with traditionally commercial content producers in a mutually beneficial relationship. Indeed as Nathaniel Tkacz notes about these discourses, 'Collaboration is literally everywhere and can be attached to almost anything, immediately giving it a positive value' that is 'beyond that of simply co-labouring' (Tkacz 2010: 41-42). Tkacz (2010) also notes that there is a gap between popular and romanticised accounts of collaboration with how projects such as Wikipedia actually operate in an attempt to enact ideals (Kittur et al. 2007;Matei & Dobrescu 2010;Halfaker, Kittur & Riedl 2011;Laniado & Tasso 2011). This process of enacting ideals is ongoing and the encyclopaedia is in transition as both a knowledge producer and web platform. Tarleton Gillespie notes of web platforms: Like the television networks and trade publishers before them, they are increasingly facing questions about their responsibilities: to their users, to key constituencies who depend on the public discourse they host, and to broader notions of the public interest. (Gillespie 2010: 348) Like other online platforms Wikipedia is a socio-technical construction that has evolved through a negotiation and formation of rules by the community. From its founding ideals Wikipedia has developed in a political context where ideals and principles scaffold the construction process (van Dijck 2013). This 'nonprofit, nonmarket business model that Wikipedia has chosen is inimically interwoven with the volunteer-based peer-production system the platform so successfully implemented' (van Dijck 2013: 148), and commercialism in this environment is consequently a controversial subject.
So while scholars like Benkler have given us a romantic view of Wikipedia as being based on peer production, on a system somehow apart from the commercial market, this is not in reality the case (Tkacz 2010). Websites are highly interconnected and this connected nature means that Wikipedia inevitably includes commercial actors. Indeed, sustainability in this environment is linked to a platform's ability to integrate content across multiple places and spaces on the web (for example Wikipedia's Facebook entries (Park 2010) and translation project with Google (Galvez 2010;van Dijck 2013).
Additionally, being conflated with other online platforms, being something other than an encyclopaedia, may reveal why Wikipedia is seen as open slather for so many marketing professionals. In using the term 'platform', which Gillespie (2010) points out is a politically charged term, we can see how it can be appropriated as a marketing 'platform,' or conflated with other 'platforms' that offer up marketing opportunities (such as YouTube, Facebook, Twitter), or indeed how Wikipedia may be packaged as part of a larger online media campaign. The difference is Wikipedia to many of its contributors and readers outside the PR sphere, is a platform for advocating the value of, and providing, free and open knowledge. This is the fine line that Wikipedia straddles between an encyclopaedia and a platform, between an institution and a community. Where an encyclopaedia has an established tradition, a platform is still being negotiated. Whereas an institution is compromised of rules, a community is a more ad hoc assemblage where members can come and go freely, and it is in this context that Wikipedia is trying to negotiate the values associated with peer production and the creation of a volunteer-led online encyclopaedia, and what commercial involvement means for its future sustainability.

Wikipedia's Core Policies as an Expression of Ideals
Wikipedia's ideals are linked to its non-profit business model (van Dijck 2013), and as an organisation free from commercial pressures it is perceived as 'free' to create neutral and objective knowledge. Setting the conditions for what Wikipedia is and its core policies -its five pillars -reflect these ideals of freedom and openness.
Of Wikipedia's five pillars, neutrality is arguably the most venerated (Greenstein & Zhu 2012;van Dijck 2013). It is the ideal to which editors aspire, a truly fair and representative article. While the possibility of this may be challenged by those editors who consider knowledge a social construction (Matei & Dobrescu 2010), it is still upheld as a core policy by most Wikipedia editors. And this ideal of the community to produce truly neutral, information is tested by the presence of paid advocates within the editorial community.
Advocacy by paid editors, in Wikipedia, is the antithesis of neutrality. The promotion of one position over another is seen as against the ideals of free and representative information. It would follow therefore that the community (which has been so good at constructing rules and norms in the past to regulate behaviour (Halfaker et al. 2012)) would want to create a policy to prevent such contributions. However the three proposals and associated votes to form such a policy all failed to achieve the support of the community. This is despite the Wikimedia Foundation (WMF) sending a cease and desist letter to the organisation found to be engaging in extensive paid editing that resulted in widespread media coverage (Wikimedia Foundation 2013a). Such a move by the WMF, presumably not only in response to some sections of the editorial community, but also in response to the threat to their brand, shows that how Wikipedia is perceived (as a hub of neutral information) to groups outside of the editorial community is equally as important as how it is constructed by the community. In this discourse in the mainstream press, paid editing is being constructed as an issue that undermines the integrity of the encyclopaedia and is against its core operating principles of freedom and openness.
In order to preserve (or perpetuate the idea of) Wikipedia's neutrality, Jimmy Wales has often called for a 'bright line' where PR professionals should never edit directly in article space, that is -any contributions they want to make or issues they want to raise should first be raised on the talk page of the relevant article an then escalated through existing channels, without ever editing any article content directly (Wikipedia 2012). In line with this, the UK's Chartered Institute for Public Relations has published the Wikipedia Best Practice Guidance For Public Relations Professionals that is based on Wikimedia UK's own draft guidelines (Chartered Institute of Public Relations 2012). However the response from some PR quarters to this approach is that it can be too slow and cooperation with editors can be difficult (Distaso 2012), leading to the conclusion that many professionals will indeed edit in article space.

The Case of Wiki-PR
A consulting business, Wiki-PR is behind one of the biggest covert editing efforts in Wikipedia's history. Banned by the community after a community-led investigation and discovery of its activities, Wiki-PR claimed to have 12 000 clients and employ Wikipedia administrators as part of its operations (Owens 2013). Rather than going through the traditional channels and protocols that Wikipedia has established for editors with a conflict of interest (posting to the talk-page, requesting an article for creation), Wiki-PR used experienced editors familiar with the policies of the site and able to negotiate the rules and norms to ensure that the articles survived the creation process. Employees created 323 fake accounts, called sock puppets, to create and contribute to pages about clients. This large-scale astroturfing resulted in several hundred articles on Wikipedia that were largely promotional in nature, and that were removed following the investigation (Owens 2013).
However the legacy from such activity remains, not only in the widespread press accounts of Wiki-PR's actions, but in how Wikipedia has positioned itself in response to the revelations of the extent of the sockpuppeting activity. At an institutional level, the WMF expressed concern that its brand and reputation as a nonprofit site of independent knowledge had been damaged by Wiki-PR's activities: The Wikimedia community of volunteer writers, editors, photographers, and other contributors has built Wikipedia into the world's most popular encyclopaedia, with a reputation for transparency, objectivity, and lack of bias. When outside publicity firms and their agents conceal or misrepresent their identity by creating or allowing false, unauthorized or misleading user accounts, Wikipedia's reputation is harmed. (Wikimedia Foundation 2013c) This event therefore demonstrates the gap in English language Wikipedia between norms around commercial involvement and actual practices. For at a community level, the response has been less decisive, reflecting the shifting values of the Wikipedia community members as they engage in discussion to define and construct paid advocacy editing and its position in Wikipedia's landscape of volunteers, paid editors and public relations professionals.

Methodology
In order to map the debates, I conducted a grounded analysis of the three main votes on paid editing conducted in the community in November 2013. These discussions formed one response to the Wiki-PR revelations and are a discrete object through which to analyse immediate user feeling in relation to a well-publicised event that challenges the encyclopaedia's ideals. It is theoretical sample, chosen to illuminate a specific response to a specific controversy rather than be a representative sample of the entire Wikipedia editorial community.
Using a Constructivist Grounded Theory (CGT) approach to the problem of mapping how the editorial community of Wikipedia is responding to the increasing presence of commercial interests and paid editors allows for new themes to emerge through the coding process that may not be reflected in dominant responses from other places. Grounded Theory works well when applied to online discussions such as these as it allows for quickly 'gaining a clear focus on what is happening in your data without sacrificing the detail of enacted scenes' (Charmaz 2006: 14). Kathy Charmaz notes that, 'like a camera with many lenses, first you view a broad sweep of the landscape. Subsequently, you change your lens several times to bring scenes closer and closer into view' (Charmaz 2006: 14). Such a close reading of all three votes revealed divisions in the community about supporting measures to limit or ban paid editing as proposed. However it also revealed the justifications offered by editors in the conversations often aligned as editors seek to negotiate what paid editing actually is.
As mentioned above, the institutional response from Jimmy Wales, Sue Gardner and the WMF was definite in its opposition to paid advocacy editing, reflecting the assumption, based on past actions that the community is against such involvement. However in favouring an open approach to the coding the data using CGT methods, a more nuanced response from the community emerged from the conversations, one that did not necessarily always fall in line with the institutional reaction.
CGT therefore offers insight into how the ideals of Wikipedia are changing as the internet changes around it. In describing these debates, the study reveals the tensions that compromise paid advocacy editing and how Wikipedia's founding principles are interpreted by those who edit the encyclopaedia a little over a decade later. As Geert Lovink (2011: 1) points out, 'The participatory crowds suddenly find themselves in a situation full of tension and conflict,' and these situations can reveal much about how platforms and collaborative projects are evolving.

Three Proposals
The three proposals analysed here are 'No paid advocacy' (NPA), 'Paid editing policy proposal' (PEPP) and 'Conflict of interest limit' (COIL) (Wikipedia 2013b;Wikipedia 2013c;Wikipedia 2013a). The three discussions and votes were carried out on English language Wikipedia in November 2013 in response to the Wiki-PR controversy. Remaining open to all possible understandings of the data, I undertook a four-stage coding process to 'separate, sort, and synthesize these data through qualitative coding ... [and]...emphasise what is happening in the scene' (Charmaz 2006: 3). Overall, 573 posts were analysed in the study. The first stage consisted of an initial round of coding where each response was coded as a support, oppose or comment along with short description of the post. In a second close reading both the posts and the descriptor and a list of key words was formed. In the third stage the key words were refined to a set of categories, and then finally each post was assigned relevant category tags. In total there were 21 categories to emerge from the discussion, ten that opposed the formation of a policy, nine that supported a new policy and two that were neutral (for example where votes either supported or opposed the policy, but called for a clarification of the policy wording).
There was a relatively large number of participants with 300 individual contributors to the discussions and proposals regarding paid editing on Wikipedia. Among the three conversations NPA was the largest vote and involved 256 individual participants contributing 408 posts, PEPP had 86 participants contributing 242 posts and COIL was the smallest discussion with 43 participants contributing 74 posts.
All three discussions were linked by an 'infobox' on each page stating that, 'In November 2013, there were three main discussions and votes on paid editing' along with a link to the other two discussions (Wikipedia 2013b). 22% of users contributed across these different discussion spaces, 16% who participated in two of the conversations about the proposals and only 6% contributed across all three discussions. Additionally in the collaborative tradition of the few doing the most, a small number of users contributed heavily to the discussions. In NPA the ten most frequent commenters contributed 16.9% of the posts, while in PEPP and COIL, the top ten contributed 49.2% and 51.4% of all posts respectively (although this was often just short replies to votes, rather than involved discussions among users).

'We are at the Barricades'
The first, and most obvious result is that all three proposals failed. Despite much debate and discussion across a variety of spaces both on-wiki and off, and the swift formation of the policy proposals, all three failed to garner enough support via the votes to effectively ban paid advocacy editors by way of a formal written policy. It became apparent in analysing the discussions that 'free' does not necessarily correlate with 'free from commercial interests' and that remaining open to contributions from all editors, paid, volunteer or somewhere in between, is more important than creating more regulatory mechanisms to assist in the production of quality, neutral content. Therefore one of the major themes to emerge from the analysis was that editors felt existing policies in Wikipedia already cover the issues raised by paid advocacy editing, the two most cited being neutrality and notability. Neutrality is expressed as an impartial point of view where articles are written from a fair and representative position (Wikipedia 2014b). Notability guidelines outline the criteria under which a topic is considered significant enough to have an article in the encyclopaedia (Wikipedia 2013e). The most common response from users to the proposals reasoned that the application of these existing policies would weed out the edits made by someone with a conflict of interest, and an additional policy is not necessary.
Further, advocacy of any sort as a motive doesn't really address edit quality. Only application of existing Wikipedia guidelines does that.
One of the issues here is our incredibly low notability standards….Sorry, but I think we need to clean up our own act before we create policies that will be used primarily to gain advantage against opponents in ideologically-based editing. (Wikipedia 2013b) The alternative view from supporters of the proposed policies, is that an explicit, new rule is needed. One that specifically bans paid advocacy editing so that a message is sent to editors that this type of commercial activity is not welcome in the encyclopaedia. Supporters maintain that traditional non-profit organisations are required to have policies on conflicts of interest and Wikipedia should be no different.
Wikipedia needs a clear, written policy on financial COI [conflict of interest], like every other major non-profit. We owe it to to [sic] ourselves, and to the public that trusts us, to get this done. I don't want to explain to my grandkids (if I ever have some) that I stood by and watched while this great experiment of our [sic] was inundated by a tsunami of commercialism. We are at the barricades, let us not back down. You have to decide if I am crying WOLF or, is the wolf at the door, here, now. (Wikipedia 2013b) While some participants outlined their support of such a policy because paid editing is against the ideals of Wikipedia, another group of editors opposed such a policy saying that preventing paid editors violates Wikipedia's core premise -that it is the encyclopaedia that anyone can edit.
Be clear and honest here, this policy change won't eliminate paid editing and COI, it will hide it. It's a deterrent to honesty and a line right through the Wikipedia slogan 'The encyclopaedia that anyone can edit'. Thanks …and yes as always the original foundation of wikipedia remains 'Welcome to Wikipedia, the free encyclopedia that anyone can edit.' If it is principles that you want I would start with, 'If it ain't broke don't fix it', followed by not eroding the two basic principles of 'Attack content not editors' and 'The encyclopedia that anyone can edit'. (Wikipedia 2013b) The last quote raises an interesting point, and one put forth by a number of editors who discussed the norm of focusing on the quality of edits as opposed to the type of editor making contributions. This reason was often given in conjunction with an oppose vote to the formation of the proposed policy, also citing existing policy as being sufficient to address the issue of paid editing.
We have policies and guidelines for how articles should be written and developed. We have built up the project to focus on the content not the contributors. (Wikipedia 2013b) This was a recurring theme among users, that a fair and accurate encyclopaedia article can be achieved by addressing the quality of the edits, not the people contributing the content. There was also the view among editors that such a policy would be unenforceable and create extra work for already over-burdened volunteers who would be required police it.
Highlighting the gap between institutional and community response only one editor referenced Wikipedia's reputation in the discussions, which the Wikimedia Foundation cited as a reason to cease and desist in its letter to Wiki-PR. Also, only one comment called for institutional involvement in this issue, suggesting that overall the community sees this as an issue it can manage itself.
Another challenge to forming an explicit policy against paid advocacy editing is that the community is still not clear about what constitutes paid editing. It can be taken to mean anything from a museum employee updating information about an artefact in their collection, or a funded graduate student contributing in their area of expertise to paid professionals who are editing for a third party to advocate a particular point of view.
Also, no one anywhere on this project has ever clearly defined the differences between 'paid editing' and 'paid advocacy', and until definitions exist then discussions probably cannot proceed. The working definition is that 'paid advocacy' is 'paid editing' which does not comply with Wikipedia community guidelines. All discussions on this topic make no sense to anyone outside this movement because advocacy in the Wiktionary sense of the term has nothing to do with its use in this small community on Wikipedia. (Wikipedia 2013b) What constitutes a conflict of interest, and indeed what threat editors with conflicts of interest pose to the encyclopaedia is still very much up for discussion in the community. It demonstrates a shuffling of values among different editors as to the place of commercial players in the Wikipedia ecology. Interestingly where commercial involvement was once viewed by the community as being in direct opposition to Wikipedia's core values (and this rhetoric is repeated at an institutional level) and should be prevented, some community members now accept the presence of paid professionals and are resigned to their presence in the encyclopaedia.
Dishonest paid editors will do it anyway, so why punish the honest ones? Or drive them to dishonesty?
We can strongly discourage paid editing but not ban it. We should try to work with the COI editors to develop a lasting relationship, not declare all out war. (Wikipedia 2013b) There are therefore values more important to the community than whether or not an editor is being paid, and these relate to the encyclopaedia's existing standards of notability, verifiability and most importantly neutrality. Participants expressed the need to differentiate between the different types of paid editing and that as long as the editorial pillars of Wikipedia are held up, the issue of whether or not someone has a commercial interest in editing Wikipedia is secondary to them holding up these ideals. 'Wikipedia is the flagship of peer production and the most celebrated open content project' (Tkacz 2010). It is the free encyclopaedia that anyone can edit and this ideal is valued by Wikipedia contributors over and above remaining free from commercial activity. The reality that Wikipedia is no longer (if indeed it ever was) free from commercial involvement, is one that many editors are resigned to. Rather than take an ideological stance against paid editing like the Spanish Fork, editors are willing to find ways to mange it based on existing ideals of neutrality and openness.

Conclusion
While debates continue to play out in the English language Wikipedia about paid editing, in other language versions, working arrangements have been reached with those editors who are paid to write for the encyclopaedia. In the German language encyclopaedia (which is the third largest version behind English and Dutch) companies can edit through a verified account (Wikipedia 2014a). Similarly advocates for paid editing from Wikimedia France welcome the input of corporate editors as they see it as improving articles that would otherwise languish and to keep information relevant and up-to-date (Wikimania 2013).
In line with this more open approach from other Wikipedias, the English language Wikipedia community is responding to the increasing presence of commercial interests and paid editors by favouring the ideals of openness and neutrality over freedom from commercial involvement. It is looking at ways of defining and regulating this involvement, but not in any way that would impede the ability of anyone to edit.
For the popular discourses about peer production that hold Wikipedia up as an ideal of free, open, volunteer-led, non-commercial activity, no longer hold in an environment where companies will want a presence on one of the world's most popular websites. And while the Wikimedia Foundation and founder Jimmy Wales are drawing bright lines around paid advocacy editing, the Wikipedia editorial community is taking steps to manage commercial involvement by looking at the variations of paid editing as they 'seek to strike a balance between stability and open-ended flexibility' (Coleman 2013: 208).
English language Wikipedia editors are still negotiating and constructing paid editing. Indeed as the nature of the web is changing and commercial activity is more overtly evident across other platforms, some editors seemed resigned to commercial activity in the encyclopaedia (Song & Wildman 2013). The question is then not how to prevent commercial involvement from paid editors (such as through the policy proposals discussed), but how to manage it. In reconfiguring their values from earlier editorial communities, editors are reflecting the changing nature of the web and separating out the values of openness, freedom and noncommercialism into a workable model that upholds the central ideals of neutral and objective information in an encyclopaedia that anyone can edit.
Kim Osman is a PhD candidate at the ARC Centre of Excellence for Creative Industries and Innovation at the Queensland University of Technology. She is currently investigating the history of Wikipedia as a new media institution. Kim's research interests include regulation and diversity in open environments, online volunteer communities, and controversies in the history of technology. E-mail: kim.osman@qut.edu.au