2010 Wikimedia Study of Controversial Content

From Meta, a Wikimedia project coordination wiki

The following is part of the Draft Study on Wikimedia Controversial Content by Robert Harris and Dory Carr-Harris. Comments on the draft or questions are highly encouraged, and should be placed on the talk page, but for now please refrain from revising the draft directly. Also, feel free to read the archived version of this page, including the original FAQ.


From almost the inception of the projects, questions of controversial, potentially-objectionable content have been a heated topic of discussion among Wikimedia editors. For years, significant and thoughtful discussions about this issue have taken place either on individual article pages, on policy pages, on general Foundation discussion groups, and specific project forums.

Different events, historically, have triggered these deliberations, and in April of this year, discussion of this ongoing issue was given an external stimulus by (as it turned out) erroneous news stories about controversial images in Commons and a heated exchange within the community about appropriate treatment of these images.

These events suggested to the Board that a more comprehensive review of the entire issue of controversial content on the projects might be well worth considering at this point in the organization’s history, a review different from those internal discussions of these issues of the past. Consequently, in late June, the Board passed a resolution instructing the Foundation’s Executive Director to commission a study to look at the whole issue of “potentially-objectionable content” on Wikimedia projects. The authors of that study were instructed to survey experience at other, similar Internet sites, to look at international experience in this area, to survey the opinions of librarians, Internet censorship experts, family-advocacy groups, and many others, as well as studying carefully the history of deliberations around these issues within the Wikimedia community. This document is the preliminary draft of that study’s final report.

As we approached our work, we were guided by the following principles, among others, as stated in the preamble to the Board resolution:

We do expect material in our projects to be educational in nature, and any material that is not educational should be removed. We see our role as making available all knowledge, not solely such knowledge as is universally deemed acceptable. We believe that individual adults should be able to decide for themselves what information they want to seek out. In the case of children, we believe that their parents, teachers, and other guardians are best placed to guide them to material that is appropriate for them, based on their development and maturity, as they grow into adulthood.

Nevertheless, we are concerned about the possibility of people being exposed to objectionable material that they did not seek out. This may include material that is violent, sexually explicit, or otherwise disturbing; culturally offensive depictions; profane or vulgar language; depictions of potentially dangerous activities; and exposure of children to material that may be inappropriate for them. We believe that the Wikimedia projects are a valuable educational resource, and we do not want these issues to interfere in sharing knowledge with present or future readers.

The Wikimedia Foundation Board of Trustees believe that the Wikimedia projects have a serious responsibility to carefully balance these interests to the best of their ability. This includes considering the interests of both adults and children, as well as understanding different cultural perspectives about what material may be offensive. It is a difficult challenge, and we do not take it lightly.

Since June, we have been engaged in carrying out the study commissioned by this resolution. We have interviewed and continue to interview many people within and outside of the community (these interviews are not yet complete – a list will eventually be presented) including experts in international censorship, representatives of other similar big-service sites, representatives of family and online safety groups, librarians, various Wikimedia community members, and people interested in the issue representing different cultural perspectives. We have surveyed studies and reports from jurisdictions all over the world that touched on various aspects of our mandate. Given the time and resources at our disposal, we cannot claim to have been exhaustive in this work, but we feel we have received a fair, representative sample of opinion in the world surrounding these issues.

We have also deliberately and necessarily engaged with you – the Wikimedia community. We have had email conversations with several of you who reached out to us, we spoke to many of you at Wikimania, we posted two sets of questions, whose answers appeared on a page devoted to the study, and, most especially, we have studied as carefully as we could the vast archive of discussions surrounding the issues around “controversial content” on the projects that go back almost as far as the birth of Wikipedia itself.

We are now at the point, as mentioned above, where we are ready to share with you the conclusions we have started to formulate, as we make final preparations to present the study to the Board in early October. We are not completely finished our work, because the feedback we receive from you at this stage is important for us in fashioning our final suggestions and recommendations to the Board. In many ways, you are the most significant experts with whom we’re interacting, because your knowledge of and experience with these issues is deep and comprehensive. We need and welcome your input.

It is also worth noting that neither us or the Board have any intention of changing the manner in which policy is made within Wikimedia projects. Here is an FAQ that accompanied the announcement of the study:

What do you say to those editors who believe that editorial policy is their purview, not the responsibility of the board or the staff?

We agree with editors who say that, and we believe that Wikimedia's current methods of developing and enforcing policy, for the most part, work really beautifully. The Wikimedia projects are a shining example of the power of mass collaboration, and nobody wants to fundamentally change anything about how the projects work.

Therefore, it is important to realize, in the material to follow, that this is a report we are presenting to you, the community, based on the traditional autonomy you have had in developing and enforcing policy. We’ll be getting into a level of detail that is not appropriate for the Board to deal with (and won’t be presented to them), but which is appropriate for you.

What we‘d like to do is present the study to you in three or four sections, pausing for a couple of days between each to collect responses and feedback. Although the complete study is full of many specific practical suggestions, we thought we would start in this first section by enumerating the basic principles from which we drew our conclusions. By and large, these are principles internal to Wikimedia that we believe lie at the heart of the projects presently, and should continue to do so. Although our mandate included us looking outside of the projects for information, guidance and experience, time and again, we were struck by the singularity of Wikimedia’s mission in the world. So we thought it appropriate to begin looking inward, with a survey of the basic operating principles that we believed should be the foundation of everything we recommend. As we begin, we would once again like to thank all of you who have weighed in on what we realize is a question long debated within Wikimedia circles. It is highly appreciated and has made our work much easier, and we hope, more effective.

Part 1: Basic Principles

As we began looking at the basic principles that we believe animate the Wikimedia experience, we noted that some of these principles are well-known and axiomatic within the Wikimedia community (although they bear re-assertion). Others, we believe, are no less widespread and essential to the community, but not quite as explicitly and as often articulated. It is the combination and interrelationship of these principles that we believe created the intellectual foundation of the territory we were asked to explore.

1. Wikimedia is Dedicated to Intellectual Openness

In the first category is the overriding principle that animates all Wikimedia efforts – the unrelenting, unremitting and rigorous commitment to non-censored openness and “intellectual freedom” (to borrow a term from the library community) that Wikimedia attempts to provide for the world. The belief in providing open information and complete knowledge to the world’s inhabitants is not only a slogan for Wikimedians – it is the principle that animates virtually every decision and activity taken over its many platforms every day. We have been told time and again as we conducted this study that Wikimedia’s commitment to intellectual freedom is not merely its mission statement – it is the key to its success in the world. The more open the projects can be, it was argued, the greater their potential success. By and large, we agree with this perspective. Wikimedia projects are trusted in the world because they are seen to be fair and unbiased (to the extent that that is possible in a world filled with contention). Open and full access to information is a principle that is both intellectually sound and practically efficient.

2. Wikimedia's Openness Does Not Change When Content is Contentious

We will note, with regard to this principle, that the manner in which this commitment to openness is executed by Wikimedia projects sets it somewhat apart from other sites also devoted to wide global use and extensive circulation of information. (Google, Flickr, YouTube,e.g.) As an educational enterprise, Wikipedia’s commitment to intellectual openness is no means to an end – it is an end in itself. It is a public service, the reason the projects exist. Consequently, Wikimedia has been less willing than others to accede to extra-institutional pressure to change its content. (This, to us, is the true meaning of the oft-quoted “Wikimedia does not censor”), Wikimedia’s educational mission forces a certain uncompromising attitude onto its projects. There is much in the world to learn about – not all of it is pleasant, not all of it is uncontroversial, some of it is disturbing, some of it is hard. But it is there, and by and large, Wikimedia is there to document it. The principle of radical openness belongs to Wikimedia’s intellectual DNA.

Does that mean that we are recommending that Wikimedia projects should not be restricted in any way, that the current status quo be retained surrounding controversial content ? Our answer is no—we are recommending an improved regime around controversial content on Wikimedia projects in this study. And the reason for that is that we believe there are another set of principles equally ingrained in the Wikimedia DNA, equally central to the projects’ success, and imperative to recognize and act upon.

3. Service to the Public is also a Key Wikimedia Value

These principles can be grouped under the heading of public service and respect. Every great educator respects both the material he or she is pledged to disseminate, and the students with whom that material is to be shared. And, as a great educational institution, respect for the reader is another key overriding, currently existing Wikimedia principle. Sometimes that core principle is explicitly expressed, but more often implicitly animates behaviour and practice on the projects. Sometimes it is evidenced in the injunction to “do no harm,” sometimes as an injunction for civility amongst editors themselves, sometimes as cautions of “not bringing the projects into disrepute,” sometimes in discussions about what might be appropriate content for highlighted content pages. But, however it is expressed, deep within the purpose of Wikimedia is a central notion of providing a unique, valuable service to the people of the world. It is solely to serve those citizens of the world that Wikimedia exists, its entire raison d’être, the reason that hundreds of thousands of individuals have voluntarily given billions of hours of their time to the projects over the years. The notion of respect for our audience, for the importance of service to the public as a goal in itself, is as much part of Wikimedia’s DNA as its commitment to intellectual freedom. Not as often, or as explicitly expressed, it nonetheless is another key to the success Wikimedia has had in the world. Care is taken within Wikimedia to present its content with the needs of the audience uppermost in our minds. Our audiences respond to this care – it brings them repeatedly to our sites.

4. Respect for the Audience Will Increase in Significance

It is our observation that this principle of public service needs to be more explicitly understood by Wikimedians as the projects move forward and more explicitly presented to the world as one of the defining features of its sites. Wikimedia, like many other websites of its stature, has been the beneficiary of a remarkable spurt of growth since it began in 2001. Almost without realizing it, the world has come to depend on Wikimedia projects in hundreds of countries, millions of times a day. Simply by following their internal policies and procedures, Wikimedians have been able to produce a powerful instrument for enlightenment in the world, almost without impediment. However, that power is now established, and the consequences of it must be reckoned with. The growing acknowledgment of the power of the Internet in general, and its great, powerful institutions in particular, of which Wikimedia is most certainly one, must force upon Wikimedia a heightened sense of the responsibility that comes with that power. Again and again over the past few months, we have been told by international free-speech watchdogs that a new wave of Internet censorship is spreading throughout the world, a backlash, if you will, against the enormous, seemingly unregulated power of the Net. Wikimedia projects cannot assume that they will be immune from this backlash. It is our view that the best defense against these forces is a clear commitment to public service by the projects, a commitment to a principle, which, as noted above, already exists as a foundational pillar of the projects. Wikimedia is eventually dependent on public acceptance and support of its mission, practices, and procedures for its continued healthy existence.

It is important to note that the two sets of principles we’ve enunciated – commitment to intellectual openness, and service to the public – are hardly mutually exclusive. In fact, by pursuing our mission of intellectual openness most thoroughly and completely, Wikimedia projects have always served the public best. The two principles – intellectual openness, and public service – almost always go hand in hand within the projects. The first by and large guarantees the second.

5. When these Basic Principles Collide, Policies Must Mediate the Differences

However, there are a very small number of cases – a minority of a minority – where that compatibility of principles breaks down, where increased intellectual openness threatens to reduce our public service, rather than increase it. Maybe it’s when we publish images considered sacred to one religious group or another. Maybe it’s when we present unprotected content of sexual behaviour or violence to children without warning. Maybe it’s when we present shock images on publicly-highlighted pages. In effect, the study we have been undertaking is made up of this minority of a minority – where the principles of openness and service grate against each other, and seem to need some lubrication. The question we must ask and answer is : Is that lubrication possible without destroying the very principles on which the enterprise is founded?

We believe it is – and not only is it possible, but it is necessary. When intellectual openness and public service can be brought again into harmony, the projects are immeasurably strengthened. Not only do we not weaken the projects by compromising the principle of intellectual freedom in those rare cases where it is necessary to do so – we protect the projects from unnecessary exposure to risk with these actions, and fortify the Foundation’s position in the world. Fortify it at a time when that position, as mentioned above, might be increasingly challenged as never before. The rest of this study will be devoted to an exploration of how the accommodation of these two principles might be accomplished to produce this result.

6. These Policies Must Begin With Openness Tempered by Service

Our basic operating principle for creating the recommendations that will follow is this one. We believe that intellectual openness is the principle with which we must begin – it is the primary principle, the bedrock of the projects, which is to be leavened only when needed by the principle of public service if accommodation is needed at all. In other words, we see openness as the primary default position for Wikimedia, to be modified only in the most restricted of cases, and then, only in the most restricted of ways. As you will see in the installments that follow that deal with more practical questions of content management on the projects, that principle – of least possible modification of the principle of intellectual freedom – is one we have attempted to follow. Where you feel we have not done so, please let us know. If there are alternative, less intrusive ways to achieve the goals we have set for ourselves, we would be delighted to hear about them.

As well, we believe the instances when modifications to intellectual openness should be considered must be severely limited to protect the overwhelming amount of content on the projects that must remain unmodified, especially that content, as mentioned above, that is demanding, difficult and contentious by its very nature. Because language matters, it is for this reason that we prefer the term “controversial content” to “potentially-objectionable content.” Objectionable content is, obviously, content to which any individual user takes issue, for whatever reason. Far from lacking an objective test, the designation of objectionable to content is always immediately apparent. If I nominate it as such, it takes on that designation. Who can argue the assertion that I take offense at pictures of balloons (if I am phobic about them), or descriptions of the mating habits of mammals. No one. Allowing “objectionable”, or, even worse “potentially-objectionable” as a trigger to determine potential modifications to Wikimedia content is, in our mind, much too broad a definition whose adoption would be dangerous for the intellectual health of the projects.

“Controversial,” on the other had, to us, implies a social process, an acknowledgment that certain types of content (say, images of explicit sexuality) generally create a reaction among broad groups of individuals, each acting independently, and without ulterior motive, that gives us more confidence that we might consider these reactions legitimate and worthy of consideration. There are objective tests for the concept “controversial,” we believe, whose use can ensure the legitimacy of consideration by an open Wikimedia culture of potential modifications to that openness.

7. Wikimedia Projects serve the Information Needs of Individuals, Not Groups

As well, we believe, it is important to note as essential the principle that Wikimedia projects exist to serve individuals, as individuals, in their full autonomy, and consequently, the projects, as a general rule, do not and should not consider as legitimate censorious demands by institutions, of any kind, political, commercial or voluntary claiming to represent those individuals, or making demands, which, in the community’s opinion, represent only their own interests. Our relationship and responsibility is to maintain our core principles in the name of and to serve individual users of our projects. That relationship should be maintained at all costs.

So, to summarize this first theoretical installment of the study, we have enunciated seven distinct principles that animated our work, and would be interested in your reaction to any or all.

  1. Wikimedia projects have and must continue to have as their fundamental goal devotion to intellectual openness and the widest possible dissemination of information.
  2. That goal must be pursued even when the dissemination of that information is demanding and contentious.
  3. At the same time, Wikimedia projects are, and must continue to be devoted to the principle of public service, and respect for our audience. The projects cannot be successful if not presented in a way that enhances their use and connection with their publics.
  4. This principle of public service and reader acceptance will increase in significance as the Internet in general, and Wikimedia projects, specifically, become more powerful and influential in the world. Increased attention to this principle is of utmost importance as the community moves forward..
  5. In the rare instances where these two principles collide, policies must be created to ease the contradictions, and allow both principles to co-exist, in somewhat modified form. In these instances, one principle should not be allowed to overwhelm the other – the need for respect cannot eliminate the need for openness, the need for openness cannot eliminate the need for respect.
  6. In the formation of these policies, the basic procedure should be to begin with the principle of openness, and modify it only as much as is needed to preserve respect and service. The default position is full openness, to be modified as little as possible in as few instances as possible.
  7. Service by Wikimedia projects is to individuals, not organized groups. Therefore, restrictions to intellectual openness will not as a rule be made when demanded by institutions claiming to represent a collection of members, whether those institutions are political, commercial, or voluntary. As well, they will generally not be made when demanded by single users representing their own uniquely individual point of view. They will be made, when they are, in response to clear and documentable responses by large numbers of individual users, unmoved by ulterior motives, when deemed legitimate under terms and conditions determined by members of the Wikimedia community.

Part Two

2010 Wikimedia Study of Controversial Content: Part Two contains more precise definitions of what constitutes controversial content and provides a set of initial recommendations by the authors.

Part Three

2010 Wikimedia Study of Controversial Content: Part Three contains no recommendations, but instead has general observations about the continuing conversation about controversial content.