Wikimedia’s proposed Terms of Use update

We could have cages, so diners are protected from each other. Who knows what they might do, sitting there with knives? That isn’t a good way to design a restaurant, to live our lives in society. --Jimmy Wales, 2005 Image (c) Nicholas Boudreau, used with permission.

“We could have cages, to protect diners from each other. What might they do with their knives? No- that isn’t a good way to design a restaurant, or to live our lives in society.
paraphrase of Jimmy Wales, 2005
Image (c) Nicholas Boudreau, used with permission.

In February, I argued in a Wikipedia Signpost op-ed that a proposed change to Wikimedia’s Terms of Use (TOU) was less than ideal. The proposal would require of all editors of Wikipedia, and all other Wikimedia projects, that they disclose any paid conflict of interest in at least one of three ways. This is, of course, a practice that Wiki Strategies has always supported (at least, respecting Wikipedia); but the specific amendment is problematic in several ways.

The Wikimedia Foundation’s Board of Trustees will be considering it this week. The amendment’s passage may be inevitable, given the strong community support (837-286). [UPDATE: Please see the insights offered below by Nemo, who questions the appearance of community support] I still think it would be a mistake; not a huge, sky-is-falling mistake, but one that (in addition to having some positive effects) will add confusion to an already complex area, rather than helping us move toward a resolution.

The main point that concerns me is that, if there is a clearly articulated requirement, and it addresses only one of many things that should be done by an organization’s staff in order to approach Wikipedia ethically, many will draw the mistaken conclusion that compliance is sufficient to assure an ethical approach. I do not question whether organizations will comply — many will, which is a good thing — but rather, I am concerned that they will be emboldened by their (mere) compliance, in ways that threaten Wikipedia in the long run.

This may seem like a very subtle point, but it’s a subtle point with very real consequences. Several recent events, in which an organization or entity has complied with the proposed rule but erred in other important ways, underscore my points. These examples would be problematic regardless of their origin; but since each is related to the Wikimedia Foundation itself, they provide a strong insight into the kind of behavior that can result when too much emphasis is placed on the specific provisions reflected in this amendment. (I also have some involvement with these examples, which is evident in the links provided.)

Let’s take a look.

Belfer Center Wikipedian in Residence project

As previously covered on this blog and elsewhere, a recent project by Harvard’s Belfer Center was designed in a way that has been broadly acknowledged as flawed. That project, however, does meet the narrow conditions of the proposed amendment; it would not have constituted a violation. The Wikipedian in Residence did disclose his identity and role on his user page. But the way he approached his editing did not tend to support readers’ ability to see his influence; he did not initiate or participate in the kind of discussion that can support healthy organizational engagement with Wikipedia. He has stated that he did not receive much guidance on Wikipedia editing; an email from Wikimedia Deputy Director Erik Moeller clarifies that he was advised against editing the article about the Belfer Center.

But when an entity like the Belfer Center is editing articles about international relations, the conflict of interest relating to the organization is hardly the most significant concern. To illustrate this point: in March 2014, the Wikipedia article on U.S.-Russian relations, one of the 63 articles edited by the Belfer Center’s Wikipedian, was viewed 30,000 times. The article on the Belfer Center itself was viewed 700 times.

Writing an autobiography on Wikipedia

Writing an autobiography on Wikipedia is not prohibited by policy, but there is strong consensus that it shouldn’t be done — and that if it is done, there should be clear transparency around it.

But last week, an email discussion thread explored questions around the Wikimedia Foundation’s former Chief Community Officer’s autobiography. Foundation executives asserted that there is no organizational policy around Wikipedia editing, and pointed out that there had been some disclosure. But that disclosure was not meaningful; it apparently went unnoticed for four years. When this issue was brought up for community consideration, a tag was promptly added to the article, informing the reader that the article may have problems relating to a conflict of interest.

Again, this is an instance of conflict of interest editing that would not have literally violated the proposed amendment, but that is considered problematic by the Wikipedia community.

Voting early and often

Somewhat related to the example above, the article on the New Organizing Institute (NOI) was nominated for deletion, due to a lack of independent media coverage. The article was clearly out of compliance with Wikipedia’s norms, and was promptly deleted; but not before four voters, whose comments strongly suggest but do not disclose an organizational affiliation, voted to have it kept. Fortunately, Wikipedia’s processes are resilient, and decisions are not made purely by majority rule; so even if 20 people had showed up with the same message, they would not have been successful.

Wikipedia’s insistence on this kind of disclosure — even in the absence of an updated TOU — is often the reason that widespread efforts to distort Wikipedia’s content is uncovered. Wikipedians often discuss the possibility that “sock puppets” or “meat puppets” are being employed to give an organization undue influence over its coverage on Wikipedia. In one of the more prominent instances, the Wikimedia Foundation sent a cease-and-desist letter to an organization that makes this a regular practice.

But when commenting on the issue, one of the founders of the NOI pointed a finger at the Wikipedia community, but not at the staff of his organization. I am not sure whether or not these edits would qualify as “paid edits” under the proposed TOU update; but if not, we have a third example of how the update would fail to protect Wikipedia from organizations editing in secretive ways in order to advance an editorial objective.

A better path forward?

Instead of, or in addition to, the TOU amendment, Wikipedians should consider creating broad standards — as contrasted with specific rules — that would clearly express broadly held values. The Wikimedia Foundation could help in important ways — but its role is not a necessary one.

I think my Statement of Ethics provides an example of the kind of document that would be most helpful. There could be a standard document, but creating it and addressing everybody’s concerns would be a huge task. Instead, it would make sense for individual Wikipedians — especially those who have worked with organizations, like Wikipedians in Residence — could write up their own Statement of Ethics documents, and engage in ongoing efforts to compare and contrast them. In addition, organizations who care about Wikipedia should create policies to guide their personnel’s approach to Wikipedia. Publishing these policies for public review would be ideal, but maybe not necessary in all cases. The Wikimedia Foundation would do well to lead the charge.

There is a field that has done this with some success: journalism. Below is the introductory section of the Society of Professional Journalists’ statement. But that document only represents a collection of individuals; it carries no special authority. Rather, its power is in the common features it shares with many, many other organizations’ similar documents.

Members of the Society of Professional Journalists believe that public enlightenment is the forerunner of justice and the foundation of democracy. The duty of the journalist is to further those ends by seeking truth and providing a fair and comprehensive account of events and issues. Conscientious journalists from all media and specialties strive to serve the public with thoroughness and honesty. Professional integrity is the cornerstone of a journalist’s credibility. Members of the Society share a dedication to ethical behavior and adopt this code to declare the Society’s principles and standards of practice.

  • Seek truth and report it […]
  • Minimize harm […]
  • Act independently […]
  • Be accountable […]

Society of Professional Journalists

About Pete Forsyth

Pete Forsyth is the principal of Wiki Strategies, and a Wikipedia expert. Full bio here:
This entry was posted in governance, paid editing, Statements of Ethics, Terms of Use, wiki, Wikimedia Foundation, Wikipedia, Wikipedian in Residence. Bookmark the permalink.
  • Nemo

    “Strong community support”? I don’t see any such thing, is there evidence?

  • Nemo, thanks for asking — I probably should have linked that in the post (and I will). By the raw numbers, I count 790 in favor (or 837 if you include “support, but should be stronger”) and 286 opposed. I think these votes were likely driven by a desire to do *something, anything* about the increasing amount of blatant paid editing. I don’t discount the intentions of those casting the votes, but I do wonder if this strong majority has fully thought through the consequences. I am wary about singing my own praises, but in this case, I do think the depth of my experience in guiding organizations’ approach to Wikipedia does give me clearer insights into the rather complex dynamics. Anyway, for the vote tally, see here:

  • nihiltres

    “Instead of, or in addition to, the TOU amendment, Wikipedians should consider creating broad standards […]”

    One of the reasons that I support the ToU amendment is that it strongly implies the “broad standards” that ought to be in place, and may, IMHO, catalyze their formation as on-wiki policy.

    Wikipedia’s paid-editing debate is really about ethics: consequentialist (duty to good content) vs. deontological (duty to good process). Consequentialism seems to me to be dominant on Wikipedia; the standards that we’d like to see in place are deontological. If the Terms of Use amendment is implemented, it gives deontological policies a foothold.

  • Nemo

    Pete, I never bothered to gauge the consensus in that discussion, as it always looked to me like an unduly blown-up drama show, i.e. best ignored. However, one thing is sure: the vote count of the most heavily canvassed “discussion” page in Wikimedia history doesn’t mean anything at all.

    I won’t even measure how low was the average knowledge of Wikimedia projects of the voters (example vote “DONT MAKE US PAY!!!”); everything happened in English, and most votes are from unregistered users, so it’s totally clear that the vote count doesn’t, in any way, represent the global Wikimedia projects consensus on the matter.

    Even a mere count with shows a very different picture. After a basic anti-socks removal of a) voters with less than 100 global edits (250+100 in favour vs. 150 not) or better 500 (241 vs. 127), or b) of those registering after the vote started (185 vs. 97), or better c) both (124+54 vs. 90 or 113+40 vs. 85): it’s clear there may be a majority, but not a super-majority. Even if majority mattered. It’s simply not serious to claim any of the proposals has strong community support based on vote counting.

    After removing voters, it’s a mere 57+22 vs. 50 (over 100 edits) or 45+14 vs. 43 (over 500 edits), a ridiculously low number of votes to even start considering the idea of affecting any project other than with the outcome (you don’t even elect a steward with such numbers; perhaps a global sysop).

  • Nemo, you make some very strong points. Thank you for your analysis.