The Signpost

Opinion

Why the Terms of Use change didn't curtail undisclosed paid editing—and what might

Contribute   —  
Share this
By William Beutler

For more than a decade, I have worked with clients of all kinds to help improve the quality of information about them on Wikipedia. I began by offering simple changes at my employer's behest, applying my experience as a volunteer, and later started my own digital marketing and PR firm, Beutler Ink, which counts Wikipedia engagement as one of its key services.

Given this background, readers may be surprised to learn that I am not a proponent of "paid editing" as such. I agree very much with editors concerned that COI editing poses a risk to Wikipedia's credibility and reputation for neutrality. That is not to say I've never edited client articles myself, but I have long preferred sticking to discussion pages, making suggestions for volunteer editors to consider on the merits. My last direct edit for a client was in late 2011.

When the Wikimedia Foundation announced its update to the Terms of Use prohibiting undisclosed paid editing in June 2014, I was a prominent advocate. Earlier that same month, I had helped to create a statement signed by ten major PR agencies committing to follow Wikipedia's rules, and I had already been publicly supportive of the WMF's decision.

From my perspective as a guideline-compliant COI editor, the Terms of Use change was the best endorsement that I could have asked for, and the dynamics surrounding disclosed paid editing have since improved. The COI guideline now recommends using the "request edit" template, and a small group of volunteers have largely kept up with the resulting queue.

But in another sense, both the Terms of Use change and PR agency statement that I led have fallen short. Though there is no reliable estimate for how much undisclosed paid editing occurs, it obviously remains a problem. Even Wiki-PR, the company whose undisclosed sock puppet operation inspired both initiatives, simply rebranded as Status Labs and kept right on going.

Indeed, undisclosed paid editing is notoriously difficult to detect and impossible to stamp out entirely. Despite the valiant efforts of volunteers on the COI Noticeboard, it's long seemed clear to me that the whack-a-mole approach is not a solution, but a stopgap.

What should be done now? I am careful to offer suggestions, as I realize some Wikipedia volunteers disapprove even of the disclosed, hands-off style of Wikipedia engagement that I specialize in. Caveat in place, I have years of experience with this topic, and strongly believe there are actionable steps the community and WMF can take to create meaningful change.

First, I believe that many COI contributors do not intentionally set out to break Wikipedia's rules. The problem is that they know little about the rules if they even know they exist, and certainly don't know how to follow them. Of course there will always be bad actors, especially those who view Wikipedia from a purely SEO perspective—*cough* The North Face *cough*—so the COI Noticeboard's role is assuredly safe.

Second, it's a mistake to think COI contributors will ever learn to engage with Wikipedia exactly as the community would like. The rising number of queries on the edit request queue proves that the interest is there, but it's simply too difficult for most outsiders to learn how to ask effective questions. Unfortunately, too many of these requests are TLDR or have unrealistic goals. This wastes volunteer time, and doesn't resolve the underlying issues.

The current edit request system, wherein someone with a COI posts a message on a talk page, affixes a template, and the request goes into a queue for a volunteer to consider at an unknown date according to unspecified criteria, is confusing to uninitiated outsiders and frankly not even that user-friendly for experienced volunteers.

Making matters worse, Wikipedia currently maintains approximately a dozen pages that serve as a possible starting point for advice to COI contributors. Should one follow the "Plain and simple conflict of interest guide" or "Best practices for editors with close associations" or "Suggestions for COI compliance"? Who's to say?

To improve the situation, Wikipedia should simplify the process and meet outsiders halfway. And to be effective, the solution will have to look less like a Wikipedia project page and more like a customer feedback form.

At the very least, the public-facing advice pages should be streamlined and the edit request system turned into a full-fledged WikiProject. Another solution might be the development of a "wizard" tool asking COI contributors to fill out input fields explaining their issue or problem, proposing a solution, and providing links to journalistic sources.

A predictable and transparent process would be helpful not just to COI contributors with realistic goals, but perhaps even more so to those whose requests cannot be satisfied. For them, simply feeling that they were heard and the system worked, even if it didn't achieve the desired result, may soften the blow and reduce the likelihood they will turn to the dark side.

Since amending the Terms of Use, the WMF has mostly shied away from COI topics, yet it still has a role to play here. After all, the WMF is used to working with outside organizations in a way that individual Wikipedians are not. It could appoint a small team to help the community evolve the process, and be a powerful voice to recommend its adoption to the broader public.

Finally, although the PR industry and communications field is no monolith, and the types of COI actors are as varied as the Wikipedia pages that concern them, I can vouch that there is considerable demand for an accessible, reliable means of interfacing with Wikipedia.

Wary as Wikipedians may be to normalize or publicize the fact that outside interests can influence Wikipedia, there's little to be gained by avoiding the obvious. And as much as I agree that direct paid editing is problematic, I hope I can persuade skeptical volunteers that a well-organized system for reviewing and adjudicating COI requests can be part of the solution.

S
In this issue
+ Add a comment

Discuss this story

These comments are automatically transcluded from this article's talk page. To follow comments, add the page to your watchlist. If your comment has not appeared here, you can try purging the cache.
Thanks Doc James, I was not aware that such work was ongoing. Harej, I'd love to learn more—check your inbox! WWB (talk) 15:16, 2 July 2019 (UTC)[reply]
I have now started a discussion regarding the merger of COI guides at Wikipedia talk:Conflict of interest#Merger of COI guides Trialpears (talk) 18:28, 3 July 2019 (UTC)[reply]
Do you find WP:JWG#COI useful, or does most of your COI involvement deal outside of journals? Headbomb {t · c · p · b} 03:44, 1 July 2019 (UTC)[reply]
It is such a good idea that I very rarely need to do any COI work with articles on academic journals. Part of the reason it works here is not just the formulaic nature of the article, but the very clear standard of notability in those fields, which diminishes the need for puffery. (And also, the very close consensus of the relatively few editors here working in this field on the article standards) The practical problem with such articles is trying to find some way of avoiding copyvio and close paraphrase. I think this would be true for many other situations also where a formulaic article could be written properly by anyone even with COI. . I would agree completely with efforts to expand this to other fields, as you suggested above. DGG ( talk ) 07:39, 1 July 2019 (UTC)[reply]
Hi, DGG! You raise a very good point about the current system working well only for simple edits. Alas, I think there is no way around the fact that some pages need more work, and larger changes require more volunteer attention. That said, I agree it is less than ideal for volunteers to make content decisions on topics outside their interest, which is why I hope that a formalized WikiProject could help. Perhaps it could attract subject-matter experts to offer comment, which COI-focused volunteers might help to implemnt.
As to your suggestion that a subject should post a "really informative and well referenced website" off-wiki, can you point to any examples? I believe this was Jimbo's original advice to MyWikiBiz, but it never really became reality, and it raises a whole host of questions, including how to format it, and what CMS to use—which is why I have long used the draft in userspace approach. In any case, there would still need to be a mechanism whereby someone on-wiki brings the issue up for discussion. WWB (talk) 15:17, 2 July 2019 (UTC)[reply]
I agree about the need for different tools for small versus large suggested updates. There's an example of a large suggested expansion at Talk:Medical Journal of Australia, where I manually went through and looked for eh diffs from the current article so that each sentence could be implemented/rejected individually. There could definitely be a tool to allow something like that of large updates or even complete rewrites. It could also give an opportunity for the original coi-editor to give additional references if a sentence was rejected because of lack of supporting refs. T.Shafee(Evo&Evo)talk 01:01, 5 July 2019 (UTC)[reply]
Following up on one of your points, it's would certainly help if we were able to get volunteer editors in business fields--I think we've set the practical notability level for businesses perhaps too high, in an (successful) effort to keep out the incompetent paid editors, and this might enable us give us more realistic coverage. But unfortunately there will still be major gaps, as there are with academic faculty, a field where we have many volunteer editors--such gaps do a disservice to those really notable, leading university PR staff or hired editors to write the articles, which they generally do very poorly, not distinguishing between what counts for notability here and what is puffery. This is similarly true even with contemporary artists , musicians, entertainers--in spite of the very large number of interested WP editors in these fields. So there does not seem any real solution unless we end anonymous editing in some fields, which I certainly do not propose. DGG ( talk ) 16:19, 2 July 2019 (UTC)[reply]
My suggestion would be: better training for writing articles! Imagine a "Wikipedia Article Writing School" supported by the WMF. And if they don't, maybe I will… WWB (talk) 14:37, 3 July 2019 (UTC)[reply]
I attempted to detail that editor dilemma of investing volunteer time in (undisclosed) paid or promotional activity in WP:BOGOF. Widefox; talk 15:03, 1 July 2019 (UTC)[reply]
You just took the words out of my mouth, Widefox. With all due respect, perhaps my very good friend DGG forgot about that, while I nevertheless have extreme respect for the dedication he invests (more than most) in quietly (more or less) going about his COI work. And while we're on the subject, something I always meant to ask you: was your shortcut an intended pun on a less polite we Bits have for 'Go away!' ? Kudpung กุดผึ้ง (talk) 20:10, 1 July 2019 (UTC)[reply]
Good question, and I think Brits may assume there's a message in the name, when there was none intended. It may detract as it was a serious attempt to understand the good faith polarised opinions of dealing with the scale of COI (especially AfDs), and look into the (economic) drivers, so the name just came out naturally. I ended up just considering this a special case of the tragedy of the commons systemic bias, which may be a better name as the bigger topic. I originally liked the focus on the volunteer's choice to be part of a BOGOF offer or not. Garrett Hardin's quote seems apt "Freedom in a commons brings ruin to all.". Widefox; talk 21:30, 1 July 2019 (UTC)[reply]
Widefox, I think BOGOF makes a valuable contribution to the discussion around paid editing, and I certainly recognize the dynamic it describes, i.e. "subsidizing the market" for paid advocacy by throwing good editors after bad. And I'm sympathetic to the idea that if WP:TNT is needed to fix a borderline notable, promotional article, that deletion should be the result more often than it is now. That said, in my reading I find it focuses on enforcement against "unscrupulous, low quality" COI work more than it offers clear suggestions toward encouraging "quality paid editors". I acknowledge the essay is offered as one view, rather than a comprehensive solution. But I wonder, do you think the scenario I describe above is compatible with it? WWB (talk) 15:21, 2 July 2019 (UTC)[reply]
Hypothetically, high quality COI work is not an article quality problem, but that's by definition. What about in practice, though? Is it even a reality? Let's assume the perfect COI editing...there's a big but...it's still a systemic bias. That bias is multiplied when volunteers are drawn in, taking finite editors away from editing other articles, additionally tilting the balance. All undisclosed paid editing is a problem by definition, per the TOU, and that's more work for volunteers to flag and deal with disclosures and problem accounts. The question of a solution to enable/encourage quality COI editing may or may not be directly related to enabling the tragedy of the commons. At least in theory. My understanding of the economics is rudimentary, but there's been success with overfishing. Maybe there's a solution here Elinor Ostrom was awarded the Nobel Prize in economics for demonstrating exactly this concept in her book Governing the Commons, which included examples of how local communities were able to do this without top-down regulations ? Widefox; talk 20:34, 2 July 2019 (UTC)[reply]
As to whether high-quality COI editing is a reality or not, you may not be too surprised to hear that I would hold myself out as an example. There are others, although not as many as I would like, and we tend not to edit client articles directly. Naturally, this requires volunteers to facilitate, but I believe the time they put in does more to improve the encyclopedia than COI/N whack-a-mole or BOGOF cleanup after-the-fact. I think there is also a good case to be made that it's more efficient, and helps to avert "tragic" outcomes. All in all, the theory is interesting, but hard data would be more so. Elsewhere I've begun to encourage independent research along these lines, but it may be some time before any of it comes to pass. WWB (talk) 14:41, 3 July 2019 (UTC)[reply]
I'm not so much interested in the individual merits of one or more editors, edits or articles, more about the collective effect per tragedy of the commons a shared-resource system where individual users acting independently according to their own self-interest behave contrary to the common good of all users by depleting or spoiling that resource through their collective action. . Is is all about the scale of the systemic bias it introduces IMHO. That's an independent issue to that of lack of bias or NPOV due to a COI, which no doubt is possible to achieve, especially via edit requests or other best practice. It's a systemic bias (see WP:BIAS). Widefox; talk 16:01, 3 July 2019 (UTC)[reply]
Hi Bri, finally getting a chance to catch up after the long weekend. I do remember seeing your essay awhile back. I am skeptical that there will ever be community support to prohibit direct editing by COI contributors; I'm reminded of the 2009 RfC which roundly rejected the idea largely for fear of unintended consequences. And in this era of WiRs and other paid non-COI contributors, the line is too easy to accidentally cross. There is also extreme resistance to changing the infrastructure of the site, cf. the decade-long debate over "pending changes". And while Wikipedia's prominence in Google certainly creates an incentive for instant UPE gratification, that is beyond Wikipedia's control. The one thing that might change this would be raising the requirements for editing across all accounts but I'm sure you'd agree that will never happen, either. Instead, I think about how similar bad behaviors were culled in the past, for instance, illegal file-sharing. First iTunes and then Spotify created a different set of conditions, including a reliability and certainty the peer-to-peer networks could never provide, and which users were willing to pay for. It didn't end all illegal file-sharing, but it did put a huge dent in it. If we are concerned about the tragedy of the commons—and I suppose I'm speaking against interest here—perhaps COI actors should have to pay something to get a fair and timely hearing? WWB (talk) 17:21, 8 July 2019 (UTC)[reply]

The ratio of promotional to non-promotional edits seems to be increasing. The proportion of PR money spent on updating public knowledge is growing too. A stable solution should channel some of the desire and funding for such edits into support for tools and editing that monitors the problem, and maintains neutrality and balance. It's good that a bit of that happens via WWB personally, but that's not nearly enough. We need to transform the market for these services by providing an alternative that primarily supports tools for review and CSB, and avoids BOGOF problems. – SJ + 15:08, 8 July 2019 (UTC)[reply]




       

The Signpost · written by many · served by Sinepost V0.9 · 🄯 CC-BY-SA 4.0