Wikipedia talk:Wikipedia Signpost/2015-02-04/Op-ed

Discuss this story

I have not encountered very many examples of paid editing. However, I have had some interaction with a disclosed paid editor who did the right thing and made requests on a talk page for some amendments to the relevant article, as opposed to making the edits himself. In that case, the paid editor was drawing attention to an article of significant importance that really did need a fair bit of work, including work over and above the amendments the paid editor was requesting. Regrettably, however, I still haven't had the time or the inclination to carry out most of that work. Bahnfrend (talk) 01:48, 6 February 2015 (UTC)[reply]

Let's start with the big question - can we stop paid editing? No. It isn't going to happen. Wikipedia's profile is to high, we allow (and encourage) anonymous editing, and the paths to get work are many and varied. Elance and ODesk are two, but there is no way we can track the bulk of paid editing. Even if you ignore jobs not offered on the job sites you still have Guru, Freelancer and Fiverr, to name but three of many. And paid editors such as Morning277 don't generally need to rely on job sites at all. Then there are companies like WikiExperts - how do we identify who they hire, much less who their clients are?

Accordingly, the next question regards what to do. We have three choices - nothing, try to prevent it completely, or find middle ground. We aren't going to be able to prevent it, and we need to try something. So the path the Terms of Use have taken is to find middle ground - allow paid editors willing to disclose a path to continue to receive compensation for editing, while providing justification for acting against those who won't. It isn't a great compromise, but by allowing clients to find editors willing to work within policy we make at least some steps towards limiting the opportunities of others.

Finally, a number of editors actively follow the main paid editing sites. How effective we are is questionable - the more skilled offenders can only have articles identified about 1/3 of the time, although that increases up to 2/3 or more for the less skilled editors. - Bilby (talk) 01:56, 6 February 2015 (UTC)[reply]

Thanks for this great op-ed. Paid editing is a huge risk for Wikipedia. It reduces readers' trust and discourages non-paid volunteers who spend increasing amounts of time, for free, fighting back the hordes of paid editors. It creates bias in articles and conflict among editors. I fear that the very existence of Wikipedia may be in danger in the long term.

In the Spanish-language version, the media has helped us uncover articles clearly authored by public-relation companies, for example about singers and other celebrities (see this discussion). I helped clean up those articles in June 2014 and have been watching them since then. Anonymous editors have come back to all of them and kept on adding "information" and embellishments. I have reverted some of those edits but most are legitimate and I have no way of proving that the editor has been paid, even if I strongly suspect it.

I am now convinced that there will always be editors willing to be sell their skills to companies, that just can't be avoided. What we can do is to deal with the clients: the PR companies and their customers. We need to make them see that paying to edit Wikipedia under cover is unethical, probably illegal and not a good business anyway because most of their work gets reverted or ridiculed. Industry-wide agreements should be signed to ban these practices, the same as multinationals in other sectors now have internal rules that forbid bribing government officials worldwide.--Hispalois (talk) 07:07, 6 February 2015 (UTC)[reply]

Yes this is a problem. Can we do anything about it? Only what we do now - delete the most blatant ones and try to engage the mild offenders. I have noticed however that most illnesses are skewed towards taking drugs instead of diet or lifestyle changes. I guess we want both, and the pharma companies give us one part, while we lack volunteers for the other part. I tend to work in the arts and sometimes it seems that the only paintings worth seeing are in museums. But we are working on the sum of all paintings -- the entire oeuvre of painters, not just what hangs in the largest museums. At the end of the day, most of Wikipedia is good, including the work of paid editors. If anything, we might be responding to the work of paid editors every time we correct peacock terms, something that happens a lot. I think that we need those paid editors to keep us responding, though I can imagine it sometimes can be depressing if you feel swamped by it, like visiting that website. Jane (talk) 08:29, 6 February 2015 (UTC)[reply]

I would like to second Hispalois's praise for this excellent and thoughtful op-ed. I feel both hopeful and resigned about the issue: hopeful, because there is much that is wonderful about Wikipedia, and most of it is unlikely to be affected by paid editing (as there is no benefit to be had on many topics, such as mine, natural history); resigned, as commercial interest is always with us. However, we should not think that nothing can be done. Companies will feel drawn to hiring editors if there is a strong benefit (Wikipedia has many readers) and negligible cost (gee, your hired editor gets blocked, has to try a new sock). We need a sanction which hurts, and there is one: we permanently ban articles on the hiring company. Obviously, this requires proof (to avoid legal action), and equally plainly, that is hard to get. Doc James is right about one thing: the companies that hire editors have plenty of money, and we volunteers by definition don't receive any, so the scales are heavily weighted against us. Still, the existence of a powerful sanction would be useful in curbing the worst excesses, as all companies would have to do would be to disclose their editors. Reputable companies already have no problem with that. Chiswick Chap (talk) 09:00, 6 February 2015 (UTC)[reply]
I am also grateful for this op-ed. The pharmaceutical company/product articles are probably the most frequently highlighted because WikiProject Medicine keeps a very eagle eye on those subjects. However, blatant paid editing is rife here, and unfortunately both WikiProject Organizations and WikiProject Business appear only semi-active, if at all. I have encountered dozens of paid-for articles and successfully seen them through AfD. Sometimes they are so blatantly advertorial that they've been speedy-deleted. When I find one, I check the articles linking and from it and the other edits by the creator and invariably find several more (egregious example). There are two tell-tale signs of paid editing:

The article springs fully formed in the first edit (usually by a "new" user), complete with a impeccably formatted infobox including an image of the company logo or the person if it's a bio.
Multiple, in fact too many, "references"—again impeccably formatted. On closer inspection, they are all press releases, press release-based, listings in company directories, "interviews" with the CEO, "articles" written by the CEO/subject but with the reference disguised to obscure the real author, articles in prominent publications which do not mention the subject at all.

Paid-for articles are significantly different from other types of COI creations, e.g. people writing "do it yourself" articles about their charity, club, business, or themselves. These are usually very messy with poor formatting, bare url references or no references apart from a link to their website, more blatantly promotional (the paid editors are usually more subtle as their goal is to fly under the radar), and often contain large amounts of copy-paste. Both types of articles are detrimental to the encyclopedia, but the paid-for ones are especially pernicious. I'd urge all editors to spend an hour a week checking Special:NewPagesFeed, pick an article that rings alarm bells or that is about a company or product, check it out, and nominate it for deletion if appropriate. If you have time, check the articles linking to or from it, and the contributions of the creator. If all the active editors did this, regardless of our main editing interests (mine is 18th- and 19th- century opera!) we could make a significant inroad into the problem. Voceditenore (talk) 09:57, 6 February 2015 (UTC)[reply]

WMF legal ought to be involved if paid editing breaking the terms of use of the site really threatens Wikipedia's reputation. I wonder if they could at least provide a FAQ about the issue. There is a fairly simple line to take, that paying for articles that are going to be deleted is throwing money away, with reputational risk. Charles Matthews (talk) 11:46, 6 February 2015 (UTC)[reply]

@Charles Matthews: Or stated shorter: what was the point of changing the terms of use if the WMF isn't going to enforce them? Ed ^{[talk] [majestic titan]} 18:54, 6 February 2015 (UTC)[reply]

Shorter, but more simplistic. If A pays B to edit Wikipedia in breach of the terms of use, both A and B are in false positions. B can be blocked (enforcement), but if you want to get at A, you need to combat that otherwise. Charles Matthews (talk) 20:16, 6 February 2015 (UTC)[reply]

That's an excellent point—thank you. Ed ^{[talk] [majestic titan]} 20:18, 6 February 2015 (UTC)[reply]

Entirely predictable result. WMF throws mountains of legal papers. Malicious editors continue business as usual; legitimate editors, scared by the WMF's legal threats and burdens, go away. WMF blames martians for the outcome. --Nemo 11:55, 6 February 2015 (UTC)[reply]

For many years we have had all kinds of problems with COI editing, but the degree to which specific edits are considered a problem seems to be mainly determined by people's political views. The Fluoroquinolone articles were completely taken over for years by people engaged in litigation against Bayer and JNJ. Information was posted to the articles on Finasteride, Isotretinoin, Alendronic acid, Varenicline, and most of the psychiatric drug articles that was well outside of medical consensus, in many cases factually incorrect. In many cases the articles or large portions thereof looked as though they had been pulled directly from the closing statements of a plantiff's attorney in a product liability lawsuit. But these inaccuracies seemed to draw only a very small level of interest or concern.

I think we would do well, especially on the medical articles, to focus on keeping them accurate and the views presented in line with medical consensus, irrespective of whether the source of the inaccuracy is paid editing, litigants attempting to tamper with the jury pool, or simply political activism. Bad medical information is equally harmful irrespective of what motivates the editor. Formerly 98 (talk) 15:21, 6 February 2015 (UTC)[reply]

Articles often become liabilities - sounding boards of negative content - when anyone can edit the article. It's a double-edged sword, not a single-edged sell out. -- GreenC 01:50, 7 February 2015 (UTC)[reply]

The most straightforward thing we could do to put a significant dent in paid editing would be to raise our notability threshold. A rather shocking majority of articles written by paid editors easily meets our notability threshold, and if they employ decent third party sourcing and do not write in an overly-promotional manner, it's almost impossible to delete the articles. I recently had an experience where I'd PRODded an obviously paid-for, barely notable article (but not quite promotional enough to speedily delete it); then a completely different paid editor requested undeletion and was automatically granted the request per our policy. I took it to AfD, and it went the full 3 rounds before there were sufficient comments to get it deleted. Again, the editor who requested undeletion was an incredibly obvious paid editor (his original username was that of the company!), but the "rules" required the article be reinstated. We shoot ourselves in the foot like this on a daily basis. Risker (talk) 02:55, 7 February 2015 (UTC)[reply]

The problem lies in that "paid editors" is too broad a terminology and we really need to distinguish between what are paid advocates and paid contributors, the first group what the this article is identifying as insidious editors who purpose is to restrict and control content being presented. The later group of paid contributors are the legitimate editors who work within sectors like GLAM and their purpose is perfectly aligned to ours making information freely available, to providing reliable citations and sharing media files from their collections. Its possible for a person to edit honestly while being paid for those edits, for the most part its a method that can successfully address some of the gender imbalance issues we read about elsewhere within signpost. What we need is to encourage honesty though identification when a person acts within their employment conditions rather than driving people into hiding with socks. This approach would make it more transarent to our readers and easier for participants to discuss issues in good faith, and it may just have the side effect of less biting of new eds who always appear to be SPA. Gnan garra 03:28, 7 February 2015 (UTC)[reply]

Yes this piece is about paid advocacy rather than paid editing. Many organizations share our goals and if they pay someone to help us achieve these shared goals there is no issue with this. Doc James (talk · contribs · email) 00:25, 10 February 2015 (UTC)[reply]