The Signpost

Opinion

Trying to find COI or paid editors? Just read the news

Who hires paid editors to write articles about themselves or their businesses? Who dispenses with paid editors and just writes articles about themselves as conflict-of-interest (COI) editors? What difference do these editors make to our readers? The easiest way to get answers to these questions is just to read the news and look for people who have been caught promoting themselves.

A scientific scandal

Sapan Desai, a vascular surgeon, had a terrible year. Last fall he was sued for medical malpractice three times, and in February he and his employer parted ways - though his former employer won’t say why. Just when things started to look much better, his life really went downhill. His coauthors and he looked like they scored magnificently, getting scientific papers on COVID-19 accepted at two very prestigious medical journals, The New England Journal of Medicine and The Lancet. All of these papers were based on an unbelievably large dataset provided by Desai’s own company Surgisphere. But soon The Guardian, The Scientist, and many other news sources began questioning the reliability of the dataset. The two accepted papers were both withdrawn in one of the biggest scientific scandals in years. One theory is that the dataset was just made up out of thin air.

The Guardian reported that an English Wikipedia article about Desai had been deleted in 2010. The deletion debate centered on puffery and unbelievable claims apparently written by Desai himself. Note that an investigation about paid or COI editors conducted purely on Wikipedia can almost never completely prove the identity of an editor; for example an editor can impersonate another (a "Joe job") in an attempt to embarrass them.

The Wikipedia article on Desai was begun on January 20, 2007 by User:OverlordvI, a single purpose account who knew many details of Desai’s life, such as the name of the consulting firm Desai founded in high school, the name of his family’s private charity, his SAT score (a perfect 1600), all three of the degrees he earned at the University of Illinois at Chicago Medical School and related campuses, the 16 medical textbooks that he had written or co-written and was selling on Amazon, and that he had earned a law degree (JD) – all by age 27 – just as he was starting his medical residency at Duke University.

Some of this unbelievable history checks out. He was selling textbooks on Amazon, and he did earn three degrees (BS, MD, PhD) from the University of Illinois at Chicago. Other details could likely only be confirmed by Desai himself or by his family. There is no evidence that he was ever a lawyer or earned a JD, not even in his 43 page resume now posted online.

A new user named User:Surgisphere began editing in 2008, the same year a company of the same name, owned by Desai, was founded. The article on Desai was deleted on July 27, 2010 following a detailed nomination by User:Crusio.

Did the Wikipedia article on Desai do any actual harm to our readers? We should not blame all of the extensive harm caused by Desai’s recent publishing misadventures on the Wikipedia article. But we did give Desai’s early publishing career a boost, if only in advertising his 16 textbooks. We likely caused harm to those medical students who bought those texts, and perhaps to their future patients. Allowing this type of article to exist on Wikipedia for over two years certainly diminishes our reputation as a reliable source.

The Ronaldo of investment banking?

A second possible COI/paid editing situation this month was published online in Efinancecareers in the story ‘’Andrea Orcel and the importance of having a positive Wikipedia page’’. The story’s author makes all the proper disclaimers that the actual identities of the Wikipedia editor involved cannot be fully identified. Andrea Orcel is a highly respected businessman who has been called the “Ronaldo of investment banking” according to his Wikipedia article, which is corroborated by at least one person. Orcel was also apparently offered the CEO position at Santander Bank, one of Europe's largest commercial banks. But the offer was withdrawn, or the deal couldn’t be completed, and Orcel is now suing Santander for 120 million euros.

The online story reports that User:MAaR11Aa 2019, whose 29 edits on Wikipedia are all to the Andrea Orcel article, has claimed twice in edit comments that “we are restructuring Andrea's Wikipedia profile to make it simpler and more organized. it is at the request of Mr Orcel.” (Both on April 15, 2020)

The Orcel article was created by User:Gallic Village in September 2018 and expanded by User:One Factor. Both of these accounts were indefinitely blocked as sock puppets of DonSpencer1 and, based on the glowing additions made, were likely paid editors. DonSpencer1 was recently unblocked and has started editing the Orcel article under that username. Taken together the users Gallic Village, One Factor, DonSpencer1, and MAaR11Aa have accounted for roughly 87% of all edits to the article and 94% of the added text.

Does this type of editing where the rich and famous appear to hire others to write a glowing article hurt anybody? It certainly hurts Wikipedia's reputation for fairness, accuracy and reliability.

+ 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.
  • Unfortunately (or, fortunately, depending on your viewpoint) it only takes one editor to publish a BLP but it takes about a dozen editors to get one deleted. I'm involved in one AfD biography right now where the page creating editor has persistently challenged the deletion votes and has managed to get the discussion extended another seven days for further consensus. And if there is no clear consensus to delete the default will be "keep". In a nutshell, it takes a lot of time an effort to remove these paid editing biographies. Yes, vanity biographies do indeed hurt Wikipedia. When somebody's name is ran through a web search their Wikipedia biography is usually a top result. Instant credibility. I'm not sure what is the solution. Perhaps a separate BLP patrol from the general New Page Patrol? Blue Riband► 00:24, 29 June 2020 (UTC)[reply]
  • Wikipedia is not a reliable source. davidwr/(talk)/(contribs) 21:21, 29 June 2020 (UTC)[reply]
    • @Davidwr: I certainly don't trust a Wikipedia article on a small or medium sized business or on a business person who is not well-known by the general public. There's way too much advertising for me. If I want advertising, I can just go to the company's website. At least that way I know for sure who is piling on the BS. If a majority of editors felt the same way I do, Wikipedia would have to quit accepting those types of articles and spend a tremendous amount of time cleaning up the old ones.. The disclaimer that Wikipedia is not a reliable source, is not however, an excuse for us to accept such garbage. Smallbones(smalltalk) 00:21, 30 June 2020 (UTC)[reply]
      • @Smallbones: Despite the garbage, these articles can be useful for canny readers who can discount the puffery and seize on any critical points that are made, especially if the subject has been convicted of a crime or been officially censured. Of course, one worries that many readers may not be canny in that way. How much do we know about how people read and use Wikipedia? Mrmedley (talk) 01:30, 1 July 2020 (UTC)[reply]
        • @Mrmedley: I don't know how much we really know about our readers. Somewhere there must be readers by country for e.g. Enwiki. I assume it's fairly similar to editors by country, which has some info referred to here (with more readers than writers proportionately from countries where there are many people who speak English as a 2nd or 3rd language, since it's a lot easier to read than write a foreign language. Somewhere I think there is a survey of readers, but not so many as to tell much about them. If anybody has specifics on our data on readers, please let us know. But beyond that all i can say is the very obvious - they are a very diverse group. Probably ranging in age from 10 to 100 (I know a 90 y.o. editor and a long time ago I met an editor of maybe 12 y.o.). Probably including anybody who can afford a computer and internet connections (and quite a few who can't who use school and library computers) as long as they have a minimum literacy level. Just incredibly diverse.
        • Given that, many readers can probably read between the lines and know BS when they read it. Others will be taken in by slick PR presentations. One touchstone is some work I did on binary options articles. IIRC an article like Banc de Binary was getting something like 300 page views per day, in all language versions, but they where spending huge amounts of money or staff time for paid editing, e.g. $15,000 for a short/medium term contract was reported on-Wiki. The thing to remember is that they only needed a few victims to scam. Press accounts start at an estimated average (or median) $20,000 loss per victim. So if they only got 3 new victims per day through Wikipedia, they were still taking in very good money. So if only 1% of our readers are not sophisticated enough to see through an obvious scam like binary options, we're still doing a great disservice to our readers.
        • Sure most readers can see through most scams like that, but that is no excuse for letting those types of article in the 'pedia. Other types of PR or even just simple bias have their own costs to our readers. Smallbones(smalltalk) 04:00, 1 July 2020 (UTC)[reply]
          • I don't think there is evidence been presented that canny readers can see the difference. I think it is the reverse, although I don't have any research to back it up, more so from my own experience. The reason that so many people are scammed on their own computers, is that the scammers responses fit the expectations of the scammed, and even those you expect to notice it, those in IT and IT security, still get scammed and it is the same kind of experience that is happening here. The type of articles that are being written by these folk, since about 2008, are getting more and more fluid in their response. It is a natural evolution, and it gets commensurately harder to spot them. I suspect that the number of folk that see them for what they are and extract value from them, is very small and those that do, are in that industry. I think the majority takes an honest approach and reads as it is. For me it is a regular occurrence to work on articles that have are subsequently speedied, Afd'd or whatnot, that I thought were genuine.scope_creepTalk 07:47, 21 July 2020 (UTC)[reply]