The Signpost

In the media

Wikimedia LLC and disinformation in Japan

Enterprise API

Wikipedia Is Finally Asking Big Tech to Pay Up by Noam Cohen in Wired broke the biggest story of the month about Wikimedia Foundation's for-profit subsidiary Wikimedia, LLC and its intended product, Wikimedia Enterprise, more commonly known as Enterprise API. While there are now over 100 news stories on the topic, Cohen is almost the only journalist who directly quotes WMF staffers.

Google and Wikipedia, according to Cohen, have long had an unspoken partnership. Google provides Wikipedia with readers, Wikipedia provides Google with content. The idealists on Wikipedia, according to Cohen, have been providing the content for free to the "rapacious capitalists" at Google and other Big Tech firms. Now the WMF is trying to charge Big Tech for faster access to the content. This more business-like relationship should result in more money to the WMF, and better service to Big Tech. Other users of the data dump of all articles, which can be downloaded every two weeks, and of the recent changes "fire hose", will still be able to use those formats for free. Those companies who need faster access, better service, or customized data feeds will pay for the privilege.

Big Tech's money should stabilize the WMF's cash inflows and allow it to pay for expected expansion in the global south. But there's a problem in getting too much money from Big Tech – the WMF could become dependent on that cash. Wikipedians will be exposed to commercial influences from our new business partners and will need to trust that they don't take advantage of us. Cohen concludes "One can only hope that [the community] finds partners worthy of such faith."

Disinformation on Japanese Wikipedia

Non-English Editions of Wikipedia Have a Misinformation Problem by Yumiko Sato in Slate The Japanese Wikipedia gets the second most page views after the English Wikipedia, but its coverage of some topics, especially those related to World War II, is lacking according to Sato. The systemic problems are exemplified by the coverage of well-documented human experimentation by Unit 731 during the 1930s and 1940s, which are described as "a theory". The Nanking massacre is described as the Nanking incident. That article contains few period photographs, with none of horrific ones that are displayed on the English Wikipedia.

The Japanese article on comfort women – women held in sexual slavery by the Japanese army – uses the Japanese word for "prostitution". Sato states the "article is not designed to inform the readers but to confuse them and cast a seed of doubt in their minds." She proposes several causes for the systemic problem, including Japanese cultural and linguistic isolation, but stresses the relatively few Japanese administrators, and the strong control they exercise.

An earlier and longer version of the article is available at yumiux.com and a video of Sato presenting Disinformation on Japanese Wikipedia (13:50) is available on Vimeo.

Videos and podcasts

In brief



Do you want to contribute to "In the media" by writing a story or even just an "in brief" item? Edit next months's edition in the Newsroom or leave a tip on the suggestions page.
+ Add a comment

Discuss this story

It's not just the Japanese Wikipedia. The Croatian Wikipedia's issues are well-known, and on Azerbaijani Wikipedia you can read a charming article about the "alleged Armenian genocide, also known as the Armenian reloction"[1] and another article about the "Genocide of Armenians against the Turkish-Muslim population in Eastern Anatolia"[2] (which did not occur according to mainstream genocide studies). Turkish Wikipedia's coverage is a bit better but not all the way there[3] I suspect that for languages spoken mostly by members of one ethnolinguistic group their Wikipedias tend to reflect the inbuilt biases of that group. (t · c) buidhe 23:33, 28 March 2021 (UTC)Reply[reply]

Yes, the article by Sato specifically mentions the Croatian Wiki and her explanations for the Japanese Wiki fits in with what you say. Smallbones(smalltalk) 00:13, 29 March 2021 (UTC)Reply[reply]
  • One positive aspect of the UCOC I look forward to is the WMF actually maybe feeling confident to simply remove the racist and nationalist lies on the smaller Wikis and ban the people responsible for them. -Indy beetle (talk) 04:36, 29 March 2021 (UTC)Reply[reply]

Haaretz commented last month that "Therefore, the articles on the Israeli-Palestinian conflict are very different in English, Hebrew and Arabic." Gråbergs Gråa Sång (talk) 09:08, 29 March 2021 (UTC)Reply[reply]

There are also similar issues on English Wikipedia. They tend to arise either in articles of special interest to a particular - generally fairly small (in a global sense) - geographic audience, to which a particular narrative appeals (and in respect of which others may have no opinion at all, often due to ignorance of the subject matter), or in articles about a subject on which speakers of English as a first language (or residents of the Anglosphere) tend to have a particular, shared, point of view that may be accepted as mainstream and unquestioned by such persons, no matter where they come from, because it is shared by so many of them. In the case of both types of article, the issues are probably more difficult to spot than on Wikipedias in less prominent languages. Bahnfrend (talk) 13:44, 29 March 2021 (UTC)Reply[reply]

At this point, the only thing that is going to solve the whitewashing problems on the smaller Wikipedias is if the WMF themselves intervene, which is long past due. X-Editor (talk) 18:42, 29 March 2021 (UTC)Reply[reply]

I recommend anyone reading Sato's article also check out Saebou's response (in Japanese) to an earlier version of the article. It looks like most of the glaring errors she pointed out have been rectified in the Slate article (such as a failure to distinguish between full and semi protection), but one that remains is the idea that the few admins "have power over what goes on in the platform". No one wants to become an admin because it subjects you to intense scrutiny and harassment with so little gain. I would attribute the prevalence of historical revisionism less to admins' exercise of power than to the lack of it. The idea of retaining and growing the editorbase and admin corps as a challenge to the project is pretty much nonexistent on Japanese Wikipedia, and the community there tends to be hellbent on protecting itself by chasing away problematic good-faith editors rather than guiding them in the right direction. Nardog (talk) 15:48, 31 March 2021 (UTC)Reply[reply]

  • I think, Sato's observation very accurately pointed out the problem of Japanese version of Wikipedia. Saebou's comment looking away from the reality of such problem and jumping on Sato's mistake which don't matter with the existance of the problem. It even distorts the facts.
Saebou explained the problem is derived from the shortage of admins, despite recently the already few admin of Japanese version has been dismissed for unreasonable reasons like User:Bellcricket or User:Triglav, further reducing the number. The view that "no one wants to be an admin" is contrary to the fact. Right-wing groups and religious right-wingers want to gain managerial status and to parge their enemy of faith acting in Japanese version. Currently, the Japanese version of the management group is already occupied by whom allowing right-wing propaganda works. They rather promoting such propaganda.
It appears to the fact of Higashi-Ikebukuro runaway car accident continues to be unable to add to the article Kozo Iizuka. Kozo Iizuka was the representative of the spiritual group Tōkyūjutsu.[1] Information hiding work by group members had become a hot topic in the Japanese media immediately after the car accident, such as the website of the group being deleted immediately after the car accident. That was one of the main reason why Iiduka was ciritisized, as like as the police didn't arrested Iiduka.
Saebou explains that the reason why the fact of car accident cannot be added is, because the Japanese version takes a strict stance against personal defamation. But in Japanese version, there exist many articles which contains gossips. It is a welcome and double standard. The fact that the content reported in WP:Rs cannot be written on Wikipedia violates "freedom of expression".
And Saebou NEVER mention to the religious background of Iiduka and to the action of groups menbers in Japanese version. In reality, the reason why the article cannot be written in Japanese version is that some managers allowing and cooperating to the violation of the policies and guidelines of the people involved in religious groups (WP:MEAT). The current state of the Japanese version of Wikipedia is perceived as abnormal in Japan, and it undermines the credibility of Wikipedia.
Saebou argued to solve the problem if new experts participate in the editing. However, experts have already participated in the editing and are blocked from unreasonable reasons such as "Tired the community". Don't bring in new experts, unblock old experts.--UikiHedeo (talk) 05:36, 7 October 2021 (UTC)Reply[reply]
@Hexirp: Thank you for your comment. But I already read what Kitamura = Saebou wrote, and I think Kitamura =Saebou's comment looking away from the reality of such problem and jumping on Sato's mistake which don't matter with the existance of the problem. Please read my bad English composition again. Thank you. --UikiHedeo (talk) 08:33, 25 October 2021 (UTC)Reply[reply]

Notable omission

It is mentioned in that piece without naming the author, the Wikipedia editor involved, or the persons with whom that editor has a conflict of interest (according to the Arbitration Committee). The Daily Dot article was not linked. The Arbitration Committee motion was also curiously absent from the arbitration report, although such things are normally covered. Mo Billings (talk) 17:51, 29 March 2021 (UTC)Reply[reply]
I noticed that ArbCom mentioned something about "using editorial discretion", but that goes without saying - every sensible writer uses their discretion.You use your discretion, I use mine. Smallbones(smalltalk) 20:52, 29 March 2021 (UTC)Reply[reply]
@Smallbones: So this was not an oversight but a deliberate editorial decision on your part to not report on a motion by the Arbitration Committee? Mo Billings (talk) 21:41, 29 March 2021 (UTC)Reply[reply]
@Mo Billings: Please understand that I'm not trying to discourage you from writing stories about coi editing. I likely have different ways of going about it or different goals, but that's normal for different publications. I will discourage you from trying to tell us what we're supposed to publish. About 3 people have done that over the last couple of years, and they all get the same answer - we're an independent newspaper and make our own editorial decisions. If you'd like to submit a piece to The Signpost' on coi editing, please email it to me. Note that I won't reprint something from Wikipediocracy. There's too much baggage there. They've outed too many people or otherwise unfairly attacked too many people.
As far as the ways we go about writing a paid editing story (you can't always neatly divide into COI/paid but I usually prefer paid) I look for classic newsworthiness, e.g about real public figures. An editor that nobody outside of Wikipedia has ever heard of usually won't work for me. It's better if it's about some issue that's known outside of Wikipedia - not some internal grudge match. But other publication have other ways of doing things. Smallbones(smalltalk) 01:29, 31 March 2021 (UTC)Reply[reply]
@Smallbones: I'm not sure what question you thought I was asking, but it definitely wasn't the one you answered. My question is very simple - did you choose to leave the Arbitration Committee's motion to sanction Tenebrae out of this issue's Arbitration report? Mo Billings (talk) 02:53, 31 March 2021 (UTC)Reply[reply]

"Plus CPAC misinformation"

Oh. Misinformation about CPAC, not by CPAC. I'm not a fan of CPAeC, but that sub-headline is going to reinforce a lot of biases in people who don't click through to the article. --Kent G. Budge (talk) 19:03, 29 March 2021 (UTC)Reply[reply]

The sub-head is "Plus CPAC misinformation" which allows both readings, same as the Reuters. I've learned to be skeptical about situation where someone vandalizes an article. Then somebody takes a screen shot of the vandalism in the few minutes that it is online and then posts the screenshot on twitter and complains about someone or somebody. Sometimes you shouldn't take it ar face value. Smallbones(smalltalk) 20:52, 29 March 2021 (UTC)Reply[reply]


  1. ^ Nippon Todo-kai introduction page (in Japanese)