In two May 6 announcements to the community Maryana Iskander announced that in January she is planning to step down as Wikimedia Foundation CEO. Iskander reached out to The Signpost requesting an interview the same day.
See the position announcement here.
The CEO succession planning started more than a year ago and the announcements show a sense of confidence that Iskander is still on the job and will handle the transition in the most effective and professional way.
So we shouldn't expect a usual WMF CEO transition? Iskander just firmly replied that there wouldn't be any surprises.
Iskander officially joined the WMF in January 2022 after a two month long listening tour. She pointed to her 2022 article in diff as the starting point of her priorities as CEO. These included:
She listed some of the major accomplishments since then as:
Have we really gained more public trust? She recognizes that the last six months have been difficult for many Wikimedians in terms of politics, and watching legal and other government actions that target Wikipedia. And there are parts of the media that are very critical in their reporting about us. But she believes that the community and the WMF are strong enough to deal with these challenges.
Iskander tells us that her only immediate plans are to continue contributing to the WMF and the community, and to successfully complete the transition by January.
Her management philosophy is influenced by the book The Starfish and the Spider. The starfish in the title refers to the starfish's ability to regrow a limb if it loses one. She doesn't want to do a "debriefing" or a review of her WMF tenure now because it's still incomplete. We will continue to check in with her over the next eight months and report any major news here. We may even do that "debriefing" around the New Year. (Send in your questions!)
We wish Iskander success in completing the transition and in her future endeavors.
Thanks Maryana! – S
See related coverage in this issue's In the media
The Wikimedia Foundation has published a draft White Paper on the subject of best practices around privacy in Wikimedia-related research. The creation of such a paper was requested by the Arbitration Committee in 2023, to convey "to researchers the principles of our movement and give specific recommendation for researchers on how to study and write about Wikipedians and their personal information in a way that respects our principles." (Phabricator task.)
Volunteers are invited to provide feedback on the current draft by 1 July 2025. See meta:Research:Wikimedia Research Best Practices Around Privacy Whitepaper#How to provide feedback on the latest version of the privacy white paper (until 1 July 2025) and meta:Research talk:Wikimedia Research Best Practices Around Privacy Whitepaper. – AK
The Wikimedia Foundation office action that blanked the article Asian News International vs. Wikimedia Foundation has been reversed, after a decision by the Supreme Court of India to overturn a lower court who had ordered removal of content it deemed "defamatory" from Wikipedia. See prior Signpost coverage and this issue's In the media. – B
The MediaWiki graph extension has been broken and disabled since early 2023 (see prior Signpost coverage at the one year mark), but an announcement on Village pump (technical) states that they will be re-enabled soon. As part of the pilot program the Charts extension has been working on Italian, Swedish, and Hebrew Wikipedias, as well as on MediaWiki.org. It is currently expected that Charts will be enabled on enWiki around 20 May to 22 May.
The WMF's 2025/8 Foundation Bulletin features a draft 2025–26 Annual Plan, as well as Wikimania 2025 location selection, WikiForHumanRights, updates on annual goals in various categories (Knowledge Equity, Safety & Integrity, Effectiveness) and more. There's also a link to the Annual Reports for both the WMF and its Endowment (here and here, respectively).
The BBC [1], The Guardian [2], and The Verge [3] cover an announcement by the Wikimedia Foundation that it is bringing a legal challenge against new regulations under the United Kingdom's Online Safety Act 2023, which according the Foundation "could place Wikipedia as a 'Category 1 service' — a platform posing the highest possible level of risk to the public."
As reported by The Guardian,
The foundation said it was not challenging the act as a whole, nor the existence of the requirements themselves, but the rules that decide how a category 1 platform is designated.
Those rules were set in secondary legislation by the technology secretary, Peter Kyle. The foundation is challenging Kyle’s decision to proceed with that statutory instrument, via a judicial review, where a judge reviews the legality of a decision made by a public body, at the high court of England and Wales.
In a separate Medium post, Wikimedia Foundation lead counsel Phil Bradley-Schmieg explained the concerns about the possible classification of Wikipedia under category 1 in more detail:
There are many OSA Category 1 duties. Each one could impact Wikipedia in different ways, ranging from extraordinary operational burdens to serious human rights risks. [... T]he law’s impact would extend far beyond the UK.
The Category 1 “user verification and filtering” duties are a good example. [...] Sophisticated volunteer communities, working in over 300 languages, collectively govern almost every aspect of day to day life on Wikipedia. Their ability to set and enforce policies, and to review, improve or remove what other volunteers post, is central to Wikipedia’s success, notably in resisting vandalism, abuse, and misinformation. [...]
However, if Wikipedia is designated as Category 1, the Wikimedia Foundation will need to verify the identity of Wikipedia users. That rule does not itself force every user to undergo verification — but under a linked rule (s.15(10)(a)), the Foundation would also need to allow other (potentially malicious) users to block all unverified users from fixing or removing any content they post. This could mean significant amounts of vandalism, disinformation or abuse going unchecked on Wikipedia, unless volunteers of all ages, all over the world, undergo identity verification.
Although the UK government felt this Category 1 duty (which is just one of many) would usefully support police powers “to tackle criminal anonymous abuse” on social media, Wikipedia is not like social media. Wikipedia relies on empowered volunteer users working together to decide what appears on the website. This new duty would be exceptionally burdensome (especially for users with no easy access to digital ID). Worse still, it could expose users to data breaches, stalking, vexatious lawsuits or even imprisonment by authoritarian regimes. Privacy is central to how we keep users safe and empowered. Designed for social media, this is just one of several Category 1 duties that could seriously harm Wikipedia.
Bradley-Schmieg also detailed how some longstanding Wikipedia features might contribute to it being classified as such a high risk social media website, due to what he called "especially broad and vague" criteria in the categorization rules challenged by WMF:
To avoid any risk of loopholes, and due to limited research, the Categorisation Regulations were left especially broad and vague. They have no real connection to actual safety concerns. They were designed around three flawed concepts:
- Definition of content recommender systems: Having any “algorithm” on the site that “affects” what content someone might “encounter”, is seemingly enough to qualify popular websites for Category 1. As written, this could even cover tools that are used to combat harmful content. We, and many other stakeholders, have failed to convince UK rulemakers to clarify that features that help keep services free of bad content — like the New Pages Feed used by Wikipedia article reviewers—should not trigger Category 1 status. Other rarely-used features, like Wikipedia’s Translation Recommendations, are also at risk.
- Content forwarding or sharing functionality: If a popular app or website also has content “forwarding or sharing” features, its chances of ending up in Category 1 are dramatically increased. The Regulations fail to define what they mean by “forwarding or sharing functionality”: features on Wikipedia (like the one allowing users to choose Wikipedia’s daily “Featured Picture”) could be caught.
[...]
As a result, there is now a significant risk that Wikipedia will be included in Category 1, either this year or from 2026 onwards.
The Verge highlighted that
Wikimedia says it has requested to expedite its legal challenge, and that UK communications regulator Ofcom is already demanding the information required to make a preliminary category 1 assessment for Wikipedia.
The BBC noted that
It's thought this is the first judicial review to be brought against the new online safety laws - albeit a narrow part of them - but experts say it may not be the last.
"The Online Safety Act is vast in scope and incredibly complex," Ben Packer, a partner at law firm Linklaters, told the BBC.
The law would inevitably have impacts on UK citizens' freedom of expression and other human rights, so as more of it comes into force "we can expect that more challenges may be forthcoming", he told the BBC.
However, Packer seemed skeptical of the Foundation's chances to prevail in court:
"Typically, it is difficult to succeed in a judicial review challenging regulations," he told BBC News.
"Here, Wikimedia will be challenging regulations set by the Secretary of State on the advice of Ofcom, after they had conducted research and consultation on where those thresholds should be set," he pointed out.
See also previous Signpost coverage:
– H
"India Supreme Court reverses content takedown order against Wikipedia operator", according to Reuters, on an earlier High Court ruling that forced the Wikimedia Foundation to take down the Wikipedia article about the court case about ANI's accusation of defamation in the article Asian News International. The article Asian News International vs. Wikimedia Foundation has been reinstated by office action. Indian Express reports that
A bench of Justices A S Oka and Ujjal Bhuyan said, "It is not the duty of the court to tell the media to delete this and take that down… Both the judiciary and the media are the foundational pillars of democracy, which is a basic feature of the Constitution". "For a liberal democracy to thrive, both should supplement each other."
The original defamation case will still be decided by the High Court unless that decision is also overruled by the Supreme Court. – B, Sb
See also this issue's "News and notes"
The headlines of two stories appearing on the same day, from different political perspectives, are telling: DC Prosecutor Ed Martin Goes After Wikipedia For Exercising First Amendment Rights (Above the Law) followed by Wikipedia Nonprofit Status Under Scrutiny From US Justice Department Amid Claims of Systemic Anti-Israel Bias (The Algemeiner) [emphasis added by The Signpost].
The next day Lawmakers press Wikipedia to clarify and enforce editorial oversight to prevent anti-Israel bias (Jewish Insider), and Members of Congress call on Wikipedia to curb its antisemitism (Arutz Sheva).
In the meantime the Interim US Attorney for the District of Columbia Ed Martin lost his position and President Trump named Jeanine Pirro as Martin's replacement for the permanent position (according to The New York Times and two more Above the Law articles).
– Sb
The American MacArthur Foundation announced the five finalists in their 100&change grant competition, which will award a $100 million grant to a single project. This video shows Jimmy Wales and Denny Vrandečić presenting the Wikimedia Foundation's candidate. Vrandečić's long-term effort has already helped to produce Wikifunctions and Abstract Wikipedia, as he has explained on diff. All five finalists are shown on this video.
Rap mogul and superstar Jay-Z (Shawn Carter) seems to know a lot about how Wikipedia works, who is editing an article related to a court case about him and who is paying them. Or maybe it's just his lawyers or tech team who think they know. You may be surprised how much they likely know, and why I think they know it.
Since this story involves several controversial court cases, The Signpost reminds our readers that anybody accused in court should be considered innocent until the accusations are proven in court. We also remind you that the identities of Wikipedia editors can never be completely proven using only Wikipedia's extensive records – even if they seem to have identified themselves. They may be spoofing or "Joe jobbing" in order to embarrass other people. This reporter is not a lawyer and is not offering legal advice.
The most controversial case involved is the criminal case against Sean Combs, better known as Diddy, who is charged with multiple sex offenses. He has been held in jail without bail since September and the trial is just getting underway. But Jay-Z's case is not about what Diddy did or didn't do. It's about a possible victim, known only as Jane Doe. She has accused both Jay-Z and Diddy of raping her at the same party about 25 years ago when she was 13 years old. She has also withdrawn her rape accusation against Jay-Z, and the details of her story have changed over time. She has said that her lawyer, Tony Buzbee encouraged her to file a civil lawsuit against Jay-Z.
In February Jay-Z filed a civil lawsuit against Buzbee and Doe, accusing them of defamation and extortion. The defamation allegedly occurred on several well known TV and radio programs as well as in court depositions. The extortion allegedly occurred during a legal procedure known as a demand letter. On May 5 Jay-Z amended his civil suit to include a reference to Wikipedia, perhaps not to identify a direct form of defamation or extortion, but as supporting information.
"129. Defendants' actions also undermined Mr. Carter's relationships, and his company Roc Nation's relationships, with their businesses in the sports and entertainment industry. For example, in violation of Wikipedia's rules, Buzbee directed his employees to edit Wikipedia pages to enhance Buzbee's image and damage Mr. Carter's and Roc Nation's reputations. Users with an IP address directly linked to the Buzbee Firm made over 100 positive edits to Buzbee's Wikipedia Page." (- PLAINTIFF SHAWN COREY CARTER'S AMENDED COMPLAINT AND DEMAND FOR JURY TRIAL) Filed 05/05/25 Page 38 of 52
It looks easy enough to check out this type of allegation. Did unregistered editors (aka "IP editors") who had a direct connection to Buzbee's law firm make 100 edits at the Wikipedia article Tony Buzbee, adding positive content about Buzbee or removing negative content about him?
This would not be the first court case that involves undeclared paid editing on Wikipedia. In a 2011 London high court case billionaire Louis Bacon won a suit gaining an order against the WMF, The Denver Post, and the publisher of WordPress to identify the authors of alleged defamation on their sites. Bacon was trying to get enough information to sue the now-convicted sex offender former fashion designer Peter Nygard. But a U.S. court would not enforce the U.K. court's order.
Diddy is one of the best known hip-hop artists in the world. He was formerly listed by Forbes as a billionaire, but his legal problems have caused his net worth to drop to a mere $90 million.
More complete information on the Diddy cases is available widely throughout the internet, including at USA Today and the BBC podcast Diddy on Trial.
Jay-Z is arguably the best and best known hip-hop artist in the world. It's not arguable that he's the richest. According to Forbes, his net worth is $2.5 billion and he is the 1,470th richest person in the world. Jay-Z does not appear to be implicated in the Diddy criminal case.
Tony Buzbee is a high-profile lawyer involved in many highly public and highly lucrative class action cases where he usually represents the plaintiffs. He's also represented Jimmy Buffet, Deshaun Watson, former Texas governor Rick Perry, Texas Attorney General Ken Paxton (in impeachment proceedings), Shell Oil, and Ford Motors. He has represented about 150 clients in several separate civil cases against Diddy. His net worth is estimated at $50 million.
It shouldn't be a mystery how Jay-Z, or his team, know about the number of edits made by IP editors to the Wikipedia article about Tony Buzbee or even about which IP editors have made the most edits there.
Many experienced Wikipedia editors know that the easiest way is go to the article's edit history and click on "Page statistics" near the top of that page or just directly type:
https://xtools.wmcloud.org/articleinfo/en.wikipedia.org/Tony_Buzbee
in the address bar of a browser.
"General statistics", the first section on the xtools page, tells you, as of May 11, that there were 204 total editors who have made 561 total edits, including 221 by IP editors, to the page since the first edit on 2012-03-21 by Pedrokelley.
In "Top editors", the third section on the page, the table tells you that the editor who has made the most edits is 12.11.122.122, an IP editor, who has made 79 edits between 2017-06-07 and 2022-05-02. The editor with the 4th most edits is 50.254.106.146, also an IP editor, who made 23 of them between 2015-03-11 and 2016-03-07.
It looks like we are in luck, tracing where Jay-Z found out that a group of IP editors working for Buzbee made over 100 edits. The 2 top IP editors made 102 out of 221 IP edits, sticking out like a sore thumb from the other IP editors (who had no more than 5 edits each). This means that it would be almost impossible for there to be a group of IP editors with 100 edits without including at least one of the top 2. But how did Jay-Z know that they made positive edits to the article? And how did he know they worked for Buzbee?
The only way to check whether somebody is consistently making positive (pro-Buzbee in this case) or negative edits is just to check their edit histories. In the Top Editors table, you can get their edit histories just by clicking on the editor's username (or IP number).
Clicking 12.11.122.122 takes you to this page: a list of all the IP editor's edits. A quick glance at the page tells you that the IP editor is a single purpose account or SPA. Seventy-five out of his 79 edits were to the Buzbee article or to its talk page
Clicking 50.254.106.146 takes you to this page. Another quick glance tells you that the IP editor is also an SPA, though the edits are not quite so concentrated on a single article.
Both these IP editors strongly supported positions favorable to Buzbee's image, including this edit by 50.254.106.146 which sources a $100 million court award to 10 Buzbee clients that is sourced only to a press release that was written by Buzbee's firm. The $100 million award had been overturned before the IP editor inserted this information into the article.
Clicking on a registered account name will usually take you to the editor's main user page, and from there just click on User contributions in the tools menu on the right hand side. For example clicking Pedrokelley will take you to their as yet uncreated user page. Don't worry! Using the tools menu will still take you to Special:Contributions/Pedrokelley! There you will see that all 32 of Pedrokelley's edits were to the same article, Tony Buzbee, and that he created the article. Click on the earliest edit which reads "02:14, 21 March 2012 diff hist +7,850 N Tony Buzbee ←Created page with 'Tony Buzbee… " to see how the earliest version of the page read. It suggests to me that it was copied from Buzbee's business website. There's almost nothing negative in the article, barely a hair out of place, and likely considered a copyright violation according to Wikipedia rules. (Compare this Wikipedia version to this version from his business website.)
Incidental information includes the talk page filled with complaints about bias and unreferenced edits. One SPA editor who was supportive of Buzbee was accused of having a conflict of interest after making this edit where they added another version of the $100 million court award story and a possible sex abuse story involving four minors – all without references.
Two days later the same editor added the $100 million court award story again after it had been reverted and added four more stories – all without references again.
Several other editors stick out from the others including sockpuppets who were blocked following sockpuppet investigations by administrators. A sockpuppet is an editor who uses two or more registered usernames or IP accounts to deceptively hide their editing. Sockfarms refer to large congregations of sockpuppets. Two editors, Th78blue and Pulpfiction621, were both blocked as sockpuppets of PapaTakaro, a fairly small and quirky sockfarm. But the editor Th34VengersHere was blocked as a member of the industrial scale Yoodaba sockfarm.
But how could Jay-Z know that the two IP editors were Buzbee employees? Short of a direct on-Wiki statement by an editor that they work for Buzbee, the main method used by Wikipedians would involve the checkuser function and be used only for internal Wikipedia purposes. But a publicly available IP address might rarely connect via either of the WHOIS or Geolocate services to Buzbee's law firm. In this case both the IP editor addresses can be traced directly to Buzbee's law firm main web page by both WHOIS and Geolocate.
Tony Buzbee and the IP editors and sockpuppets who appear to have worked for him seem to have broken Wikipedia rules, for example by not declaring that they are paid editors. The evidence is very strong. They were not contributing to Wikipedia in good faith.
But if we are looking at a violation by Buzbee et al of biographies of living persons (BLP) policy against Jay-Z, something appears to be missing — a direct statement by them disparaging Jay-Z. All of the seven editors whose histories were examined in the previous section stopped editing by September 2022, a full two years before there was any clash between Buzbee and Jay-Z. While the editing to the current Buzbee article is not calm, neither does there seem to be any BLP violation.
Our BLP policy of course is not the same as the laws against defamation. That is something that the courts are designed to handle.
I'm not primarily an English Wikipedian; most of my Wikipedia time is spent contributing to Swedish Wikipedia or explaining the encyclopedia to the Swedish public. But I still hang out here. I fix mistakes I come across while reading. I illustrate articles, dabble in policy debate, take part in some talk page conversations, even write the occasional English article. Mostly I haunt Articles for Deletion, where I keep an eye out for anything related to Sweden, to help hunt down and contextualise sources to ensure we can save notable articles.
Usually, it's a simple task of expanding the article a little bit, adding a few sources to make sure key information can be verified elsewhere, and letting people know it's no longer the same text as was taken to AfD.
Sometimes it's a frustrating exercise for everyone involved.
Ideally, the encyclopedia wouldn't rely on journalism. Journalists are typically not subject-matter experts; their speciality is the craft of journalism – writing, storytelling, interviewing, quickly absorbing key points of a topic and presenting a digestible version to their readers. They have deadlines. Newspapers have budgets which rarely allow them to truly take in a field of knowledge. This is not to disparage journalism, which plays a crucial role in society; it's just that "create the best possible encyclopedic sources" isn't part of the journalist's job description.
Wikipedia editors have the luxury of working at our own pace, gathering and synthesizing information from diverse, more authoritative sources. But newspapers are accessible. Easy to find. Easy to read. They write about people, organizations and topics which might not have been covered by academic research or other, slower forms of non-fiction writing. And so we constantly fall back on them in our discussions on whether an article is notable or not, verifiable or not.
Google News often omits Swedish news articles. If one relies on Google News, it's quite possible to do a decent WP:BEFORE – and attempt to see if an article can be sourced before one takes it to Articles for Deletion – and come to the conclusion that no significant coverage exists, although there might be dozens of relevant articles. This is true for most languages I've reason to look for news in. When searching for Finnish news, where I have no specific skill or access, I have far more success searching for the topic and individual news outlets than Google News.
To further complicate the issue, outside of the Swedish public service – national radio and television – there's a strong paywall norm. Even when speaking Swedish and being familiar with the Swedish media landscape, looking for articles can be fairly hopeless without access to Swedish newspaper archives.
Besides the individual news outlets, there are two significant Swedish newspaper archives.
The most useful one for our purposes is Mediearkivet, a for-profit collection of almost all modern Swedish newspapers, as well as many other media publications, and some other Nordic sources. It is comprehensive for current news, but coverage gets more and more sketchy as we go further back in time. For a few newspapers it's decent into the 1990s or even 1980s, for others it's less useful already in the early 2000s.
Access is typically institutional. Anyone – or at least most – with a Swedish university account should have access, both students and staff. Others, like me, have access through Wikimedia Sweden, who are covering the costs for a number of Wikipedia editors who depend on it for their work on the encyclopedia. Most Swedes lack this privilege.
The other significant archive is Svenska tidningar, the newspaper collection of Kungliga biblioteket (the Royal Library). It goes back to the 1800s, though some newspapers have been scanned and others are still awaiting digitalization. The main issue isn't the partial coverage, but access: For copyright reasons, it requires you to go to the library itself, or one of the libraries which have access. For me, living in Sweden's third largest city, the library network has two computers which can be used to connect to the service.
Many newspapers can be read through the library, from home, without having to leave the comfort of one's own couch or desk. Those of us who subscribe to a newspaper might have their entire archive at our disposal. But this is typically more useful – especially for reading newspapers through the library – when you have already found the texts and just need to be able to read them, than when doing a thorough search of what's available.
Beyond access it's a matter of understanding the context – what sources are reliable in which situations? There is no such thing as a "reliable source" without contextual qualification. A source can be excellent at one topic and far less reliable in another situation. I've been told that we can trust The New York Times because it's a reliable source, in a conversation about a news piece about events in Sweden, where they don't have a foreign correspondent, in a story where it's fairly clear that they've done a rewrite of an article in The Guardian which in turn was a rewrite of an article in Expressen. On the one hand, award-winning investigative journalism, on the other hand Swedish tabloid journalism made worse by the game of Telephone: We should not treat these the same way, even when they live in the same newspaper. Similarly, figuring out the relationship between a source and the topic at hand can be difficult enough for those of us who sometimes deal with them professionally. Sometimes the best possible journalistic source is the small local newspaper.
So I patrol Swedish-related articles on Articles for Deletion, to see if I can source and save articles about notable topics; I'm less inclined to spend my time on articles where I assume I won't find anything useful and which will be deleted whether I participate or not.
I usually try to find an online equivalent to link to. Sometimes they are paywalled. I often fail, and the only thing I can reference is in the archive, impossible to link to. These I treat as print sources – it's not like we expect every piece of information we refer to to be easily accessible online. Yet some editors tend to ignore anything unlinked in their source assessments, as if it weren't there at all. This is a slightly bizarre experience: Having spent quite some time skimming newspaper articles, discarding most as irrelevant, adding a few, pointing to them in the conversation, and it's like they don't exist, in a world where we still print books.
On the other hand, it's frustrating for anyone trying to do an independent source assessment, when it's implied that it will be more or less impossible for them to reliably assess what's being presented.
First, it means that it's important that I'm not alone in both having access to these sources and having Wikipedia:WikiProject Deletion sorting/Sweden on my watchlist. It'd be folly to expect everyone to have access to all sources we reference – but the Wikipedia model is definitely helped by the handful of people with the ability to check these articles who keep an eye on the Sweden-related AfD conversations. Second, it demands that we not only add the sources, but also describe them in the discussions. Length, to what degree does it cover the topic at hand: is the source generally considered reliable? Is the newspaper national, regional, local?
It's a fine balance: Trying to convey an assessment where one feels confident one has a far better understanding of what's been written and in what context than those without the same access, without attempting to monopolize the conclusion of what this means for the editorial discussion.
Starting on May 19, the Wikimedia Foundation's sites will set a unique first-party cookie on users' browsers - a practice that is very common on the web, but has so far been avoided on WMF sites for privacy reasons. WMF stresses that its new "Edge Uniques" system will implement this cookie in a privacy-preserving way
, based on
a process that reads, verifies, and discards a copy of the cookie "at the edge" of our computing systems – meaning the first point where visitor traffic enters our network. This process minimizes the time this uniquely identifying information about a logged-out user will be present in our system to seconds and typically milliseconds. Because we are not storing this cookie in our traffic logs or databases, we cannot create user profiles that could be used to track readers' usage behavior over time. This solution will provide a standardized, privacy-preserving framework that staff and volunteer developers can use when implementing new features, bots, tools, and gadgets that require continuity or analytics.
The benefits of this solution are described as follows:
1. Improve user experience through A/B testing [...]
2. Enable protection against DDoS attacks [...]
3. Understanding our visitor trends
[...] we need to accurately count how many visits our wikis receive on different types of devices and in different geographies (among other dimensions) [and the new cookie will enable a more precise count than the currently used "Unique devices" solution]
Despite the efforts to discard the unique value of the cookie almost immediately (which still sets WMF apart from many other websites), some privacy concerns remained, with one former WMF engineer pointing out that We've gone from a system where its impossible for WMF to track people, to one where we rely on trust for WMF to do the right thing. Sure the software is open source and whatnot, but there is no way for the average user to verify that the deployed software is the software it is supposed to be.
In response, the Foundation added a canary page that can be used to verify what Wikimedia Foundation does with the Edge Unique cookie.
– H
The Wikimedia foundation is now rolling out an update where you're able to get a QR code for a page on some Wikimedia projects (Special:QrCode). – V
A new Wikipedia has been introduced in Nupe. This is a language primarily used in the North Central region of Nigeria.
Bot Name | Status | Created | Last editor | Date/Time | Last BAG editor | Date/Time |
---|---|---|---|---|---|---|
DreamRimmer bot III 9 (T|C|B|F) | Open | 2025-03-21, 04:50:16 | 1AmNobody24 | 2025-04-03, 13:25:14 | Never edited by BAG | n/a |
Bot1058 10 (T|C|B|F) | Open | 2025-03-07, 17:49:45 | Wbm1058 | 2025-03-27, 15:40:30 | Anomie | 2025-03-09, 00:12:51 |
AnomieBOT 84 (T|C|B|F) | In trial | 2025-04-11, 00:29:13 | Anomie | 2025-05-11, 22:31:28 | DreamRimmer | 2025-04-29, 14:45:51 |
C1MM-bot 3 (T|C|B|F) | Extended trial | 2024-12-12, 04:42:12 | DreamRimmer | 2025-05-12, 03:34:43 | DreamRimmer | 2025-05-12, 03:34:43 |
CFA (bot) (T|C|B|F) | In trial: User response needed! | 2024-12-31, 05:00:34 | CFA | 2025-03-27, 02:24:46 | DreamRimmer | 2025-02-14, 17:06:11 |
CanonNiBot 1 (T|C|B|F) | In trial: User response needed! | 2024-12-17, 12:50:01 | DreamRimmer | 2025-02-14, 17:07:41 | DreamRimmer | 2025-02-14, 17:07:41 |
SodiumBot 2 (T|C|B|F) | In trial | 2024-07-16, 20:03:26 | DreamRimmer | 2025-01-26, 08:10:11 | DreamRimmer | 2025-01-26, 08:10:11 |
AussieBot 1 (T|C|B|F) | Extended trial | 2023-03-22, 01:57:36 | DreamRimmer | 2025-02-02, 11:19:10 | DreamRimmer | 2025-02-02, 11:19:10 |
Ow0castBot (T|C|B|F) | In trial | 2024-11-14, 01:51:38 | Primefac | 2025-02-26, 14:33:55 | Primefac | 2025-02-26, 14:33:55 |
TenshiBot (T|C|B|F) | Trial complete | 2025-04-20, 22:26:31 | Tenshi Hinanawi | 2025-05-08, 17:58:13 | DreamRimmer | 2025-04-29, 14:43:25 |
RustyBot 3 (T|C|B|F) | Trial complete | 2025-03-25, 18:34:23 | Rusty Cat | 2025-05-10, 23:10:04 | DreamRimmer | 2025-05-10, 15:59:26 |
Latest tech news from the Wikimedia technical community: 2025 #20, #19, & #18
{{subst:lusc|1=User:Jeeputer/PCBIndicator.js}}
While each RfA will be unique in some way, I'm not sure I can add much to what previous RfA debriefs have touched on in terms of the experience itself, but what I would like to reflect on are my feelings about "mop work", how they changed over the last 15 years and how that led me to being interested in being an admin.
To my experience at least, there's now a more or less overwhelming admin culture of remaining civil and responsive, as against uncivil and unresponsive. There's a greater consciousness for and enforcement of admin accountability. I won't pretend it's perfect but compared to when I first started editing almost 20 years ago, there's a noticeable culture shift. To be simplistic, we're far closer now to being staff at a public library than brothers in a fraternity house. That shift was important for me in deciding to make a run at RfA.
My first period of intense engagement with Wikipedia was around 2008–2010 and I often found admins lacking in civility; there appeared to be a culture of combativeness associated with the work (which was not necessarily the admins themselves, but the culture in which admin work had evolved).
To be blunt, it was not attractive in the sense that it was something I thought I would enjoy doing. I felt there were more than a few admins who acted in a privileged manner, who viewed their status as admins as proof of their correctness. Wikipedia feels very different now and, to misphrase Groucho Marx, I found a club I'd hope would have me.
I think if there's any advice I'd give others thinking about this process, it would be to sit back, read and watch. For as long as possible.
If you think you're ready to run, wait another six months or a year. I waited two years before fully deciding after being asked. I didn't feel underprepared for the RfA — I’ve stalked the noticeboards probably far more than I should (in the sense that maybe I should enjoy real life more!). I was aware that there are areas I'm not strongly familiar with, but I felt I had a strong knowledge of Wikipedia’s core tenets (thank you AfD!) and having followed RfAs over the last six or so years, it's clear that being an admin is about having the honesty and self-recognition that one will never be a master of everything.
When admin elections were first announced, I had thought of participating but hesitated. I realise there's quite justifiable criticisms of RfA, but for better or worse, I decided that RfA's particular brand of transparency appealed to me, warts and all. I'll continue to support admin elections, but for me, an RfA felt the right approach. The start of the RfA was fairly anxiety inducing, but the fact that a couple of editors and admins who I respected a great deal were quick, early supporters, gave me a lot of relief.
Ultimately, I was surprised how much I appreciated the comments from all those who participated; it reinforced my sense of the positive contributions that a supportive culture can create. We really underestimate (underappreciate!) how much human kindness and collective effort basically keeps Wikipedia going day after day.
Max Lum, Ed.D, MPA (ICOHBuzz), was employed by the National Institute for Occupational Safety and Health (NIOSH) from 1995 until his retirement in 2011. Prior to his retirement he served as Communications Director, and subsequently he returned to NIOSH as a consultant. He grew up in New Jersey and attended the University of Southern California. He died peacefully on November 19, 2024.
Max's role at NIOSH included spearheading the agency's communications on emerging media platforms, including social media. It was Wikipedia in particular that held a special place in his heart. Through his advocacy he succeeded in bringing in Wikipedians in Residence at NIOSH, whose responsibilities included contributing media collections and large datasets directly to the Wikimedia projects and training NIOSH staff on how to best contribute to Wikipedia.
Max advocated passionately for active participation on Wikipedia: not just within the United States government, but with occupational safety and health agencies all over the world. He presented at conferences in the United States, Denmark, Singapore, and elsewhere, on the value of Wikipedia to workplace safety agencies. He loved to travel, and he loved to tell people about Wikipedia while doing so. Directly or indirectly, Max helped to create new Wikipedia editors and give existing ones a renewed sense of purpose.
He will be sorely missed, especially by the Wikipedians who had the pleasure of working with him, and now carry on his work.
As an online organization, The Wikimedia Foundation obviously makes heavy use of software and hardware. Unlike most organizations of its size, however, the foundation tries to document or open source as much of the technology as possible. While maybe not for everybody, looking at the technology behind Wikimedia can be interesting enough that I decided to give it its own section.
As a foundation that runs one of the largest websites in the world, it's not a surprise that Wikimedia has a very complex tech stack, which is totally documented on Wikitech. To be totally blunt, there's parts of this that I just don't understand. I'm only going to look at the parts that I actually understand and think is noteworthy, but Wikitech has way more detail about everything that I'm going to write and not write about in this section.
To monitor the health of the Wikimedia infrastructure, Wikimedia has 100s of publicly available dashboards at Grafana. Some of these dashboards link to documentation about what the metrics mean, but others don't. Of course, you probably aren't going to spend much time looking at these dashboards unless you already understand this stuff.
Whenever possible, Wikimedia uses free and open source (FOSS) software, which anybody can contribute to. Some of the code is hosted on Gitlab or Github, but most of it is hosted on Gerrit. Once you get a developer account to get access to Gerrit, there's tons of different projects that you can work on. Information about contributing to the infrastructure can be found directly on Wikitech, while information about the other projects can be found on the Wikimedia Developer Portal.
On top of the Git host, most software projects will have an associated Phabricator page. This is used for managing work, but people who don't develop can also use it to report bugs, issues, or to request features. It's somewhat analogous to the issue page for a Github project.
Yup, Wikimedia has its own cloud computing platform, which is for developers who wanted to create programs to improve the projects. To provide hosting, the Wikimedia Foundation offers a cloud service called Toolforge. True to Wikimedia form, anybody can make an account and start writing their own tools, as long as they follow a fairly basic set of rules. However, this is far from the only cloud service offered by Wikimedia. There's also PAWS, which hosts Jupyter notebooks for analysis of Wikimedia projects and small bots, Cloud VPS, which is similar to Toolforge but is an Infrastructure as a service solution instead of a Platform as a service solution, and Quarry, a web interface to run SQL queries against Wikimedia projects. There's also Superset, which is also a web interface for SQL queries, but with the additional benefit of being able to create dashboards.
We tend to take it for granted that the Wikimedia projects are always up and accessible. However, that isn't true for everybody. Some countries have censorship laws that prevent citizens from using these websites, while other people have unreliable internet that makes it impossible to access any website whenever you want. Thankfully, the Wikimedia Foundation dumps every single project every 2 weeks, as well as various statistics. Not only can you download the contents of these websites, you can also download the revision history and statistics as well. This also makes it good for research and archival in the unlikely event that Wikimedia goes down. These dumps can't be read directly, but they can be read with FOSS tools such as Kiwix.
I already mentioned that the Wikimedia projects are run using a piece of software called Mediawiki, but I never actually explained what that is. Mediawiki is a free and open-source software (FOSS) system that was developed in 2002 to better run Wikipedia after the pre-existing software was found to be too limiting. Knowing how to use Mediawiki well can massively change your user experience in ways that you wouldn't expect from how old-fashioned the UI looks. It also comes with an API that can be used to scrape Wikimedia pages.
MediaWiki is essentially a no-code solution to modify a website. Instead of using HTML, it lets you write in plain-text and has built-in functionality to let you link to different pages on the website. Specifically, it uses a markup language called wikitext, which ironically doesn't necessarily have to be text (a change made to accommodate Wikidata). All changes made to a page are recorded in a public revision log, which can be used to revert bad edits.
There's no such thing as a one-size-fits-all solution. Very often, users or admins will want to customize the UX for MediaWiki to either fit their own needs or the needs of the community. For that, you usually turn to extensions, which range from mere cosmetic changes to adding new functionality.
Templates are pages designed to be included in other pages. This is useful for when you need to frequently repeat an action while editing a project.
Namespaces are groups of pages, which are all connected by their name. When using a Wikimedia project, the pages that you're going to look at the most are the ones that make up mainspace, which is the name space dedicated for whatever the project is actually about. However, there's usually also a namespace for all the user pages, a namespace for discussion about the articles in mainspace, and a namespace for templates. As a general rule of thumb, you'll have to do a lot of searching to truly become acquainted with all the namespaces that a project has to offer.
An important feature of Mediawiki is user rights, which restricts or grants rights to certain groups. Mediawiki allows for admins to create groups and assign them rights, but there are also groups that are created by default. The 2 big ones are the admin and bureaucrat groups. Like the name implies, users assigned to the admin groups, well, administrate the website. They modify the CSS, ban malicious users, and do whatever else is needed to keep the community functioning properly. The bureaucrats are similar to the admins, except they can add or remove anybody from any group, including the admins.
There are many aspects to the movement that are deeply important, but not part of the main projects whatsoever. Some of those things will be given their own section, but this part of the blog post will exclusively deal with the miscellaneous stuff that doesn't neatly fit in any other category.
This is the wiki for the movement as a whole, rather than any individual project. Here, you can find details about what Wikimedia is doing, how it's doing it, and what it plans to do in the future. This is also a good place to find events and documents explaining various initiatives that Wikimedia has tried or are currently ongoing. While contributing to this wiki might seem harder than the official projects, Metawiki is always looking for volunteers to help translate content. There's always a lot that needs to be translated, even for common languages, so it's a good entry point for bilingual people who want to help out the movement. Alternatively, if you're the kind of person who's better with computer languages than human languages, you can also volunteer to provide tech support to the various communities that make up Wikimedia.
These are the admins of admins. They can access any public Wikimedia project and change the user rights of anybody. The idea is that they can serve as admins to projects that have yet to appoint their own admins and to act in emergencies where the proper admin is incapacitated or too slow to act. There's annual elections, but only admins who have at least 600 edits on one project and 50 edits made in the last month can run. Similar editing requirements also apply to anybody who wants to vote, but you don't need to be an admin. To be elected, candidates must receive at least 30 supporting votes, and at least 80% of the votes must be supporting.
Most Wikimedia projects have different language versions. Rather than go through the effort of making a version that nobody ends up using, potential new versions are instead prototyped on Wikimedia Incubator. From there, the community can contact a group called the Language Committee for approval to become a new project. In true Wikimedia fashion, anybody can create a new language version of any project besides Wikidata, which is language-agnostic, Wikifunctions, also language-agnostic, Wikiversity, which hosts new language versions on Beta Wikiversity, and Wikisource, which hosts new language versions on the multilingual section of the website. The languages that get prototyped depend on the project. Wikipedia has already covered all the major languages, so all that's left is languages that just barely have enough speakers to justify a new language version. Meanwhile, the less popular projects often have major gaps, particularly Wikivoyage, which has major languages like Indonesian and Czech stuck in the incubator.
The Wikimedia projects are created through a proposal process. People interested in creating a new project propose it for the community to debate on, create a demo, and hopefully get proper recognition. Similarly, Wikimedia projects that seem to be dead can be removed by the community if someone proposes to delete it.
Wikispore is an experimental project where people can create their own miniwiki centred around particular topics like art or biographies. This makes it similar to Fandom, but without the ads or bloat. Because Wikispore is a collection of separate wikis, searching for things can be hard. However, it has a small but decently sized community, and by far the top contender to become a new Wikimedia project.
At this point, I've hoped that I've convinced you that the Wikimedia movement is massive. Because of its scope, there's a lot of potential for research about online sociology, data science, and natural language processing. As part of this, Wikimedia has a dedicated page where researchers can publish the work that they've done or are in the process of doing. A fairly large number of these projects are commissioned by the Wikimedia Foundation itself in order to create products and improve the user experience, which offers an exciting insight into what Wikimedia projects might look like in the coming years. There's a monthly newsletter if you want to receive regular updates about the research that goes on (this newsletter is also published as a section in The Signpost).
People have different interests, which means that the level of information you can get about certain topics can vary widely. Because of this, there's occasionally campaigns to encourage efforts to edit articles in underrepresented areas and programs to help teach people not used to Wikimedia how to participate. These campaigns and programs can be found at the Outreach Dashboard, which also contains information on how to create and run a campaign or program of your own. These campaigns will openly publish metrics about their impact and level of participation, which is helpful for seeing if your campaign is the worth the effort it took to run it. Some of these campaigns have prizes, so it's definitely worth looking through and seeing if there's one that you find interesting.
Despite what your middle school English teacher might have said, Wikipedia is actually a pretty decent source of information. More importantly, if you're in the 3rd world, you might not have access to more traditional forms of information like books. To that end, the Wikimedia Foundation has been putting a lot of effort into promoting the use of Wikipedia (and the other projects) in educational settings. As part of its commitment to transparency, the programs funded by Wikimedia are publicly listed alongside their goals and the institutions running them. The flagship program is Reading Wikipedia in the Classroom, which teaches teachers on how to use Wikipedia as an educational resource. A pilot program that educated 7000 teachers in the Philippines, Bolivia, and Morocco found that the program drastically improved teachers's view of Wikipedia and their willingness to use it in the classroom. This program has only been done in 7 countries, all of them 3rd world, but the training material can be freely accessed and used by everybody.
A very closely related initiative is the Wiki Education Foundation, an organization created by the Wikimedia Foundation in 2013 in order to handle the pre-existing Wikipedia Education Program. The Wiki Education Foundation has programs for universities to assign Wikipedia writing assignments, to teach researchers about how to use Wikipedia, and for institutions to develop an initiative to add their specialized knowledge to Wikipedia.
Finally, there's WikiLearn, an initiative by and for Wikimedia users to learn about leadership within the community, such as teaching grant recipients how to run a program safely or to help onboard people who are considering whether or not they want to run for a seat on the WMF Board of Trustees. This is done by offering courses that can be used by anybody for free, even if the course is only meant for a niche audience. The courses are currently only made by staff and trusted affiliates, but once a governance model is made for WikiLearn, there's plans to let everybody create courses.
People like community, and while most Wikimedia communities are obviously online, there's still an appetite for meeting people IRL. The most prominent Wikimedia meetup by far is Wikimania, an annual conference where editors and interested parties can learn and discuss about various issues surrounding the projects. However, this is far from the only way to meet other editors IRL. There's tons of meetups happening in major cities around the world, which you can find at a dedicated Wikipedia page, and if there isn't one near you, you can always start your own.
This is a movement to work with Galleries, Libraries, Archives, and Museums in order to provide content for the Wikimedia projects. This is often done by having these institutions appoint a Wikipedian in Residence to write articles, but they can also integrate Wikimedia content into the institutions themselves, like by pairing exhibits with QR codes that link to relevant Wikipedia articles. The people behind GLAM maintain their own outreach page for people who want to contribute and have a monthly newsletter for people who want to monitor the progress of this movement.
Next issue: Part 3, The Wikimedia Foundation and The Other Groups
This article was originally published on April 30, 2019. Several other editors contributed ideas for the article including those who contributed to WP:JOKES.
After last month's very serious discussion of humor in The Signpost, Wikipedians responded in a way that I did not expect. They started telling jokes. A few of them were actually pretty good. A search through some user pages and articles reveals that humor still lives on Wikipedia. I've hijacked some of it for this column. More will be found in the comments section. My appreciation to Levivich, Atsme, and EEng.
Wikipedia's not for the meek
You need a de-stress technique
Sip tea with biscotti
Go fish – try karate
But edit war? Blocked for a week!
"Has the jury reached a verdict?" the judge asks.
"Yes, your honor," the jury foreperson replies. "The result of the discussion was jailify."
The wikilawyer jumps from his chair. "NOTAVOTE!" The judge shakes her head: "CONSENSUS has been reached."
"REVERT!" exclaims the wikilawyer. The judge shakes her head: "0RR."
"DRV!" the wikilawyer demands. The judge shakes her head: "No FORUMSHOPPING."
"There is NORUSH!" argues the wikilawyer. The judge shakes her head: "The deadline is NOW."
"DTTR!" the wikilawyer asserts. The judge shakes her head: "TTR."
"NOSPADE!" pleads the wikilawyer. The judge shakes her head: "SPADE."
"IAR!" the wikilawyer shouts. The judge shakes her head: "DROPTHESTICK."
"NOTBUREAUCRACY!" retorts the wikilawyer. The judge takes a book from a shelf: "PAGs."
The wikilawyer cries "ANYONECANEDIT!" and tears out a bunch of pages.
"BLOCK!" orders the judge.
As the court officers move to handcuff the wikilawyer, the defendant bolts for the door, yelling, "Wikipedia does not need me!" - L
![]() |
How To Avoid Pricks
|
When you land in a place that is prickly at best,
And feathers get ruffled – you've disturbed someone's nest; |
An editorial artist or cartoonist.
No captions, funny words in bubbles, or artistic talent required.
Benefits: standard Wikipedia vacations, insurance, and pay (up to twice your monthly average, provided it doesn't exceed our annual budget).
Apply at the Newsroom talk page or leave a sample in the comments section below. -S
Still waiting for Jimbo and Larry to walk into that bar? Sorry, that screenplay's still in development. But this is a wiki, so feel free to add your own version in the comments below, or to edit the versions of others.