Wikipedia:Wikipedia Signpost/Single

Following the closure of the nomination process for candidates, online ballots for the December 2025 ArbCom elections have opened on November 18, with the Arbitration Committee looking to seat nine members for the upcoming term. Eligible voters can cast their vote through the SecurePoll system until December 1 – see the October 20 and 10 November 10 issues for more.

In other election news, the call for candidates in the December 2025 admin elections will take place (November 25–December 1), before moving to the discussion phase (December 4–8) and, finally, the voting phase (December 9–15). – O

Users Bobby Shabangu and Michał Buczyński have been elected to the WMF Board of Trustees as part of the 2025 elections. Shabangu was elected in the second round of the STV ranked vote, having collected 2,258 preferences, whereas Buczyński, who had originally won 1,298 votes, was elected in the fifth round. The two new Trustees will be appointed at the next Board meeting in December 2025.

The latest BoT elections went ahead as expected despite a recent controversy involving former candidates Ravan Al-Taie and Lane Rasberry, who the Board had both removed from the slate for the elections before the start of the community voting period. The BoT faced criticism for their unprecedented decision, with Board member Victoria Doronina later announcing that she had suspended herself from most of her activities until the end of the year.

See both the Special report and Interview sections from our October issue for further context on the elections and the controversy surrounding them. – O, B

The Wikimedia Foundation has published its audit report (in US dollars) covering the period from July 2024 to June 2025. Highlights include record revenue, investments in product and technology, funding advocacy and awareness for Wikimedia projects, limited internal expense growth, and improved revenue diversification.

Key stats (rounded) from page 6 of the report:

• Total operating support and revenue: $209m (up from $185m)
• Total operating expenses: $191m (up from $178m)
• Net assets at end of year: $297m (up from $272m)

For frequently asked questions see the FAQ on Meta-Wiki. Additional questions can be asked on the FAQ's talk page. See also the Wikimedia Foundation's blog post.

Reporting on Wikimedia Enterprise is included in the audit report, but there is also a separate report for Wikimedia Enterprise, which states: "We are pleased to share that not only does this report show the first complete year of profitability, but even more importantly, that earned revenue has now fully repaid the initial investment in the project from previous fiscal years."

The Wikimedia Endowment became its own standalone 501(c)(3) in September 2023 and now files its own audit report, which will become available in a few months' time. – AK

Back in June, the Community Affairs Committee (CAC) of the Wikimedia Foundation Board of Trustees (BoT) had assigned the Sister Projects Task Force (SPTF) to "update and implement a procedure for assessing the lifecycle" both of a proposed Wikimedia sister project, meta:Wikispore, and an existing project, Wikinews. Two separate public consultations followed, with discussions continuing until August 15.

On November 25, BoT Chair-elected member Lorenzo Losa published the results of the consultations and the Task Force's consequent recommendations to the Board, which you can read in full on the Meta page. In summary, the Board was advised to keep the current technical setup of Wikispore, and consider options to integrate the project in the Wikimedia Incubator.

No immediate changes should be made to Wikispore's current technical setup. The current setup is functioning and supports ongoing experimentation, and bringing Wikispore into the WMF technical setup at this point of time would add necessary technical constraints that would not well support Wikispore's evolution.

Exploring options to more closely align Wikispore with the Wikimedia Incubator is encouraged, potentially as a dedicated subdomain or pilot project for a fixed period (e.g., two years). This would simplify processes and allow lessons learned in Incubator to inform Wikispore's trajectory, if this can be done while not restricting Wikispore's growth.

Initially proposed by user Pharos in 2019, Wikispore is a project that is intended as an extension of the already existing meta:Wikimedia Incubator, in order to develop and test in a "safe space" any of the other proposed sister Wikis that are consistent with the copyright and NPOV principles of the WMF.

On the other hand, the Task Force proposed to cease the activity of Wikinews permanently, while encouraging "exploration of new paradigms for Wikimedia news content" with the help of community of new members.

Archive all editions of Wikinews, preserving their content.

The implementation (including the timeline, archival method, etc) are the responsibility of the Product & Technology department, but the process should be sensitive to local project contexts and follow inclusive, transparent processes.

The Wikimedia Foundation should support and provide resources to groups exploring new paradigms for Wikimedia news content, such as the proposed "Wikinews Pulse" centralised multilingual headline portal. It should set a fixed timetable (e.g., one year) for these pilot initiatives to demonstrate progress, after which results should be publicly reviewed and further recommendations made.

Wikinews is an official Wikimedia project based on news reporting and citizen journalism that was first launched in November 2004, following an online vote on Meta: Wikipedia co-founder Jimmy Wales intended the project as a way to write each story "as a news story, as opposed to an encyclopedia article". However, Wikinews has always struggled to gain momentum in comparison to other WMF projects throughout the years: at the time of this issue's publication, the platform is active in 31 languages, with just over 700 active editors across the board. – O

In a post on the Wikimedia Foundation's tech blog, titled "Unifying our mobile and desktop domains", WMF engineer Timo Tijhof describes the recently completed work to remove the separate mobile domains for Wikimedia projects (e.g. en.m.wikipedia.org). Among other benefits, this "led to a 20% improvement in mobile response times" for readers who access Wikipedia and its sister sites coming from Google (i.e. a majority of readers overall).

The post recaps how the separate mobile domains had been introduced back in 2008 when this was a common practice among major websites like the BBC, IMDb and Facebook, which however has long become outdated. What triggered current conversion, though, was that "above all, we had reason to believe Google stopped supporting separate mobile domains." This was found to have caused huge slowdown for many Wikipedia readers (which apparently had gone undetected for over a year):

Google introduced a new crawler in 2016, and gradually re-indexed the Internet with it.[...] This new “mobile-first” crawler acts like a mobile device rather than a desktop device, and removes the ability to advertise a separate mobile or desktop link. It’s now one link for everyone! Wikipedia.org was among the last sites Google switched, with May 2024 as the apparent change window.[...] This meant the 60% of incoming pageviews referred by Google, now had to wait for the same redirect that the other 40% of referrals have experienced since 2011. [...]

Unifying our domains eliminated the redirect and led to a 20% improvement in mobile response times.[...] This improvement is both a recovery and a net-improvement because it applies to everyone! It recovers the regression that Google-referred traffic started to experience last year, but also improves response times for all other traffic by the same amount.

As part of the conversion work, WMF engineers also fixed two major SEO failures involving Commons that likewise appear to have flown under the radar for a while:

1. In response to two Community Wishlist proposals (one by User:TheDJ and another by User:Prototyperspective),

Tim Starling found in June that only half of the 140 million pages on Commons were known to Google. [phab:T400022] And of these known pages, 20 million were also delisted due to the mobile redirect. This had been growing by one million delisted pages every month. [...] Tim and myself [Timo] disabled the mobile redirect for “Googlebot on Commons” through an emergency intervention on June 23rd. Referrals then began to come back, and kept rising for eleven weeks in a row, until reaching a 100% increase in Google-referrals. [...] The index had likely been shrinking for two years already.

2. In addition, it was found that videos on Commons had been almost entirely absent from Google:

"We also found that less than 0.1% of videos on Commons were recognised by Google as video watch pages (for the Google Search “Videos” tab). I [Timo] raised this in a partnership meeting with Google Search, and it may’ve been a bug on their end. Commons started showing up in Google Videos a week later.

– H

• The Wiki Science Competition gears up worldwide: The 2025 edition of the Wiki Science Competition, a global photography competition for students, researchers and other volunteers that aims to expand the catalogue of science-related media to Wikimedia Commons, has officially begun on November 1. The WSC will be hosted internationally, but several national contests are taking place across the globe: see this page to look at all of the contests available and get involved.
• Milestones: The following Wikimedia projects have reached milestones in October and November 2025:
  • 1,000 articles: Rakhine/Arakanese Wikipedia
  • 20,000 articles: Assamese Wikipedia, Oriya Wikipedia
• Articles for Improvement: This week's Article for Improvement is Emergency management, which will be followed by Computer algebra system (beginning 1 December). Please be bold in helping improve these articles!

Jimmy Wales, in Germany to promote his book The Seven Rules of Trust, answered the same question four times in one minute in a Jung & Naiv video interview hosted by Tilo Jung – before walking out of the interview. The question posed by Jung, known for a faux-naïve interview style inspired by Stephen Colbert, was: "Are you the founder or co-founder of Wikipedia?" Wales's answer was: "It doesn't matter."

The incident attracted press coverage in Germany –

• Berliner Zeitung: "„Dümmste Frage der Welt“: Co-Gründer von Wikipedia verlässt Interview mit Tilo Jung nach Sekunden"
• Der Spiegel: "Wikipedia-Gründer Jimmy Wales nennt Tilo Jungs Fragen »dumm«"

– and further afield, including British tabloids:

• The Daily Mail: "Furious Wikipedia co-founder storms out of interview after refusing to answer basic question")
• The Daily Express: "Moment guest storms out of huge interview and leaves host speechless: 'I'm done''

The Times of India noted that the incident has widely circulated on social media (see e.g. a Reddit thread with over 6,700 comments). – S, AK

An article by The Intelligencer titled "Grokipedia Is a Warning" finds that Elon Musk's "Wikipedia clone is ridiculous" but "also a glimpse of the future":

Grokipedia, and Musk's AI projects in general, invite us to see LLMs as powerful and intrinsically biased ideological tools, which, whatever you make of Grok's example, they always are.

A somewhat similar argument is made in a London School of Economics blog post by Patrick Gildersleve (author of several peer-reviewed research publications about Wikipedia). He argues that Grokipedia has essentially fallen flat, with little Google visibility and dwindling traffic. But Gildersleve notes that Wikipedia is operating in an ever more hostile political environment, and describes Grokipedia as

a warning shot for AI’s real impact on Wikipedia and open knowledge. [...] AI is a threat to Wikipedia, but Grokipedia itself is little more than a politically charged sideshow to the deeper battles underway in the digital knowledge ecosystem.

An article by disinformation scholar Renée DiResta in The Atlantic, titled "The Right-Wing Attack on Wikipedia", states:

The free internet encyclopedia is widely used to train AI. That's why conservatives are trying to dethrone it.

As for the Grokipedia concept, she says:

It's pure algorithmic output with no community, no transparency, no clear process for dispute resolution. The irony is striking: Even as Musk and his friends attack Wikipedia for supposed bias, he is building something far more opaque and unaccountable.

As a concrete example, DiResta highlighted serious issues in the Grokipedia article about herself:

The remarkably thorough article about me contains nonsense that conspiracy theorists entered into congressional proceedings—including claims that my former research team at Stanford Internet Observatory censored 22 million tweets during the 2020 presidential campaign. [...] I reported these issues via the Suggest Edit tool included in Grokipedia’s user interface—so far, to no avail. On Wikipedia, I could appeal to an editor by dropping a note on a Talk page. But Musk’s version misses what gives Wikipedia authority: human consensus.

However, in a post on her Substack several days later, DiResta reported that Grok had resolved these:

Grokipedia also has hallucination issues—a known challenge with AI, but a particularly big one for an encyclopedia. I experienced this firsthand [...] I flagged the errors for the chatbot. Tonight, around two weeks later, just as I was getting ready to publish this post, it finally fixed them.

Journalist and novelist Stephen Harrison discusses Wikipedia and the role of an editor community in a Slate podcast titled "Wikipedia Enters the Culture Wars" (transcript).

– AK, H

"Scandals Erased, Editors Paid: How Big Law Firms Try to Control Their Wikipedia Pages" – that's the headline of a law.com article (published back in September, archive) that told its readers:

A deep analysis by Law.com shows how some law firms pay editors, flout the rules, whether consciously or not, and remove controversies to curate their image on one of the world's most popular websites.

The article looks at the editing history of several law firm articles. It gives examples of firms using PR consultants who openly disclosed their work in line with Wikipedia's rules and quotes a commercial editor who makes undisclosed edits. It also covers the politicisation of law firms:

Law firms with connections to U.S. President Donald Trump or his executive orders often show evidence of this on their Wikipedia pages.

The Verge published an article on Jeffrey Epstein-related Wikipedia editing. The article mentions and indeed quotes from a March 2020 Signpost piece by User:Smallbones. See the current issue's Disinformation report for more. – AK

A recent article in Nonprofit Quarterly, titled "We Stood Up: Organizing at a Feminist Nonprofit", details the author's critique of a 501(c)3 organization whose "founders were so assured in their politics that they placed the word 'feminism' in the organization’s name", but that

over time, and with new leadership, politics were treated more as an impediment than as a part of our mission. For example, developing a material commitment to complement the organization’s Black Lives Matter statement was a challenge, as was sustaining a disability justice praxis, or taking a clear stance against apartheid, genocide, and settler colonialism.

Leaning both on provocation and intention, I began to assert that we ought to rename the organization, replacing "feminism" with "women" to more accurately reflect our focus on the representation of careered, cisgender women. (The organization, not surprisingly, did not budge.)

The name of the organization containing the word "feminism" is not directly provided in the article, but the author's biography on NPQ leaves little doubt which organization it is referring to: "She was terminated from her nonprofit job during a union campaign; she continues to organize with her former coworkers in Art+Feminism Workers United!"

(The Wikipedia article about Art+Feminism currently describes it as "an annual worldwide edit-a-thon to add content to Wikipedia about women artists, which started in 2014." According to the most recent Form 990 for Art Feminism Inc, the nonprofit had a budget of about 0.5 million USD in 2022/23; see also meta:Art+Feminism User Group. Its current grant request for the Wikimedia Foundation's "Wikimedia Community Fund" is for $1.3 million USD over three years, i.e. 2026–28. It mentions that the organization currently has three staff members – two full time and one part time – and that "[t]he biggest change [regarding staffing, from previous years] is sunsetting the Regional Ambassador Program. [...] This also resulted in the layoff of the Program Director role where the majority of the role was managing this program." A separate document details the rationale for this change.)

A more recent post by the aforementioned "Art+Feminism Workers United" follows up on the NPQ piece, clarifying that the organization had indeed justified the NPQ author's firing by "citing a 'lack of work' due to the elimination of the regional ambassador role as the reason", but still attempts to put it into the context of wider threats to open knowledge:

Across open knowledge sectors, the threats to our work are ongoing: whether it's recent threats at the Wiki North America Conference, the far right attacks on Wiki groups like Art+Feminism, or the undemocratic removal of two candidates, including the only woman and the only openly queer candidate, from the Foundation’s Board Election ballot, it can feel intimidating to do this work in a public sphere; but more than ever, the time to ensure A+F is operating in alignment with our values is now.

In an earlier post from around July 2025, the collective – all three of them still employed at A+F at the time – had focused less on external politics and more on internal organizational issues, which they claimed to have caused a concerning decline in the organization's impact:

Since 2019, the number of Art+Feminism events has dropped by 80%, and attendance has declined by 82%. Our global footprint, which once spanned 43 countries, now extends to only 27. Even our core Wikipedia contributions, new articles and edits, have collapsed by more than 94% in just the past year.

– H

• Mammoth mummy sex-change: 404 Media reported on a paper just published in the journal Cell, "Ancient RNA expression profiles from the extinct woolly mammoth". The main result of the paper was that 39,000 year old RNA can now be examined by scientists. An interesting sidebar is that the mummified wooly mammoth Yuka has been reclassified from female to male. 404 mentioned that this news is so new that even Wikipedia hadn't reported it yet. This oversight has now been corrected. – S
• Praise for pictures and photographers: Some favorite wildlife photos of the Wikimedia Commons Pictures of the Year contest were shared by AOL and Bored Panda [1].
• Stop scraping me, bro: Wikimedia would like AI companies to use its paid API product and properly attribute scraped content. A recent Wikimedia blog post to that effect sparked coverage in TechCrunch as well as an article by CNET.
• "How to save Wikipedia from AI": An article in the Chicago Tribune (widely syndicated) takes another Wikimedia recent blog post that reported a reduction in human pageviews, attributed to AI use, as its point of departure (cf. Signpost coverage: "Wikimedia Foundation reports 8% traffic drop since last year due to 'the impact of generative AI and social media'"). It argues that students need to be taught to click through to sources: "Education is at its best – more magical than an LLM ever will be – when it moves us from being passive recipients of knowledge to active explorers who reason about what we see. The challenge we face is to move from being passive users of AI to engaged citizens who can verify AI responses that, as a common expression goes, are 'frequently wrong but never in doubt.'"
• Q: Where do babies come from? A: Nazis.: "Priest horrified to learn he is Himmler's grandson after recognising Nazi's mistress on Wikipedia" (The Independent)
• Annual rankings start coming in: Visual Capitalist gives us "Ranked: The Most Viewed Wikipedia Pages of 2025 (so far)".
• ... and in reverse: 20 Great Articles and Essays about Artificial Intelligence at The Electric Typewriter includes "Wikipedia's Moment of Truth" which appeared in July in The New York Times.

• No theories on Jesus' sexuality, please: Christian magazine Charisma questions the article Sexuality of Jesus, whose lede says it covers "alternative and fringe theories" on Jesus's sexuality. Charisma objects, saying the Wikipedia community "allow[s] activist contributors to insert unorthodox views under the guise of academic balance".
• Saintly Wikimedia donors: Canadian news satire site The Beaverton pokes fun at Wikimedia's ongoing banner campaign, imagining a reader who achieves sainthood by responding "not with apathy, but with a one-time contribution of the minimum listed amount ($2.75)". But with fame comes scrutiny, and a fictitious journalist calls her "a fake and a phony! ... My sources have shared that she didn't even mean to donate to Wikipedia, she just accidentally clicked on 'Payment Method' and her Google Chrome extension already had the credit card info filled out. And it wasn't even her credit card, it was her dad's!"
• "Wikipedia's Antisemitism": An article published by Orthodox Jewish outreach group Aish HaTorah welcomed Jimmy Wales' call to action regarding the Gaza genocide article, and said additional bias, antisemitism, and falsehoods could be found in the articles Jewish Supremacy, Well poisoning, Palestinian genocide accusation, Al-Ahli Arab Hospital explosion, and Jews.
  • See above: Ashley Rindsberg in Spiked also highlights Israel, complaining of "The hostile takeover of Wikipedia". Deborah Lipstadt tells the Jewish News Syndicate she has "lost faith" in Wikipedia's neutrality on Israel, adding: "It is pretty close to, but not yet identical with, the BBC."
• American History X, but from a Nazi's AI's point of view: "Grokipedia cites the neo-Nazi website Stormfront as a source 42 times", NBC News reports based on Cornell University research that analyzed 883,000 articles. The Stormfront-love ironically included its citation six times in Grok's rewrite of the American History X overview. The NBC writers noted "Wikipedia and Grokipedia rely on some of the same sources, with 57 internet domains showing up in the top 100 source lists for both encyclopedias. But the two sites often diverge, especially when it comes to sources Wikipedia editors consider unreliable or even blacklist", another example being conspiracy theory site Infowars.
• Wales gets Forked: The New York Times's Lulu Garcia-Navarro interviewed Jimmy Wales in the NYT podcast Hard Fork, titled "How Wikipedia Is Responding to the Culture Wars" (part of the book tour, see last issue). The podcast runs almost an hour long, with no walkouts as far as we could tell.
• Superpowered, and lawsuit-proof: "Wikipedia's army of volunteers is its superpower", says The Chronicle of Philanthropy. The distributed volunteer base makes it "hard to sue even as it has been drawn into partisan political wars" (though we admit we aren't gods).

• Ketchupy kudos: Congratulations to the editors of Ketchup, which article plays a part in the plot of the Keanu Reeves film Good Fortune, noted in reviews appearing in The Independent and The Wall Street Journal (also syndicated by Hindustan Times).^[1]

This article uses Jeffrey Epstein's emails (which were recently released by the U.S. House Oversight Committee) to follow up on a Signpost article from March 2020 about Epstein. These emails are not part of the Epstein files, which are required by Epstein Files Transparency Act to be released by the DOJ within 30 days of passage of the act, i.e. by around December 19.

1. Optimizing good information to push down the negative: using new content (blog, websites, press releases, public network profiles, google images, etc) and search engine optimization. Changing the Wikipedia profile. Notes Wikipedia below.

...

WIKIPEDIA

This is a tough nut to crack. And I need to do more research on this. Wiki comes up first on the Google list due to its powerful domain and contains totally lopsided and damning content on you.

But we'll crack it. On the surface, Wiki is controlled by a morass of copyediting geeks who have nothing better to do than to discuss reference tags etc. It is also 'the people's' encyclopedia, dictated by the tyranny of the majority—so no objectivity at all.

Current Strategy - 1. The Reputation group have a copyright team dedicated to balancing the content on Wikipedia and gradually sectioning the bad stuff down to sub categories in the profile. They don't seem to be able to eliminate the bad though. 2. So, I think it's worth seeing how people like Prince Andrew managed to create an absolutely stellar profile—or Bill Clinton for that matter. No mention of his impeachment or lying to the Grand Jury. It barely touches on Lewinsky etc. 3. There are also protected pages on Wiki but I need more time and research.

...

— Christina Galbraith to Jeffrey Epstein, 12/16/2011, HOUSE_OVERSIGHT_025233 (via Jmail)

On November 12, 2025 Republican members of the US Representatives on the Oversight Committee released about 20,000 email records from convicted sexual predator Jeffrey Epstein. Seventy-two of these records referred to Wikipedia in some way, but not all are about whitewashing with Wikipedia, e.g. here's one about genetics. About 20 of the emails include Epstein and his hired wikiwashers discussing their plans, strategies, and even progress reports on editing two Wikipedia articles, Jeffrey Epstein and Jeffrey Epstein VI Foundation. They include Epstein's complaints about the cost of the wikiwashing.

In March 2020 The Signpost published the article "Jeffrey Epstein has asked me …" by this reporter using Wikipedia's extensive edit histories to document how this wikiwashing was done, and how it affected people outside this encyclopedia. That earlier article is republished in this issue at From the archives for easy reference.

This article shows in detail, in the words of Epstein and his employees, how their planning affected the articles and how other Wikipedians resisted their efforts.

The emails come from a disorganized database given to the Oversight Committee by Epstein's estate (see 404 Media, "The Epstein Email Dump Is a Mess"). They are occasionally crude or vicious, and often lack proper grammar and conventional spelling. I've generally left them as is, without copy editing, to preserve their spirit, except in cases where a few changes are needed in order to understand them. Mostly, though, they are reasonably calm or even boring discussions of how to whitewash Wikipedia and the rest of the internet. This article has greatly benefited from the work of others who provided data and new information. User:Dflovett first reported the existence of the email records about Wikipedia on his Substack blog "Did Jeffrey Epstein edit his own Wikipedia article?" He also performed the difficult task of sorting out the files and formatting several of them into readable form. The Verge, wrote an excellent article How Jeffrey Epstein used SEO to bury news about his crimes which gives additional information and interpretation from an expert. Later, Riley Walz and a collaborator also did a remarkable job in taking the disorganized data dump of 20,000 emails, and putting it in a readable format at Jmail. An earlier attempt at journaliststudio.google.com can be checked for additional people who were copied on the email, or in one case for an attachment that wasn't included in Jmail.

The Epstein article was created on August 4, 2006 featuring Epstein's criminal indictment for the solicitation of a minor for prostitution and included a reference to the charge within the first 4 edits. The first apparent whitewashing occurred in July 2007 by an anonymous IP editor 63.165.175.250 who removed the text about the charge and the reference, and added complimentary information about Epstein. They made a dozen similar edits through October 2007. These methods of whitewashing are very simple – just remove controversial, but cited facts, and add complimentary information. When other editors put back the cited information, the whitewashers just use the complimentary information to push the cited facts to the bottom of the article.

These edits took place before the earliest available email evidence, but the strategy described in the emails was the same: brute force removal of cited information combined with the addition of complimentary info to push the ugly facts away from the readers' view.

Epstein's email exchanges with several practitioners of online reputation management (ORM) begin in 2010. Their strategies offered to Epstein were not limited to Wikipedia, but aimed at censoring the entire internet. Most of these strategies were as simple as they were brutal. Any website or newspaper article in the top dozen items from a Google search that told the truth about Epstein's crimes were to be forced off that page by complaints to Google or the website owner or by other tools in the search engine optimization (SEO) toolkit. Any websites which presented complimentary information on Epstein were to be promoted on the Google results page, perhaps by linking these articles to each other or writing new articles with those links. Also articles were to be posted that simply confused matters such as one about Jeffrey Epstein (plastic surgeon). The strategy was not limited to Wikipedia, but since this encyclopedia was generally listed as the top article on the Google search page, it was of special importance.

As Al Seckel wrote on December 10, 2010:

Wikipedia was an important victory, as it will always be at the top of the search engine results. (N)ow the head lines do not mention convicted sex offender or pedophile. Instead, Philanthrophic work, Epstein Foundation, Promotion of Scientists. … Your wiki entry now is pretty tame, and bad stuff has been muted, bowlerized, (sic) and pus(h)ed to the bottom. … We hacked the site to replace the mug shot and caption, and now has an entirely different photo and caption. This was a big success.

We pushed the Edge all the way up to the front page, where it was previously buried on page 5 of google search.

We have promoted the other jeffrey epsteins, and other pages are also filled with your material.

— Al Seckel to Jeffrey Epstein, 12/15/2010, HOUSE_OVERSIGHT_022216.txt.pdf (via Jmail)

Al Seckel may have been the first reputation manager to work for Epstein. His low pricing for his services to Epstein calls into question his experience in ORM. In another context, The Telegraph calls him "a top-notch charlatan".

His connection to Epstein may have been made through Seckel's domestic partner Isabel Maxwell, Ghislaine Maxwell's older sister. Ghislaine was Epstein's former girlfriend who was convicted for child sex trafficking.

Epstein and Seckel argued about the cost of the whitewashing. Epstein complained

I was never told never, that there was a 10k fee per month,, you inittaly said the project would take 20.. then another 10. then another 10..
— Epstein to Seckel, 12/16/2010, (via Jmail)

Seckel responded, in part,

My initial estimate, given to me by Pablos, was 25, and that was based solely on his quick look at the situation, and not knowing what was really out there. In fact, as I have repeated many times, the job was far far worse that originally expected, you have a dedicated group of people trying to undo and damage you, including now, they have started up in the last week full force again, as it is obvious that they can't fux with your wiki page any more as we blocked that..

...

Then, there was the issue of trying to create additionally on you a positive web presence, with the science and org sites.

I spend literally four months of non-stop work, creativity, and my own political capital to get you this so far, which saved you not only time, but countless dollars, and isn't something that can be readily bought. ...

We were trying to fix up your mess. I didn't create it. Just thought it would be something to help. This was NEVER about trying to pull money out of you, and fact, we have don't everything possible to keep the costs down considerably.

I must talk to you about the island thing asap. When can we do that?

— Seckel to Epstein, 12/16/2010, (via Jmail)

The Verge notes the argument about the price (Archived), and quotes an SEO expert, Rand Fishkin, saying that the cost should have been $100,000 to start with more than a $10,000 monthly maintenance fee. "The prices just looked insanely low to me. Here's a billionaire who supposedly is worried about his reputation as a fucking pedophile coming out in public arguing over a few thousand dollars. Honestly, the chutzpah is insane."

Epstein's dissatisfaction with Seckel's efforts (and prices) seems to be reflected in an apparent ORM proposal from Osborne & Partners LLP, a UK PR firm, dated 14 June 2011. It is long and detailed and emphasizes material in the U.K. press. Apparently this proposal was not accepted. (This link is not to the Jmail site, because Jmail does not include attachments to the emails.) The following year Osborne, who was also a venture capitalist, reportedly lobbied his contacts at Epstein's behest to install Jes Staley as CEO of Barclays Bank. Staley was Epstein's long time banker at J.P. Morgan & Co. and his work with Epstein was involved with his departures at both Morgan and Barclays.

Another long and detailed apparent ORM proposal was sent to Epstein by Christina Galbraith on Dec 16, 2011. She was recommending the firm Reputation.com and stated

You need people who eat, drink and sleep algorithms and search engine optimization. I've researched several companies and the most skilled, discrete, award winning and comprehensive is a company called, www.reputation.com based in Silicon Valley.

...

I would serve as a liaison between you and Reputation. I would monitor their progress and provide them with all positive content that they need. They would make sure that the content has maximum algorithmic potential. You could keep your positive content simple (using you Edge.org summary which is good), or expand your content with updates on current science work and scientists. The benefit of the former, is that it keeps you more anonymous. The benefit of the latter is that is algorithmically associates you to a larger pool of known scientists, further pushing up positive content. My advice would be to combine the two: a simple repetitive bio summary of you but with a larger list of scientists added on. But I would ask the Reputation team about this: saying that the main goal is to enhance anonymity and algorithmic associations at the same time.

— HOUSE_OVERSIGHT_025233 Christina Galbraith to Epstein, 12/16/2011 (via Jmail)

Reputation.com may not have been hired by Epstein, but Galbraith certainly was. She served as his in-house publicity director and published a dozen or more press releases at PR Newswire giving her name and work phone for journalists to contact, according to The New York Times.

After this proposal, Galbraith was an influential figure in the ORM project, being copied in 4 email threads on the topic and serving as a go-between for Epstein and the reputation managers.

Galbraith has an interesting background. Her father Evan Galbraith was successful on Wall Street and was appointed as the U.S. Ambassador to France under President Ronald Reagan. He was also a friend and supporter of William F. Buckley and the National Review.

Buckley died a month after Evan Galbraith and Christina Galbraith wrote an appreciation of their long friendship in the National Review (archived) in her father's voice. In 2013 she also wrote a puff piece in the National Review about Epstein's donations to Harvard University.

Galbraith's emails with Tyler Shears, a reputation manager who worked for Epstein in 2014 and 2015, were very specific suggesting that she was managing or closely monitoring him.

Im at a Kinko's.

...

I see that you're boosting non-website url's -- are you sure this is the right approach? (vs. boosting Jeffrey's sites?) the .org and foundation sites are slipping down and the USVI and science are permanently off the first page.

Thanks for you input.

Christina

— Galbraith to Tyler Shears (via Jmail)

Jeffrey Epstein and his often changing group of online reputation managers planned, edited, and tracked their whitewashing of Wikipedia as part of a larger effort to keep reports of Epstein's crimes off the internet. The effort to whitewash Wikipedia was one of the key parts, often the first step, of these efforts. Given the large amount of money spent and the number of whitewashers hired, it is surprising that Wikipedians were at all successful, but they continued to update the articles and helped frustrate Epstein's coverup. Though Epstein was partially successful for a decade, he could not succeed in the long run.

This type of whitewashing is powerful. Given the upcoming legally required release of of the Epstein files, you might expect to see similar reports in the future, perhaps in the next issue of The Signpost.

Wales, Jimmy (2025). The Seven Rules of Trust: A Blueprint for Building Things That Last. New York: Crown Currency. ISBN 978-0-593-72747-8.

The Seven Rules of Trust was released on October 28, 2025. Not many books have the claim to fame that this one does: Jimmy Wales is featured prominently on the cover (with co-author Dan Gardner in much smaller lettering). But does this book actually have something fundamentally important to say about Wikipedia? I would argue that it depends on the audience reading it. I read this book with my Wikipedian hat proudly on and that affects my perspective quite strongly. Therefore, I think a more casual reader may find the book to be more insightful than I did.

For the sake of transparency, I am declaring the conflict-of-interest that I have with Jimmy Wales that is already on my userpage. I'd also like to mention that I received a free copy of the book from him about a week before the book was officially released. Despite this connection, I don't think I have let this cloud my review. Feel free to take my opinion with a grain of salt regardless. I won't be offended if you don't trust my review to be purely objective.

So what is this book actually about? As the title implies, it's about trust. While the book lists both Jimmy Wales and Dan Gardner on the front cover, I perceived the book to be written in "Jimmy's voice". I say that because there is a lot of first-person language like "I". There are plenty of personal insights about Wales' life and there are none about Gardner's. The back cover only includes a photograph of Jimmy Wales.

Wales opens the book by talking about a very personal experience. His daughter had meconium aspiration syndrome and he had to make a decision as a parent on whether to accept an experimental treatment for the condition. I can only imagine the stress of that situation. He couldn't really find anything other than scientific papers that were unintelligible to laymen like him. Wales had this experience about a month before launching Wikipedia. He describes it as fortifying his belief in this website's importance. It's a touching story and I wholeheartedly believe its veracity. I'd personally advise more caution in the strength of our medical content than Wales does when he says "Worried parents will never go through a similar experience today because Wikipedia has excellent articles" on page 16 because our disclaimer about such content exists for a reason. I also admit to being a bit wary that Wales does not mention Larry Sanger until page 152. Another personal experience that I appreciated reading about was Wales' tribute to Jo Cox.

I would argue that this is not strictly a book about Wikipedia even if there are Wikipedia-related anecdotes scattered throughout it. I think that The Editors by Stephen Harrison (Signpost reviews) actually features Wikipedia more prominently despite technically only being "inspired" by it. In contrast, The Seven Rules of Trust reads like your standard business-related book with Wikipedia laying the foundation. There's numerous pages that have absolutely nothing to do with Wikipedia, discussing Uber, AirBNB, Quaker Oats, and the subreddit r/ChangeMyView, to name a few examples. Even Jeff Bezos gets a positive mention. What all of these subjects have in common is something Wales perceives as being relevant to the matter of trust. I'm not saying these reflections aren't meaningful. I was simply expecting to see more about Wikipedia. The limited content about the hopes and dreams people had with Web 2.0 made me long to read more. Wikipedia really is something magical when compared to the toxicity that runs rampant across most of the Internet and the book is at its best when it focuses on that.

I really enjoyed the mini-biographies of Wikipedians Annie Rauwerda, Emily Temple-Wood, and James Heilman. I also enjoyed the much briefer quotes of other Wikipedians. I know that plenty of Wikipedians were interviewed while this book was being written and I'm disappointed that this type of content was only present in a handful of pages. I admit to also being confused about why for-profit companies were repeatedly held up as examples of institutions that were trusted. These comparisons seemed out of place, akin to apples to oranges. I think a crucial part of what makes us different is our independence from corporate interests. I think other large collaborative projects like the Internet Archive, the Great Backyard Bird Count, Distributed Proofreaders, and Open Street Map would've been better comparisons.

I'm genuinely shocked that page 47 made it through the publication process. There, Wales describes the Lynching of Horace Maples as part of the dark history of his hometown, Huntsville, Alabama. I understand the argument that Wales was trying to make. There's a dark side to the Internet and not everyone comes together to do something great. The Gamergate analogy made sense. But then Wales started writing about the lynching as an example of volunteerism and focuses on that aspect way too much for my liking. I'll emphasize that Wales sees this incident as evil. But the framing felt off, made me wonder if anyone at Penguin involved a sensitivity reader at any point, and felt like it cheapened the horror that this history contains. I encourage people to read the full page that prompted my reaction here. I'd argue that it's actually worse in context than without it, but it's possible interpretations on this may vary. For even more context, this is the following page. As I said, I understand the argument he was trying to make here, but I don't think this was a good way to go about it.

Chapter 4 gets across the principle of assume good faith decently well. However, I do think the example written in the book takes the practice farther than how it is actually applied. Wales describes a situation in which a new editor erases a paragraph in an article about American politics and replaces it with slanted content cited to an obscure blog. He believes that the way Wikipedians would react to this situation would be to revert the edit (correct) and then open a discussion on the talk page explaining why they disagree with it (this generally wouldn't happen from the editor doing the reverting). I think there's a sincere value to writing edit summaries like this when dealing with good faith edits. But I also recognize that most editors will simply offer a quick link to the relevant policy and see that as sufficient. Whether that's how things should be is a different conversation, but I'm worried this example will give casual readers unrealistic expectations. I do agree with the message that it's difficult to have constructive conversations if people are making personal attacks. Civility is important.

I admit that I'm also not a big fan of the AI tool that Wales describes working on in the conclusion of the book. I think the answer to people not understanding our radical transparency is to get better at teaching the public how to engage with Wikipedia and how we work, not to have an LLM summarize and name the editors that participated in the discussions that led to the current consensus. I get why someone would think it's a good idea to make our processes more understandable this way but it'd only do so on a surface level. I worry that the results would cause more harm than good.

Overall, I'd say the book is a decent lighthearted read, even if I can't ignore its very noticeable flaws. Despite that assessment, I don't hate this book. I'd rate it 3/5 stars. It's difficult to get across when something is done well because you don't have too much to say when it is. I think critiques just stand out to me more in comparison. I read a lot of books and that likely makes my standards higher as well. I'd suggest people borrow this book from a library if they want to read it. $42 CAD works out to 20 cents a page. That's a hard expense to justify for the average person.

The Signpost sat down with Wikipe-tan, the ... mascot of Wikipedia, to have a conversation.

The Signpost: Who are you?

Wikipe-tan: I'm Wikipe-tan, official mascot of Wikipedia.

TS: Official or unofficial?

WT: I don't care. That's the dumbest question in the world.

TS: Really?

WT: Absolutely.

TS: There seems to be a dispute.

WT: There's no dispute. I don't care. So yeah, say what you like. It doesn't matter.

TS: Well, isn't that, when it comes to Wikipedia, a problem? What are the facts?

WT: Yeah. [laughter] Well, it's not a fact. It's an opinion. So, you can have whatever opinion you like. It doesn't matter.

TS: But for you, you're the official mascot.

WT: Can I just say again, it doesn't matter. I've answered your question four times.

TS: ...

WT: You know what? I'm done. Thank you. Thanks.

TS: What's going on?

WT: It's stupid. Don't ask me stupid questions.

TS: How is it stupid?

WT: It's a stupid question.

TS: It was the first question.

[Wikipe-tan walks out of the room.]

Every year, Wikimedians from around the world gather for a new edition of Wikimania, the best-known and important convention involving Wikipedia, Commons, Wikidata and all the other WMF projects. However, not as many people know that several national and regional communities host their own yearly meetings, too: among them, there’s the Italian community, which recently met at the latest itWikiCon, held in Catania from 7 to 9 November 2025.

As reported by Wikimedia Italia staff member Simona Cannataro, almost 90 people took part in this itWikiCon, with more than 30 sessions being hosted within the rooms of Palazzo Biscari, where participants also had the opportunity to join edit-a-thons and other activities in their spare time.

The sessions included in-wiki related themes like advocacy activities, the upcoming Wikifunctions and Abstract Wikipedia platforms and the constant push-and-pull between Wikipedia and AI. More practical applications of Wikimedia projects were explored in detail, too, including the digitization of cultural heritage, university programs and partnerships with local associations. The 2025 itWikiCon was also the right occasion to celebrate the latest round of WikiRicci, the yearly prizes dedicated to frequent contributors of the Italian community, as well as the birthday of Wikidata, and a special collection of postcards headed to contributors of the Iranian community.

Finally, throughout the week and the days leading to the meeting, participants had the opportunity to visit some of the most iconic locations in Catania, including the Teatro Massimo Bellini.

The 2025 itWikiCon is just the second national event ever hosted in Southern Italy – following the 2023 edition in Bari – and the first one held on an island rather than the peninsula, specifically Sicily.

The southern regions of Italy do offer particularly rich cultural and natural heritage, but suffer from a chronic lack of economic resources and a shrinking population. According to the latest edition of a report by Fondazione Migrantes, an internal institution of the Episcopal Conference of Italy (CEI), more than 1.6 million people have left the country from 2006 to 2024, with Sicily being the region that registers the highest number of Italian citizens living abroad – over 844,000 people; the report also found that over one million citizens have moved from Southern Italy to Central and Northern Italy in the last ten years, with 48.5% of them being aged between 20 and 34.

The WikiSud collective, which was first created at the 2023 convention in Bari, has aimed to build a more solid network between Wikipedians who still live – or come from – Southern Italy, by bolstering up the number of active volunteers, initiatives, organized groups – including Wikimediani in Sicilia – and GLAM institutions. These ambitions were highlighted during a session at the 2024 Wikimania in Katowice, and have now been brought to the organization of the latest itWikiCon in Catania.

A member of the convention's organizing team, Giovanni Pennisi, shared partial results of a poll sent to participants with The Signpost. The participants were mostly active on Wikipedia, Wikidata and Commons; over 25 people said they had taken part in a national convention for the first time; while the majority of participants were aged 31 or older, about twenty of them were aged between 18 and 30; some users from Switzerland, France and Belgium took part in the event, and over twenty people received a travel scholarship. Finally, many of the participants stated that they felt "more like a part of the community", that they were "more motivated to contribute" and that they understood more clearly "the issues of Wiki projects and the community".

The success of the latest itWikiCon served as a step forward towards the ambitious goals of the WikiSud collective. It showed that, even in territories where volunteers face more systemic obstacles and many people still feel forced to leave in order to get better chances for their studies or their professional careers, local communities can still thrive and work together to bring a glimpse of hope and positive changes for their lands.