Wikipedate on trial!

Gitz, mounting a defence that may be a little too passionate...

On 1 July some it.wiki users reported certain "conduct contrary to the UCoC outside Wikipedia" to the Arbitration Committee (ArbCom), asking to remain anonymous for their own protection. The ArbCom granted the request for anonymity and, by an "almost unanimous" vote (that is, by a majority), decided that the report would be heard by the full committee. No arbitrator considered themselves involved and therefore bound to recuse themselves.

On 14 August, I received an email from the Universal Code of Conduct Coordinating Committee (U4C) informing me that they had initiated a private (confidential) investigation into alleged doxxing and harassment on the Wikipedate forum. The U4C sent me 53 links in support of the allegations and asked me not to disclose them, without specifying which passages the filers regarded as violations of the UCoC. Some links point to blog posts and others to entire forum pages, each containing dozens of comments by different users. It is therefore impossible to tell which specific forum comments were being reported, or why. Moreover, some of the material is irrelevant: the first link is to a post that does not criticise Wikipedia or Wikipedians and does not mention any username other than my own; the last is to a blog that I do not run and to a post on which I did not comment.

To make sense of this mass of irrelevant or overly generic material, I had to enlist the help of ChatGPT and Claude, and then review the results myself, which highlights the absurdity of the situation: an accusation that requires the assistance of a non-human intelligence just to be understood — or rather, guessed at.

I tried to reply anyway, but inevitably my response was too long and boring (what on earth do they expect me to reply to?). This is the short version, which I hope is also a bit more engaging.

What Wikipedate is

Wikipedate is my blog, which I started on 21 August 2023. It is intended "for active it.wiki users and for anyone interested in free knowledge and online democracy". Since then, I have published around 80 articles, mostly my own, with occasional contributions from others.

I have always tried to create a welcoming environment while also setting limits. As early as September 2023, I wrote in a post that

criticism of the conduct of Wikipedians (...) is more useful and more convincing if it avoids insulting people. This is not just a matter of politeness and good taste, but of substance: Wikipedia could not exist without the passionate work of users who spend their free time blocking vandals, explaining how the project works to "newbies", making difficult decisions in conflict situations, and protecting and expanding the encyclopaedia. Even when they make blunders or come across as arrogant or hostile, we should not forget the respect they deserve.

The forum was launched in November 2023 and, to date, has attracted contributions from several hundred spam bots... We also have around sixty registered users, including about a dozen diehards. From day one, the forum has been governed by one fundamental rule:

The purpose of this site is to criticise Wikipedia, not people. Always ask yourselves whether you are criticising someone's actions or words, or simply insulting them.

We are particularly strict on doxxing: we do not republish the personal data of it.wiki users merely because that data has been voluntarily disclosed. We only mention an identity that has already been made public when it is relevant to the matter under discussion, such as in the case of a conflict of interest. In a recent thread, for example, I pointed out that Pierpao had revealed his real name on-wiki, but I asked that it not be mentioned on the forum because he might not have appreciated it, given that our pages are indexed by Google. His name has never been published on Wikipedate.

What is it for?

The purpose of the Wikipedate blog and forum is plainly not to harass or to doxx, but to document, discuss and criticise the activity on the Italian Wikipedia, particularly that of its administrators. We do it mainly because we enjoy it, but one can still ask: is it useful to anyone? Here are a few episodes we can be proud of.

Dark plots and invented plots. In May 2025, forum moderator TrameOscure ("Dark Plots") discovered that the article on the early twentieth-century Italian novel Manoscritto trovato in una bottiglia (Manuscript Found in a Bottle) contained a plot summary and a list of characters bearing no relation whatsoever to the book. The whole thing was propped up by nineteen precise footnotes: clearly an AI hallucination. Once notified, the author, an it.wiki administrator, replaced the invented text with a version that was still wrong. In the end, another forum user did the most surprising thing of all: he read the source and rewrote the article (thread, discussion).

Manosfera. This article, which exists in 26 languages, was speedily deleted by Gac as manifestly non-encyclopaedic, after Mastrocom had moved it to draft and deleted it as a "neologism that does not exist in Italian". The author opened a discussion at WP:RA and Superspritz turned him down flat: "administrators have no editorial duties (...) If the sources are not there, neither is notability". On the forum I shared a number of Italian academic and journalistic sources showing the opposite; four days later Mastrocom published the article.

Speedy and rash deletions. Giacomo Verde is an artist whose article was speedily deleted by eight different administrators, once even as a draft, despite the fact that ETS had published a collective monograph on his work in 2018 and numerous obituaries had appeared after his death in 2020. The article now finally exists, although its author has left the project (thread). The same goes for Claudio Colica: the article was speedily deleted three times and the draft three times, by five different administrators. In the end, the article was created and the deletion discussion closed unanimously in favour of keeping it (thread). Other speedy deletions include Italian brainrot, deleted three times and now present in over thirty language editions (thread); Max Angioni, deleted five times by four different administrators when he was already a presenter on the popular TV show Le Iene (thread); Maria Esposito, the co-lead in the television series Mare fuori, who was deemed non-notable by Kirk (thread); Matteo Mancuso, a guitarist praised by Al Di Meola and Steve Vai, whose article was speedily deleted by three different administrators (thread). Wikipedate has raised the alarm more than once about the many speedy deletions carried out without the required grounds — a damaging practice that few it.wiki users seem to care about.

Real-life patrolling. An administrator used to warn users, on and off Wikipedia, that evading a block would constitute a criminal offence under Italian law. I disputed this claim on the forum (comment 1 and 2). The legal helpdesk of Wikimedia Italia subsequently rejected the claim at the request of a temporary account (legal opinion, comment by the temporary account); ArbCom ruled unanimously that this off-wiki conduct breached the UCoC's prohibitions on harassment and abuse of authority (decision), and the community removed his administrator rights (removal discussion). Prior to this, on Wikipedate we had flagged several questionable actions by this former administrator (1, 2, 3, 4, 5), and I believe our criticism was justified.

The name in the block log. In 2022, an administrator published the real name of an anonymous user in the block log (log). During his 2023 reconfirmation, he described that user, who had said he was autistic, as the "village idiot", whose behaviour was reminiscent of artificial intelligence (reconfirmation). Nobody objected on it.wiki; indeed, another admin defended him strenuously in private correspondence with me. The doxxed user complained on the forum about the disclosure of his real name, and I, with his authorisation, reported the matter to Trust & Safety (ticket 69590). Shortly after a post mentioning the episode was published on Wikipedate — written so as to preserve his anonymity — his real name was finally removed (post by Arturo Toglie; see my comment at the bottom).

The Portuguese model. In July 2025, an administrator and arbitrator proposed that only registered users should be allowed to edit, arguing that the Portuguese Wikipedia had done the same without suffering a decline in contributions. On the forum, I pointed out that a change of that magnitude ought to be tested against the data. I examined the figures collected and published by the Wikimedia Foundation (WMF) and found that, on pt.wiki, the number of net, non-reverted edits has been falling for years. I asked the WMF staff for clarification, and they warned that, if the downtrend continued, it "could be catastrophic for the future of the project". I published it all on the forum and flagged it directly to an it.wiki user. The proposal was rejected (thread).

Rules and procedures. Above all, though, I am proud of our contribution to the debate on the rules and procedures of it.wiki:

Since October 2023 I have repeatedly criticised the absence of a public procedure for appealing against blocks on it.wiki (post 1, 2, thread). Such a procedure was adopted in March 2024, with a clause requiring users to wait until half of the block period had elapsed (six months for indefinite blocks) before lodging an appeal. I then criticised this limitation (post) and it was removed in May 2024 (discussion).
Since 2023 I have been pressing the point that it.wiki lacks a rule comparable to WP:INVOLVED, which is strictly enforced on en.wiki (post 1, 2, thread). In January 2024 Civvì opened a discussion — long and difficult, as it turned out — which led to the creation of Wikipedia:Terzietà in November 2025. Back in November 2023, a former administrator observed that the principle had almost never been applied on it.wiki "until a certain user 'dusted it off' recently", probably referring to my activity on Wikipedate.
Before it.wiki set up an Arbitration Committee, I had argued that the lack of such a body was a structural weakness of the Italian Wikipedia (2022 discussion, post, thread). Since ArbCom has existed, I have argued that it should be able to remove administrator rights (post 1, 2, thread). As is well known, after the troubled business of Pierpao's removal, the question was at the centre of an intense community discussion, unfortunately inconclusive, which I commented on in a recent post.
By contrast, the position I have argued for years in favour of creating on it.wiki a "real" system of requests for comment on article content and the project's rules has remained without practical effect. In my view, such a reform would help to avoid or resolve many conflicts and might even counter the shrinking of the editing community (see this post for more details).

I am not claiming that these outcomes are due to Wikipedate; I am claiming that the site addresses real governance and content-quality issues on it.wiki, and that many of the questions we raise there are later addressed in similar terms in the community's own discussions and decisions.

Where we did not go wrong

Phyrexian: criticism or hounding? Recently, a substantial part of our criticism has concerned Phyrexian's editing and actions (post 1, 2, 3, 4, thread 1, 2, 3, 4, 5). I accept that the volume and the duration of this coverage are probably the strongest argument against us — assuming that this is in fact what we are being accused of, which I do not know. I will not go back over the individual criticisms here; what interests me is the question of principle, and that of the rules.

Authority carries increased responsibility. Anyone who holds the administrator tools exercises a power that affects other people's activity, and must therefore expect closer scrutiny, including from outside the project. This, however, does not justify any and every criticism, as even a real problem can be pursued in a persecutory way. Nor is the fact that a criticism is documented with diffs enough in itself to make it acceptable. Criticism that is harsh, persistent or unwelcome may be legitimate; criticism that is well founded and documented may be intimidating.

Therefore, drawing the line between legitimate criticism and harassment or hounding may not be simple. However, the UCoC offers a clear criterion on this issue, albeit a debatable one, when it refers to behaviour "intended primarily to intimidate, outrage or upset a person", or behaviour "where this would reasonably be considered the most likely main outcome".

Intimidating, outraging or upsetting Phyrexian was not my intention, still less my primary one, nor do I believe it was the main outcome of what we wrote. What I wanted to do was flag problematic conduct, prompt some reflection on his administrator rights and possible editing restrictions, and obtain answers from the other admins on the merits rather than systematic defences. I wanted to highlight a tendency among administrators to close ranks, which makes it more difficult to correct mistakes and apply the rules impartially. And I also wanted to show how a procedure for requests for comment on article content, which it.wiki still lacks, could have prevented many issues and counteracted the abuse of power and influence by a single administrator. Besides, many of the issues raised on Wikipedate were later independently raised by other users during his twelfth reconfirmation, which ended with the removal of his rights. This does not prove that we were right, but it does suggest that our concerns were not baseless.

When I considered some of the criticism directed at Phyrexian to be unfair, I said so explicitly (comment 1 and 2). Moreover, our coverage essentially stopped when his administrator rights ended. This seems to me indicative of what we were trying to achieve: not harassment or hounding, but documented criticism of Wikipedia's governance.

.mau.: doxxing? The user .mau., an administrator and arbitrator, speedily deleted the it.wiki article about himself. The obvious conflict of interest was discussed on the forum, and it is fortunate that it was, because on it.wiki the matter was disposed of quickly: a fellow administrator wrote that nobody was better placed than .mau. himself to assess his own notability. When a user objected, another administrator assured him that the deletion had already been discussed at length in the admin-only channels.

Was this doxxing on our part? .mau.'s real identity had been public for years, inside and outside the project, and he had confirmed it again in the very discussion we were commenting on. So much so that the question came up again openly on-wiki at the April ArbCom elections, costing him re-election.

If discussing it on a blog constitutes doxxing but discussing it on-wiki is fine, then by definition external criticism of the projects is incompatible with the UCoC. In that case, it would be better to say so explicitly, either in the UCoC or in the decision that will close this proceeding.

Where we did go wrong

Wikipedate does not tolerate doxxing or harassment, but it does allow comments that would rightly be sanctioned as "personal attacks" on it.wiki. There is nothing odd about that: the standards of conduct governing discussion on Wikipedia cannot simply be transferred to an external site devoted to criticising Wikipedia. On Wikipedia, Wikiquette and the rules against personal attacks exist to enable people to collaborate on writing an encyclopaedia together. On Wikipedate, however, nobody is required to collaborate with anyone else or maintain the same level of collegiality and civility. Therefore, we can allow criticism that is more forceful, sarcastic, mocking or even coarse, provided it does not degenerate into doxxing or harassment, or into gratuitous and hateful personal attacks, especially when aimed at particular users or identifiable individuals. This is the standard we have set ourselves and strive to apply scrupulously.

However, with the help of ChatGPT and Claude, I reread the vast amount of material sent to me by the U4C and found some passages that may not live up to that standard. I am not linking to these passages here, in order to avoid giving further currency to offensive expressions aimed at particular users. I will, however, list them all, as I am certainly not trying to minimise this issue.

Of the 53 links sent to me in support of the allegations, at least seven contain questionable comments and one is a guest post. The first comment refers to an administrator as being "in the throes of senile delirium" and "out of his mind", while the second, also about this user, states that he "non sembra in grado di intendere e di volere" ("does not appear to be of sound mind") and is "pathetic". Another administrator is said to "not understand a fucking thing and need things explained to him with pictures", and I myself compared him to a "career politician", accusing him of insincerity. A third administrator is called "a disgrace to Wikipedia", and a fourth a "clown". A user who is not an administrator is called a "lecchino" ("bootlicker"), with the insinuation that he had ingratiated himself with some influential admins. Finally, there is a guest post that insults all Wikipedians — "nerds who never made it" is one of the kinder expressions used — with unpleasant sexual innuendo and further insults aimed at particular users ("camerata amministratore", using the Fascist form of address, and so on). Satire or not, I would not publish it today.

I set out these examples because self-criticism without examples is of little value. I have not yet altered the material in question, so that the U4C can examine it in the form in which it was published, but I intend to remove the unacceptable passages once the case is over — or sooner, if the U4C prefers. I am sorry: in these cases moderation failed and, in one of them, I failed personally.

However, this remains an unpleasant but small sample drawn from several thousand comments. Isolated cases in which we did not live up to our moderation standards do not demonstrate that the primary intent or main outcome of Wikipedate is harassment or doxxing. Moreover, I had never previously been made aware of these cases: nobody had complained to me about those passages or asked me to take action.

In any event, when a problem arises — whether we notice it ourselves or someone else points it out — the moderators step in. When a user opened a thread on an administrator's alleged conflict of interest and included information about his family, TrameOscure removed the post immediately. I restored it after removing the personal information and explained the reasons for our rule against doxxing. And when Civvì criticised the tone of some of our comments on it.wiki, I opened a thread on moderation that resulted in shared criteria (1, 2, 3) and the removal of offensive passages (1, 2, 3, 4, 5, 6, etc.). I have also had occasion to pull up a regular user, one of the forum's mainstays, over the expression "minorato mentale" ("mentally defective"), telling him that insults should be avoided, especially when they are discriminatory. By contrast, nobody on it.wiki had batted an eyelid when a similar expression ("bambini ritardati", "retarded children") was used by an administrator during his own reconfirmation; the criticism came precisely from the forum's users (1, 2).

A matter of public interest

None of the filers ever approached me. They were not obliged to do so, and nobody expects an alleged victim to go knocking at the alleged harasser's door. I can say, though, that if they had, they would have found me just as willing to engage as I was with Civvì and others (such as Cosma Seini here). They would have received an answer on the merits and, where appropriate, action to remedy or mitigate the problem.

Instead, here we are with a private investigation into the line between external criticism of the Wikimedia projects and harassment under the UCoC: an issue that concerns not only me, but anyone who writes about Wikipedia outside the project. Is it not a problem that an issue of this kind, intrinsically public, should be decided in secret not only — ça va sans dire — by the ultra-secretive Italian ArbCom (the "arbitrary committee"), but also by the U4C, whose proceedings are in principle supposed to be public? Is it not a problem that I was sent 53 links, with no indication of which passages the filers regarded as breaches of the UCoC and without a single precise question being put to me?

I have never been interested in knowing who the anonymous filers are, and I bear them no ill will. I imagine they took offence at some inappropriate comment, or perhaps merely at some legitimate criticism. But a report of this kind, handled in this way, personalises an issue that is not personal. I have written and repeated this more than once (for example here): "it.wiki's admins are not the problem. Despite the occasional poor showing (...) I would say, without exception, that they are acting in good faith and are genuinely committed to building and defending the encyclopaedia". The problem is not the admins, former admins, or the filers. Nor is it the Wikipedate users who criticise them off-wiki, often with good arguments. The point is that, like any community, Wikipedia needs checks and internal review procedures, but it also needs public discussion, even when that discussion is unpleasant or unwelcome and takes place on external platforms, in the media or other venues that the Wikimedia movement does not control.

That is why the forum remains open even to those who detest it. We invited Actormusicus, who had criticised us harshly, to write a post on the blog setting out his views in more detail and with greater visibility (thread). He did not take us up on the offer, but the invitation still stands for everyone.

Wikipedia is a free encyclopaedia — or we would like it to be. It would be strange if freedom ended where criticism of Wikipedia began.