I had reported here on a study I was undertaking of steward action. This was not a collection of gripes against stewards. The vast majority of actions studied were at least arguably proper. There were a very few exceptions, almost entirely regarding Vituzzu, but I had not drawn that conclusion, it was simply that the data came up.
I had invited correction of any errors or information about missing data (i.e, if some action *looks bad* but was explainable by missing data, this could be provided or, at least, it could be said that there was a justification -- which any steward could verify --) and I was not warned that there was anything improper about this study.
Yesterday one of the pages was deleted by Vituzzu. It was a list of involved IPs in what I call the "Santarelli" case. There was no data there that was not readily found in public logs. There are similar collections on SSP pages on it.wikipedia, as I recall.
Vituzzu gave a reason: privacy violation. I've seen this before. I was in communication with Daniele Santarelli, who has been accused of being behind various alleged sock puppets, by Vituzzu. The real Santarelli acknowledges certain edits as being his, but denies any other editing. This was consistent with the data I'd seen; the other IPs were not from his location, or, at least, unclear.
Since the data was all from public logs, created by the steward, I requested that Vituzzu email me the data. I did not request undeletion. I also requested that the generally sane global admin PiRSquared17 provide me with the wikitext. I simply didn't want to lose the work, I had saved all the other pages, but not that one. The global admin declined because of the allegation of "privacy violations."
(Some of these pages represented weeks of research, and the vast, vast bulk of it was just there for completeness.)
Today, the steward Billinghurst dropped onto my talk page. This was the conversation:
DanielTom dropped by after this with a comment that could get him blocked, unfortunately.What the hell are you up to?
Do I see a similarity to Don Quixote here? You seem to be running around picking fault and analysing action of stewards for no bigger purpose than to either pick a fight, or to prove a contentious point? Or is it to just be troublesome? I am not seeing any particular benefits. What ultimately do you think that you will achieve by such an analysis? What do you think is likely to be the result of nitpicking analysis? — [[user:billinghurst|billinghurst]] 15:07, 24 May 2014 (UTC)
:I have deleted all that rubbish, any further attempt to provoke people and/or engaging in collecting informations about actions without consent will lead to an immediate and definitive block on this wiki. --[[User:M7|M/]] 15:50, 24 May 2014 (UTC)
::(edit conflict with above) Billinghurst, my view is that we cannot design policy and assess performance without study of what is actually done. You have stated that review is welcome. That's what I'm doing. It is not "nitpicking," it is presentation and analysis of actual steward behavior. In a very few cases, I report behavior that might be questionable. In many more cases, I report behavior that may not be covered by policy, but where, easily, policy may need extension to cover the relevant situations.
::Can you give me an example of the specific "fault picked"?
::M7, I am accordingly, ceasing the study here, pending an appeal of the deletion. Thank you for making the matter completely clear. --[[User:Abd|Abd]] ([[User talk:Abd#top|talk]]) 16:02, 24 May 2014 (UTC)
:::I will block you if you just "ask" for undeletion. Is that clear? --[[User:M7|M/]] ([[User talk:M7|talk]]) 16:07, 24 May 2014 (UTC)
::::Again, thanks for being clear. It is quite useful. --[[User:Abd|Abd]] ([[User talk:Abd#top|talk]]) 16:10, 24 May 2014 (UTC)
It is very, very clear: it is not allowed to review steward behavior. Not even if the goal is to improve policy. Even worse is criticism of steward actions.
I had examined something on the order of 6000 steward actions. Only a handful were seriously questionable. There had been errors that the stewards promptly corrected. But there were still some quite remarkable actions by one steward, and they are obviously protecting him.
The deletion log, while it still exists. Notice that M7 deleted and Barras restored "for suppression." Whether or not the log remains after suppression, I don't know.
Now, the deletion reason given for the one page mentioned was privacy. However, the full study was deleted with this reason:
It is quite unusual to request revision suppression for pages "intended to harass/provoke" unless the pages include seriously offensive language. It can be argued that the pages contained data that could be used to identify a person real-world, but the only known person in that category was very aware of the study and was consenting to it.(Mass deletion of pages intended to harass/provoke other people, with many collateral damages.)
The actual identification was made by the steward whose action was documented. And it was incorrect. And, I assume, still stands in the logs and edit histories as documented.
No, this is totally obvious. The pages had existed for weeks, they were announced on the Forum (seeking participation and correction), and there had been no warning on my talk page or serious objection. One of the pages deleted was a Talk page with friendly comment by a steward whose actions had been described on the attached page, and a global sysop, as I recall.
No, at least one or two stewards realized that this data showed abuse as the community would agree if it is brought to them. I had seen this early on, and began the study because of two incidents I'd seen, both involving the same steward. As the study proceeded, data collection was, I recognized, warped toward collection of data on that steward. Seeing this as a problem, I made the study generic. That study actually showed that the vast majority of steward actions, as far as I'd gotten with review, were at least arguably proper.
I had not come to the point of a consideration of the specific steward. I'd started by looking at Billinghurst's actions within the 5000 studied first. That's because Billinghurst was the most active steward, only one other steward even approached his lock count. There was no claim that Billinghurst had done anything improper.
There is some routine action that is clearly well beyond what was contemplated when lock policy was originally written and discussed, but it can easily be argued that this is *necessary*. But the nature of global locks is such that, when action goes well beyond that original intention, as expressed in policy, restraints or protocols become necessary.
Stewards do not want any restraint, obviously, and the only steward where I saw a *pattern of abuse* had explicitly complained about being impatient with policy. However, this might not have made it into the final draft. I was not planning on filing an RfC on Vituzzu. The plan was to propose policy changes, and RfC only if needed, and not to focus on some alleged Bad Guy.
Anyway, without community support, this is totally dead. Billinghurst has mentioned that the issues I was addressing are covered by the Ombudsman Committee.
That is preposterous. The Ombusdman Committee does not have the labor to do the kind of study I was doing. They would review *specific cases*. No, what is needed is policy review, and that is the purview of the community.
I'll go to the Committee, I assume at this point, but only to deal with the massive page deletion, the threat to block me if I merely request deletion according to standard procedure, and suppression, if that is done.
These were abuse of tools, as serious or worse than anything I'd collected, and certainly more obvious.
If Wikipediocracy wants the page texts, let me know. I have most of it.
(What I'd seen in the past was that Vituzzu was the most abusive steward, by far, with his actions being reversed by other stewards, on occasion, while he continued to complain (and even continued to lock after his actions were reversed). But, then, M7, several times, was the first to jump in, in discussions on-wiki, with reinforcement of Vituzzu's actions as "obvious" and any questioning of them as disruptive. M7 also revealed that while there was no on-wiki discussion on a matter I took to stewards, there were hundreds of emails between stewards. Now, stewards are allowed to communicate off-wiki because of privacy issues. However, off-wiki decision-making is obviously being done far outside that necessity.
Vituzzu and M7 are both Italian administrators as well as stewards. And the massive cross-wiki disruption I had seen originated with that wiki. Basically, I saw a quick-to-indef block, which commonly can lead to sock puppetry, as a naive user is completely outraged by the insanity of it all. Then escalation to global locks, by Vituzzu, merely based on *editing elsewhere* without disruption, then more sock puppetry as the user attempts to circumvent this, and the rest of the community, not realizing what has happened and not understanding, then blocks the user for "sock puppetry" or "block evasion" -- even where the user wasn't blocked locally, and then anyone looking at the CentralAuth pages sees blocks on multiple wikis, hey, this must be a really disruptive user! He really shouldn't be socking, right?
(Yet I found explicit statement by a steward that the registration of alternate accounts globally is not contrary to policy, and discussions where it is claimed that global locks are not global bans have explicitly said that users are free to create new accounts if locked, as long as they don't repeat "disruption." But what is being seen, constantly, is that alternate accounts are being blocked on sight. Usually that is because of a plausible claim of intention to run a spambot, the accounts are sleepers, but .... this usage creates a huge number of locks, and problematic locks are then hidden within a landslide, unless someone actually studies the flow. (Or attacks a specific steward, and specific actions, which is then vulnerable to a claim of "cherry-picking")
That indef/sock/massive blocks/locks sequence, as far as I have seen, *exactly what happened* in the Santarelli case. And what appears to be a major academic source, independently published, academically edited, on heresy in the early modern Catholic Church was, of course, unilaterally blacklisted by ... Vituzzu, probably because Santarelli is the appointed managing editor. Along with all of Santarelli's books being removed from bibliographies as "cross-wiki spam." And Vituzzu revert warring with local users over it and threatening to use steward tools to get his way. Vituzzu's actions explicitly claim that the books were spammed by the author, and that, then, outs the editors who placed the links, or attempts to. So he is covering up his own privacy violations.
And in one case, Vituzzu actually locked a user who had revert warred with him (and Vituzzu had reverted others besides that user, and the user was merely removing a speedy deletion template Vituzzu had placed. This was on en.wikipedia, and also on nl.wiki, so the user was following policy and Vituzzu was violating it.)
However, I was very careful with those pages, as they evolved, to take the focus off of Vituzzu and onto actual steward behavior, overall.
What this was doing, though, was leading to a situation where Vituzzu's actions would stand out like a sore thumb.
So what are they doing? Why are they protecting Vituzzu? It is what communities do, they protect their own. If dissent in the steward community appears, it can reduce confidence in the entire concept of stewardship. There is nothing surprising about this, but there is nothing surprising about death, either.
This is where communities go if there is not constant vigilance. The "leaders" take over. Organizations ultimately are weakened and often fail as a result.
Billinghurst was handling the vast bulk of global account locks in the last 5000 locks in 2013. Basically, what happens as a few individuals take on a massive work load, is that they are led to "own" the process. The logs only showed a tiny fraction of the work that Billinghurst had to do to identify all those real or alleged "spambots" without any edits.
Vituzzu actually had relatively few actions, comparatively, that surprised me. That makes the relatively large numbers of questionable actions much more significant.
Wikipedia has a huge torrent of spam coming in at all times. Yet it does not allow the kinds of actions that stewards are routinely taking globally. These actions are made necessary by the community leaving this work to stewards, instead of the work being spread out over a much larger community. It's a structural issue that leads to ownership and burnout.
And they shoot the messenger.