Has WR discussed ways to improve the Wikipedia admin structure?

I am thinking one drastic improvement would be some automated system, such as adminship after 2000 edits, 6 months, or maybe both.

The obvious benefit would be to get rid of the political wrangling, the power of admins under the current system, the starting of new accounts to sneak through, the risk of sockpuppet admins, etc. There may be better metrics than edit counts or time on Wikipedia, but those together could work.

Of course you could maintain the zero-tolerance policies on wheel warring and other desysopping structures to weed out those who caused problems.

I think it’s fairly obvious why such a change is unlikely from within, due to the incentives of either current or aspiring admins. And yet without it, the factionalization and wikipolitics seems likely only to worsen. I wonder where this has been given thought.

All admins should be required to pass a course in ethics, and be held accountable by registering under their real names.

Moulton's on the right track. There has to be a fair degree of transparency and ethics.

Give the tools to people who have demonstrated a clear ability to deal with people diplomatically & fairly, and have x number of edits, and have done this, that and the other.

If it's automatic, I think it would introduce even more chaos into the system.

I submit that a project truly devoted to non-elitism would have strict term limitations on all management positions.

The fact that such a reform will never happen on the initiative of the Elite Cabal itself is of course obvious and telling.

Jon Awbrey

That's one approach. The other is to treat admins more like regular editors so that adminship is no longer such a big deal.

What would this mean? After 2000 edits and 6 months, an editor is allowed to delete pages and block other editors. If they show they can't handle it, the power gets taken away.


That could work as well. On the other hand, it might be more efficient under the automatic approach, to not have to cut good editors off.

The trouble is that is how the system is already supposed to be ("it's just some tools, it's the mop and bucket): adminship does not represent any acquisition of knowledge or insight, yet in practice, that is what they acquire in WakiWikiWorld. The janitors get to tell the teachers what to do.

Admins are explicitly stated not to have any extra weight in editing discussions, yet they have it, and also they get to set policy arbitrarily whilst the minions argue for months for obvious and sensible corrections and get abused as trolls by admins for their troubles.

One result of a semi-automated system would be to lessen the number of people who hold authority, at least over other long-term editors, down to the arbcom, checkusers, etc. who can be identified. That wouldn't solve everything, but I think it would help quite a bit.

You underestimate the influence of unofficial power structures. For example, Tony Sidaway is as powerful as most admins, despite a lack of any admin tools (I don't count the IRC channel). If you made more people admins, the cabalisti would just be even harder to spot. Even if your plan worked, giving power to fewer people helps how?

I guess I don't mind the unofficial power structures so much as arbitrary and gameable structures that are built into the system. As far as I know someone like Sidaway may be a problem if you disagree with him, but otherwise isn't a systemic issue... or at least one I could say much about. He even uses his real name!

The benefit of reducing the people with high-end authority is to make it less relevant and level the playing field, and just to make adminship less important. Frankly I don't see how any admin needs a block button that other experienced editors don't have. That kind of thing should go to ArbCom, where at least some kind of quality check can be maintained.

Of course once you have this structure it's hard to imagine getting rid of it. Most editors probably think Wikipedia is too big at this point to allow equal editing. The problem is that this compromises the entire wiki structure, while pretending to leave it in place. If you're going to have a wiki, I think you should have it.

But the structures built into the system are not all that gameable. The problem is, the system was designed with a Wikiutopia in mind, instead of falliable humans in mind. As a result, the formal structures are nearly indistinguishable from the formal ones. You talk about administrators having authority? Well, that's not part of the system that was set up at all. Administrators having special authority to decide things is completely an invention of the administrators themselves.

But it is a natural outgrowth of the method of election, or at least a natural weakness.

I could say the current system is almost entirely gameable, in that I'm not sure it serves any purpose other than the game. A person passes RfA, and what does this mean, that they are competent and noncontroversial? They're then given the power to delete pages, block editors, and lock down pages. I understand the original purpose, that they didn't want to give these to everybody, but beyond that I don't see the great benefit of the way it's done.

Yep; all your comments are well-said. There are many non-admin users who seem to wield more power than most of the 1300-whatever admins, but TS is a particularly good counterpoint because he's even open about his identity unlike, say, MONGO. The problem isn't the bit, nor is it even the way sysops are minted--the root evil is the culture.

I think emphasis placed on adminship is weird on Wikipedia, and it's also weird on this thread. It isn't much power, when you consider how any somewhat influential admin can reverse your actions and dodge implications of wheel warring. If we had 6000 admins on wikipedia, I guarantee we'd still be talking about the same folks.

The problem is that the community generally lacks accountability. The most influential admins just happen to leverage the institution's excessive faith in human nature.

That's like saying Godzilla doesn't have much power because he might be reversed by Megalon.

I didn't write this. Another moderator made an innocent technical boo-boo and deleted my actual post, inadvertantly subsituting the above. Apologies to those who may be waiting for me to defend or elaborate on the above quote, a result of mistaken identity. I will now attempt to reconstruct what I actually posted:



I disagree, for the same reason that I oppose term limits in real life politics: there is absolutely no guarantee that anything will actually be improved by them. In fact, you run the risk that some capable and fair-minded admins would be removed, and lord knows they are scarce enough at Wikipedia. Term limits are generally brought about by populist rage, which indiscriminately dislikes whomever is in charge. What I would prefer, would be a sysptem that holds admins to a much higher standard of conduct than the rank-and-file editor, so that it would actually be easier to de-sysop an admin than to ban an established editor.

Sure, but can Megalon dodge implications of wheel-warring? From what I remember, he could barely dodge a Jet Jaguar atomic karate-chop... OTOH, implications of wheel-warring may be easier to dodge than one of those...

It does seem like an overgeneralization to say that WP admins are either too powerful or have very little power - it probably depends on where, and in what context, the attempt to exert power is made. And also who happens to be watching, how many friends he/she has, and what sort of mood they're in at the time...

What's more, even if admins (or editors in general) were somehow compelled to become more accountable by exposing their real identities, that might only create a special class of WP'ers who would be able to get away with more abuses, simply by virtue of their having nothing to lose in real life if their WP activities were to become known to their friends, families, teachers, customers, business associates, and the fellow who picks up the trash on Thursday mornings. And let's not forget that there are plenty of people merrily editing away over there who would just love some extra attention and notoriety, under their real name or otherwise.

Wheel-Warring in WikiDrama, like political give and take everywhere, follows an oft-observed model. The model presented here applies in general to all WikiDrama at any level of intensity, from a simple reversion to clamorous kerfuffle and brouhaha. It has 5 stages.

1. Mimetic Desire for One's Point of View
One editorial clique establishes their Point of View as an editorial objective and other editors react with a countervailing drive for their complementary Point of View.

2. Mimetic Rivalry for More Prominence
Now the editorial cliques begin competing for prominence. Whatever winning strategies emerge, the less experienced editors copy them. To survive in Wikipedia, an editor must become deft at gaming the labyrinthine rules of the system.

3. Skandalon
Skandalon is a Greek word that means "taking the bait." It's the root of "slander" and "scandal." In the rivalry for editorial dominance, if one side can goad the other into committing a foul, the opposing editor can be neutralized or even eliminated from the game. Thus begins a Wiki-War, fought on the editorial battlefield, in which the goal is to demolish and disempower the other side. Skandalon is what makes it so hard not to take the bait, so hard just to walk away. It's so easy to bicker and goad. The give and take escalates.

4. Scapegoating and Alienation
Eventually one editor crosses some arbitrary threshold of civility where another Admin feels compelled to intervene. It's essentially random which side crosses first, but often it's the more disgruntled minority, which uses harsher language to maintain parity. Whichever side goes over the arbitrary line becomes singled out, and the others who kept their trolling below threshold are sorely offended. They rudely chastise the miscreant, sending him or her to the Oblivion of Time Out.

5. Consensual, Irrevocable, and Sanctioned Banishment
To appease the rabble, the ArbCom determines the standards of civility and visits banishment and page-blanking on the outcast. Then everyone issues a sigh of relief. This escalates the polarization to the next higher level of examination in online culture.

The 5-stage pattern repeats at all levels of Wikidrama and for all rivalries and editorial competitions. The most vicious attacks are reserved for people highest up in the power structure. Jimbo Wales, ArbCom, and Wikipedia Review all follow this model. Well, actually, almost everyone follows it.

At every point in a battle of WikiWits, the dynamic is somewhere in the 5-stage model, which repeats endlessly.

The only way to arrest the Wikidrama is to adopt the conscious goal of de-escalation and run the model backwards toward constructive dialogue. Giving up the desire to be dominant, avoiding the temptation of skandalon, avoiding Requests for Comments, avoiding authorized and sanctioned banishment.

A common type of outcast is a person who bears witness and speaks the truth to power.

Wikidrama, left to itself tends to escalate over time.

We need to think our way out of verbal vendettas by mindfully running the model backward, de-escalating editorial power struggles and moving toward open dialogue.

At every stage of the model, we need to be mindful of the dynamic we are caught up in, and consciously elect to run the model in reverse.

With this Systems Theoretic Model of the dynamic structure of argument, debate and dialogue, we can discover the optimal strategy to drive the system in reverse toward better practices and more accurate articles.

It's pure science, pure reason, and pure common sense. These methods of thought all reach the same insightful solution to getting along.

It's time we learned it so that we can discontinue the mindless practice of Wiki-flogging ourselves to death. It's time we learned, reviewed, reflected, and meditated on the Mimetic Reconciliation Model.

I didn't say that I was CooCoo4DaConcep, I simply indicated Yet Another Glaring Inconsistency (YAGI) in the Wikipediot Litany Of Lunacy (WP:LOL).

Whenever you make the argument in Wikiputia that Admins shouldn't have so much power in a genuinely non-elitist system, they and all their knee-jerk Minions jerk their knees in goose-stepping unison with the reactionary reflex protest that that People Who Know Da Ropes should naturally have whatever power they can rope in.

What is that?

That is an argument from expertise, the very kind of reasoning that they Howl In Protest Against when it's expertise in anything but Gaming Da Game.

Jon Awbrey

Wikipedians have a neurosis about subject matter experts that seems incurable.

Doubleplus, they seem to have some difficulty deciding who qualifies as a subject matter expert.

This is one of the reasons why they have wasted years and years worth of Ediot-Hours on policy pages like WP:NOR, trying to Robotize what are by and large no-brainer decisions to SME's.

Jon Awbrey

I am talking about admin rights, but my argument is about the way it corrupts Wikipedia's culture. This happens initially through a no-rules election process where succesful candidates need a super-majority-plus of the vote. Some wikipedians put a lot of effort into constraining the reasoning on these votes, but people certainly aren't compelled to justify their votes as they would be in a tenure review or anything comparable I could imagine in an academic setting. Thus, the clear message is that becoming an admin is about being someone the vast majority of people are at least vaguely comfortable with. Certainly this fills the role of a crude barrier, but seems awfully inefficient, with all the problems of having an anonymous and virtually-no-entry-barrier vote on who gets a position. I don't think it's consistent with overall neutrality in the project.

The problem also continues once these elections take place, to cause many of the issues that people here are talking about in terms of Wikipedia culture. Once a person has won a standardless election, it shouldn't be surprising that some people give them an equally standardless halo of authority, or equally importantly that they start to act as if they have one themselves.

I would support strict rules for the removal of admin powers, in fact I think this would much better track the underlying purpose. If someone blocked another editor abusively, or locked a page they were editing, you could take it away and make them fill whatever the requirement was again. That would make a lot more sense to me.

The only requirement I see is popularity.

Popularity does not generally equate to competence, diligence, and ethics.

The competent, diligent, and ethical player is often remarkably unpopular.

The first question I would ask is this. Why the hell does enwiki need over a thousand admins? Why do they need to ban so many users and delete so many pages? That is just wasted energy.

It's a consequence of having adopted a dysfunctional regulatory architecture from the gitgo. Since the system is not self-regulating, it needs an ever larger police force. But the police force is itself also not self-regulating, so the whole system is corrupt and dysfunctional.

In other words, a fish rots from the head down.

I don't think Wikipedia needs thousands of administrators, although one effect of the current system is probably to make adminship into more than it needs to be. However, Wikipedia has a policy question of who exactly should have the block, delete and protect features. Should they go to everyone, or some people, be assigned based on an open vote, or assigned in another way?

The system of an open vote seems idealistic, and maybe alright as a start, but far from a great way to run a larger system. If instead they assigned it much like the AutoWikiBrowser, I imagine they could avoid the whole dramafest of the current system without much else lost.

AWB is assigned by a very small group of AWB developers. If adminship were granted based on the judgement of a half-dozen users, there would be a hue and cry.

People do tend to focus too much on the admin tools, when some of the monobook tools are just as useful (and at least one has the same effect as an admin tool).

I find it obscenely unethical of WP to consider a trivial
thing like wheel-warring worse than defamation,
privacy violations, and other matters seriously affecting
real people.

If an admin wheel-wars to unblock someone, should the
person be blocked just to punish the admin who wheel
warred? That is incredibly dehumanising to the victim
of the block. Is making admins feel powerful, like their
decisions cannot be reversed and must be respected,
more important than not making a good person feel
unwelcome?

Worse yet, supposing an admin wheel-wars to delete
defamatory, privacy-violating, or otherwise severely
harmful content. Should the admin be punished for
doing the right thing because of some strange notion
that having one's power challenged is worse than
being defamed or having one's privacy violated.

It's utterly unethical!

They've replaced morality with systematic power-tripping!

The truth of the matter is that the BLP paranoia that's currently rampant on the project means that the automatic adminship train has officially left the station. There's no way the ultra-paranoid folks would ever consider allowing such information to just anyone.

Well, they could break up the admin permissions.
Protect/unprotect are probably fairly safe to give out,
as is the ability to view Special:Unwatchedpages, or
whatever it is. The delete button might be less
dangerous than the undelete one - better yet, they
should oversight pages with defamation and privacy
violations so they don't have to worry about who sees
the deletion log, but that doesn't stand much chance
of happening. The block button can do a lot of
damage, but as long as you don't buy the having-one's-
power-challenged-is-worse-than-anything BS, the
unblock button is fairly safe. Ipblock-exempt should
really be granted separately.

Still, as long as the system favours power-tripping over
morality, no change you make to how people gain
power will do much good.

I think the suggestion that admins should use there real name or disclose it on their userpage is a good one. The concerns about stalking are honestly overblown - I can only think of one instance I've heard of (and not verified) where someone was subjected to what legally constitutes stalking. Out of a thousand + admins, thousands of editors, and millions of readers, that's not a big fucking deal. And the occasional nutcase that emails you is dealt with quite handily - it's called the trash folder.

I registered with part of my real name, and ended up getting crazy emails from BoxingWear/George Reeves Person, who is a certified crackpot. Some months later I gave an interview using my full name, and crackpot started sending emails to the Dean of Faculty office at my college. What happened? The emails were so incoherent and insane that the DOF asked me about them, and when I explained the situation they said something along the lines of "What an insane person! Is there anything we can help you do about this?" End of story, no harm done. The potential harm in revealing one's real name is not as great as the paranoid would have you believe.

That's probably a factor, but also nonsensical since the current system lets anonymous people through anyway.

ArbCom could do it, or bureaucrats. The point would be that it wasn't an election, but a question of whether you met some objective measure. People could object as well, but the presumption would be similar to AWB: if you're not abusive, then there is no reason to withold these features.

If there were a thousand Godzillas, and hundreds of Megalons, perhaps.
This must be correct. Wikipedia's hierarchy behaves as if non-editors have no stake in what's said about them, whereas crossing certain influential admins is a grievous sin. The fact that wheel-warring taboos can transcend policy, law, and ethics is much more poisonous to the culture than silly RfA straw polls.

There are plenty of admins who have disclosed their
real names who still engage in defamation and
privacy violations.

Granted, if you are going to be defamed or have your
privacy violated, there is some consolation in knowing
that at least you know who did it. But it would be
better if the defamation or privacy violation had not
occurred at all.

But, when complaints of defamation are labelled as
'harassment', there's really no hope you whether you
insist on real names or not.

And which is worse - using an obvious pseudonym, or
stealing someone else's name and letting everyone
think it is yours? The latter could be used to ruin a
person's reputation.

The focus needs to be on the person's morality, not
on what they are called.



Only one? On the internet, you mean? If you have
only heard of one instance of stalking in real life, you
would seem to have lead a sheltered existence. Not
that there's anything wrong with that, good for you if
you have.

On the internet... well, does threatening to turn some-
one black and blue count as stalking? I heard that
happened to someone. (Not an admin.)




{{{One}}}

So many over there have no ethics,
no basic sense of human decency.

Hence my suggestion of requiring all participants with any level of administrative power to pass a college level course in ethics. You cannot obtain a college degree in journalism without taking such a course. The same training should apply to Wikipedia.

Not everyone has money for that sort of thing, so it
would be classist. (Not everyone has legal ID, either.
1/3 of the world's births are not assigned birth
certificates.)

More importantly, ethics can't be taught. Ethics flows
not from the mind, not from the soul, but from the heart.

Consider Oscar Wilde's The Fisherman and his Soul.

Anyone who cannot learn ethics has no business exercising power over others.

Ethics is not learnt, but realised and felt.

That will come as a shock to my colleagues in academia who teach ethics.

Classroom ethics teach ideas about ethics but but does little or nothing to instill ethical behavior. Part of the problem WP faces is the extreme youth and limited life experiences of their cadre of admins. If a developmental model of ethical/moral behavior is to be given serious consideration it might need to be accepted that WP admins simply do not possess the raw material needed to achieve the kind of ethical behavior that would be acceptable in the position.

Ethical reasoning is learned, just as all methods of reasoning are learned. It's troubling to contemplate the notion that those handed supervisory power on Wikipedia are not intellectually mature enough to acquire the requisite knowledge and skills essential to the ethical exercise of their duties.

No amount of lecturing can drive home the point
that starting violence is bad like getting beaten
and forcibly drowned for a few minutes.

Fighting back? Well, starting a physical fight and
defending oneself are two very different things.
But then, obviously some can't tell the difference.

But if all the experience one has with violence is
watching some films filled with glorified battles,
simulating some fights on those little computer
games for pure pleasure, and hearing a few
lectures on the topic, then maybe one would see
threatening people with violence for the good of
the encyclopaedia as a helpful thing to do. Who
knows what goes through people's heads that
leads them to such conclusions.

One does not learn to reason by listening to lectures.

Studying ethics is like studying any subject that requires the development of higher-order reasoning skills.

Students in ethics don't learn what to think. They learn how to think.

It may be learned but that learning may have to occur within parameters of set sequences and during critical periods.

The longer such learning is delayed, and the longer a person exercises power without acquiring the ability to reason ethically, the harder it becomes to repair the damage, both to the individual in question, to those impacted by unethical decisions, and to the reputation of the system in which the parties are jointly embedded.

Absolutely agree.

Nor do one's professors physically attack one so that
one might learn that attacking is wrong. That wouldn't
be right.



So is it then not from the ethical, but from the evil, that
people learn ethics?



If someone beats a person up, that person must either
conclude that he or she is evil and deserved the pain,
or else that it was wrong of the other to beat the person
up.

No higher-order reasoning required.

But then, why do so many innocent people think the
former?



No. Reasoning can help you find the most effective
way to destroy your enemy, but you must feel with your
heart to know if it is right to do so.

It also takes higher order reasoning to discover how to teach ethics to those who are ethically challenged.

If they have no heart, they will never realise ethics.

But how is it possible for someone to have no heart?

I may as well ask how it is possible for there to be so
much cruelty in the world.

Maybe people lock their hearts away in iron boxes,
so they can't hear the thumping.

Perhaps some of them suffer from Oxytocin Deficit Dishonor.

I'm only aware of one case of actual stalking resulting from someone's work on Wikipedia. That's not to say that no other incidents have occurred, but those who see the stalking bogeyman everywhere are generally short of specifics.

ETA: And no, threatening someone is not the same as stalking - it's threatening. Quite a lot of threatening happens on Wikipedia, from all levels of users. Whether that's something that would be exacerbated, prevented, or not affected by disclosing real-life identity is a question I can't answer. But the proponents of anonymity do not generally base their arguments on incidents of threatening. They are almost singularly concerned with the possibility of stalking.

Okay, thanks for clearing that distinction up.

Well, stalking or no, threats of violence are morally
depraved.



Hard to make a believable threat of violence against
a person if you don't know where to find said person.