It is Time to be Incredibly Blunt Because You're Either Tuning Out the Community or Blind to the Truth of the Matter
(Please don't take anything I say as an attack. It isn't intended to be. I'm only observing what I see going on here.)
There is external context to why people are upset with the new policies. A person was removed for what people believe was a cross between either rash stupidity.trigger finger on the part of a moderator or on the more extreme case outright malice. Of course I'm referring to Monica. They got de-modded for something about a pronoun discussion gone south. Certainly not what I'd expect to be the source of a massive community upset, but I digress. I'm going to describe the issue in a slightly different issue as a timeline of events. This is the perspective of the community as I've observed mind you. If one day it were revealed that Monica were pulling wool over all our eyes, obviously this timeline would be wrong. However, perception is worth far more than the truth when you are dealing with politics and keeping a community happy (spoiler: they really aren't happy).
- Monica expressed dislike towards some new policy regarding pronouns.
- A fellow moderator or community manager or someone decided they felt this was Monica expressing prejudice feelings towards LGBT and other groups desiring the new policy.
- They requested she be removed under the grounds that people like her are simply not welcome.
- The request got expedited and she was demodded without any chance for appeal or community input.
- The people at stack exchange in an attempt to cover up the situation refused to provide her any information.
- The people at stack exchange either in error or in an attempt to cover up the situation spoke to the media and gave false statements that were degrading to her image.
- stack exchange creates new policies for removing moderators that may or may not help in future circumstances.
This is more or less what I pieced together. I'm going to now tell a fictional timeline of the B & O Railroad Company. I got that from Monopoly in case that spot on the board is named after an actual railroad. This is a fictional story.
- Foreman on the railroad wants to be able to get more tracks built in an expedited timeline.
- Foreman decides to contact a government official to try and get permits approved faster.
- Government official states that it can't go any faster due to various red tape and policies.
- Foreman bribes government official to bypass any need for inspections.
- Company is eventually charged with corruption.
- Company attempts to write inflammatory articles to try and discredit the government official.
- Company is legally forced to add more anti-corruption practices and to have a monthly inspection to prevent and discourage future bribery.
Now, while obviously one situation involves way worse problems and could potentially get people killed, if we only consider the question of which situation I would be more likely to believe could have a repeat offense, I would sadly have to say stack exchange. This is of course assuming the community isn't horribly misinformed (if we are, then provide the proof that corrects that belief).
Why is this?
Because bribery and other forms of unethical business practices can result in criminal charges and major fines, and a government will actively prevent that business from doing any further repeat offenses.
In other words, I don't necessarily trust that foreman to not want to bribe someone. I trust that he was arrested and thrown in jail and that anyone thinking of doing the same will be prevented by the routine inspections.
Basically, I would find a previously corrupt and weeded-out railroad more trustworthy and safer to interact with than the stack exchange following this incident. I am very glad that I am not a moderator, nor would I ever desire to be following this.
How does this relate to stack exchange?
You have no oversight. You did something comparably unethical (though obviously not even remotely as egregious) or made a mistake that at the very least harmed trust. Nothing prevents you from repeating the same incident. There are no consequences. We can't stop you. Clearly if this was a massive mistake, you can't seem to stop yourselves either. Policies are nothing more than things you can show off to the new guy being hired and say "hey, this is what you do in your job" if you have no means by which to punish people for not following the polices. Us being upset is clearly not an example as many comments in this question indicate that every angry person leaving the site would have little to no impact in the grand scheme of things.
Without consequences that force you as a company to be accountable for such actions n a truly meaningful way, there is no way that any policy will ever work. It is a pacifier, and we are not infants. We do not need to be treated as such.
With all of that out of the way, what can you do to improve?
First off I propose 3 "stages" of how to remove moderators:
stage 1. Full Transparency:
All evidence and all claims made against the moderator should be released publicly. The discussion itself should also be public. If an issue of revealing personal information comes up, and the person does not wish to divulge that information, then the information can be censored. Ideally the reporting member or staff member should fill out a formal complaint. If that complaint is deemed worthy enough to be heard, then a formal report should be posted to a meta post (locked from being editable) and possibly locked from being commented on and a basic summary should be provided of the post. Upvotes and downvotes could also be turned off. It can be completely locked to external interference. Just make it public so people can't randomly make up claims. kthx.
This step should never be skipped, ever, period. Even in the case of an emergency, a full and comprehensive report should be posted explained exactly what happened. If it is a legal issue and must wait until after the user is demodded, do so. But be specific. If a hack was performed or a bot went rogue or a user was a pervert stalking children, give details of what hack was performed or what bot broke down or mild details on the nature of the stalking. I respect that a child being stalked obviously makes disclosure more difficult. Granted, in the case of a child predator, I expect the moderator in question to be going to jail, in which case their mod status being revoked via a bypass of the policy is whatever. To be honest, I'd expect that to be a situation of permanent network ban without appeal. No matter what, give as many details as legally possible (including direct links to specific instances of the behavior and/or chat logs) to avoid confusion either from the community or the moderator. This will make a lot more sense when I get to stage 3.
stage 2. Moderator Defense:
Following the creation of such a meta post, there should be a period of at least a week whereby the moderator is expected to answer questions and bring any other people into the conversation they wish to bring in. Furthermore, it should be expected that the moderator provide an answer post to the meta post stating precisely why they believe they should not be demodded (or that they concede and wish to step down). This should be done properly and taking the person's schedule into account. If the person has a vacation and they won't be online, then that's a reasonable thing to take into account. Assume reasonable good faith. If it will be an extended period before they can be online to discuss it, then a temporary demod would be done just to prevent lying mods from getting out of having to undergo a formal hearing. In other words, don't allow blatant abuse of the system, but try to be cordial.
stage 3. Community Vote:
No matter what the case a community vote should be taken to allow the moderator to effectively be "impeached". The vote requirement could be based on the number of people voting. I don't know if a simple > 50% majority would work. I don't know what the turnout for such a vote would be. Even if the moderator has to be removed for legal reasons, a vote should still be held.
The only time ever period whatsoever that stages 2 and 3 should be violated is if every moderator that the sub site unanimously agree to demod the person, AND the user will be given a permanent stack exchange network ban following the demod.
In other words it should not be emergency behavior that triggers a bypass of stage 2 and 3. It should be criminal behavior. If the behavior could actually result in the person being sued by stack exchange or going to prison for cyber crimes, then obviously the community has no say. Arguably not even stack exchange has a say. You can't prevent the government from effectively demodding someone by proxy through the method of sending them to prison.
What are the consequences?
There have to be some. Otherwise abuse can and will inevitably occur. It is the nature of human beings. A system has to be designed such that any one malicious or foolish person cannot cause damage in a large manner. It is why the United states has a Bill of Rights containing seemingly innocuous and obvious things we would never question and why monarchies have tended to fall out of style in the modern world.
The community managers are employees, right? I propose that the first offense of not following the policy be met with a formal write-up to go on their personnel record. After a second time do whatever you do whenever someone violates a policy twice such as the policy of not mouthing off to your boss. Treat that policy as an actual important thing and not just a general guideline for the purposes of organizing chaos. Failure to follow that policy will in every occasion result in the community being upset and angry.
I close with the reminder that every time you violate the policy and I notice, I will remember that story of the B & O Railroad Foreman and how it is performing better at avoiding repeat offenses.