|(English) This is an essay. It expresses the opinions and ideas of some Wikimedians but may not have wide support. This is not policy on Meta, but it may be a policy or guideline on other Wikimedia projects. Feel free to update this page as needed, or use the discussion page to propose major changes.|
I served as a Wikimedia steward from February 2014 to March 2015. While I believed that I would pass confirmation, I chose not to return for a second term for personal reasons. I figured that I would post some thoughts about my time as a steward, since it's hard to know what the role really entails without actually having gone through the process. This was largely inspired by w:en:User:AGK/ACE2012. Note that these are all my opinions, that may or may not be shared by my former colleagues.
What can you add to the team?
The ideal steward has experience being a bureaucrat, a checkuser, and an oversighter; they may have even served on an Arbitration Committee. They have edits across several wikis, and serve as a global rollbacker, a global renamer, and a global sysop. They are administrators on at least a few wikis. They are available for several hours a day. They have experience on big wikis and small wikis. They know several languages and are effective communicators in all of them. They have OTRS experience, are highly trusted in their communities, and have never made any serious mistakes. Besides their service in these areas, they also write content on several wikis.
Does this "ideal steward" exist?
No. Even if one devoted every waking hour to Wikimedia, it simply is not possible.
Perhaps a better approach is to look for what a candidate will add to the team. Do they have skills in at least a few of these areas? Or once they are elected, will they not do anything because they have no experience in any of the areas that stewards work in? Very few stewards work in all areas, and some only focus on a few areas; stewards do a wide variety of tasks, from renaming to counterspam/countervandalism to CU requests to granting and removing permissions. And as far as languages, while there are areas where those skills help, there is plenty of work to do that does not involve languages. Google Translate usually suffices for most steward work (since it is translating from another language to your own), except for the more rare of the 200+ languages Wikimedia has (which usually no steward knows - but in those cases they often speak English or another common language).
As far as languages, I'm mostly monolingual (I can get the gist of something in Spanish) but Google Translate was sufficient for me and I rarely found myself disadvantaged. I would say the most problems I've ever had was dealing with the Croatian Wikipedia issues (but that is not a major language) and the Azerbaijani issues (but that was well after my steward term).
Stewards are not an ArbCom
Stewards are not a global arbitration committee. They can only act in clearly-defined cases according to policy. They cannot globally ban users they don't like, nor cannot desysop people whenever they want. They can only act according to the wishes of local communities. If you want this sort of power, then being a steward is not for you.
There also is no global arbitration committee. Which sometimes leads to some awkward interactions with the rest of the community, looking for some sort of leadership. Sometimes stewards have had to take actions at the edge of their remit, asking for a consensus on Meta to take difficult actions (i.e. Chechen Wikipedia). Other times, they have to bring matters to WMF as they concern matters with legal repercussions. I will also say that the superprotect matter was also quite an awkward time for us, with us having the technical power to undo the controversial actions, and being tasked with implementing community consensus, yet risking consequences from WMF if we did so. In the end we were not used as "battering rams" as I personally had feared that we might.
Thankfully, WMF has taken on the worst of the worst of the harassment cases with the introduction of global bans. However, stewards are sometimes consulted on these matters, which involves confidential and sometimes, downright disturbing information. It requires you to be of sound moral character, to say the least. I am also told that communication with WMF has increased on other matters since I stepped down.
Can you devote enough time to the steward role to make an impact? The number one reason that stewards are not reconfirmed is due to low activity. Have you been on Wikimedia so long that you are burning out? Are you looking at any upcoming life changes that will affect your availability? Can you remain active in your other on-wiki responsibilities? If halfway through your term you find that you are unable to do so, are you willing to step back in those other areas or resign from them entirely?
Do you refuse to use IRC or private mailing lists? If so, that will be a significant handicap for you as a steward. For as much as people like to hate on both of those, because of privacy concerns, the difficulty of having many user talk pages and noticeboards across many wikis, as well as timeliness, that is where most crosswiki work takes place.
As a steward, you will be subscribed to four mailing lists: global-sysops, global-renamers, checkuser-l, and stewards-l. Global-sysops is rarely used. Global-renamers got about 30-40 emails a month, from my short time on the list. However, checkuser-l and stewards-l can approach 150 emails a month, each, though it can be highly variable (and one can drop all the lists except stewards-l, though it is not recommended). Are you prepared to read through and stay current on critical issues?
Can you work with other people? Can you listen to what your fellow stewards have to say? Can you work to come to a consensus with them, or are you unwilling to compromise? Remember, what benefits Wikimedia the most is having people of many different opinions, so that we have a true balanced outcome; it's not all about "winning" battles. I would have to go even farther than this and say that the single largest cause of dysfunction during my term were some stewards who insisted on a particular course of action and were completely unwilling to compromise.
Can you respect another wiki's policies? On occasion, we do have wikis that do inappropriate things (more on that later), but the vast majority of these local policies should be respected, regardless of whether you agree with them. Can you step back and let local users handle matters when appropriate?
On occasion stewards encounter some really nasty stuff - not nearly as bad as what the English Wikipedia ArbCom receives, but sometimes harassment from enwiki goes crosswiki. Stewards do often see the "nasty" side of the site through oversighted content and checkuser-l, which can be disconcerting and wreck your fuzzy notion of Wikimedia if you are not prepared for it.
Sometimes people that arrive in the IRC channel are upset because they have been the victim of outing. Or there are emergency situations where decisions have to be made. Can you remain calm enough to act in those situations? Can you calmly tell them that sadly, there often is nothing that stewards can do about the situation, per policy? Or will you tell them to get lost because you can't or won't do anything, and provoke them to leave uncivil messages on your talk page, other stewards' talk pages, and every other noticeboard on Meta?
You will make mistakes
There are so many policies and norms to follow, and sometimes you will get it wrong. Can you learn from your mistakes? Can you admit in public that you were wrong, and reverse your actions? Or will you force another steward to make the hard call after a long and drawn out discussion where you maintain your innocence in the face of consensus?
Some of our policies and tools are complicated, and have legal consequences (especially the CU and OS tools) - if you do not have experience with the tools already, do you think that you will be able to learn how to use them? Or if you do not want to focus on those areas, can you learn enough about the tools to respond in emergencies?
Or, sometimes you will say or do something that offends another steward or "established" community member such as a functionary. Are you willing to apologize and at least try to restore the working relationship, or will you be at odds with them until the next election?
Too many problems, not enough time
There are many problems that Wikimedia wikis face. If you try to fix all of them, you will drive yourself crazy within the first month. These could be problems from spam not being deleted, to POV content all over the wiki, to autotranslated horrible articles, to even administrators acting badly.
Focus on what is important, and what can be fixed, rather than trying to fix everything all at once, which will result in nothing getting fixed.
English and German Wikipedias
Stewards are allowed to perform emergency admin actions, desysops, CU and OS actions on all wikis, in the event that local users are unable to respond. In practice, that almost never includes the English or German Wikipedias, the two largest Wikimedia wikis. Stewards who have performed inappropriate (or sometimes even appropriate) actions there or even on a few other wikis have received a lot of backlash, and in some cases have lost confirmations. (Since this has been used to say that stewards absolutely cannot do anything on those wikis, that is patently false; they almost never can, but they are able to if local users are unable to respond).
Also, if you are from the English Wikipedia, prepare to face a bit of extra scrutiny when you run. For reasons detailed at w:en:User:Rschen7754/You represent the English Wikipedia!, global users are suspicious of people from en.wikipedia. Note that a user with solely en.wikipedia adminship/crat/CU/OS rights, weak foreign language skills, and little crosswiki experience is almost certainly going to fail (in fact, this has happened in recent elections).
You cannot perform a lot of steward actions on wikis where you are an active community member. What is "a lot" has varying interpretations, and you may need to adjust your approach in some cases. Typically if you are an admin or have ever run for admin there, people start looking suspicious. Some are able to say, "well, I'm not actually active in the community/left years ago" and people are okay with it. (Meta has some leeway because all stewards are part of Meta, and test/test2/testwikidata/mediawiki.org/outreach are sometimes given leeway since they are not full communities or adminship is very easy to obtain). After inactivity, this is another common reason as to why stewards are removed.
However, as a steward you frequently do mediate between other users on your homewikis, and assist them in crosswiki matters. But sometimes you have to step back and make sure that you are not inappropriately influencing discussions related to your homewiki, or at least be transparent about your bias.
As far as private information, stewards can see all oversighted and deleted content on all public wikis, including their homewiki, and have access to the CU wiki and checkuser-l. They can see CU logs, but cannot actually use the CU tool without granting themselves local CheckUser (or CheckUser on loginwiki), and this is almost never done on wikis with local elected or appointed CUs. They should not be using another wiki to run CUs on users from their homewiki, unless the issue is crosswiki and affects more wikis than just their own (and all CU actions are logged) - though the last sentence is more of a standard that I imposed on myself, to avoid additional controversy.
Outing and harassment
Generally, the steward role by itself does not result in off-wiki harassment. Sometimes if you lock a certain troll (especially those from en.wikipedia) you can get some onwiki harassment, or some extra emails. Most of the users you deal with will live in other countries, since you cannot perform most actions on your own homewiki; they won't be paying money to harass you in real life.
With that being said, if you already face off-wiki harassment, being a steward will not make it go away, and may make it worse.
Time to go!
Usually at least half of the stewards you are elected with will have resigned around the 3-4 year mark. At least half of those remaining will have a significant reduction in activity due to burnout or other Wikimedia or life responsibilities.
You won't be a steward forever. The longest anyone has been is just over 11 years. Can you decide for yourself when it is time to go? Are you likely to stay editing for a long time after your election? Are you willing to drop other rights as your activity declines? Or, will you hold onto your steward hat for longer than you should, even though you are barely getting reconfirmed, until it is forcefully taken away from you by a failed reconfirmation?
Warning: you could be desysopped or banned for running
Added in early 2020
This is mainly aimed at potential candidates from the English Wikipedia but I suppose it could apply to any candidate.
Over the last several steward elections, 2 of them have had at least one candidate (who ran a serious candidacy) who has both:
- Not only lost the election, but gotten one of the lowest support percentages among all the candidates
- Gone on to be desysopped and/or banned due to an Arbitration case or motion against them within the next 12 months after that election
Clearly their self-perception is a bit off, to say the least. In fact, it could be argued that their candidacy drew the attention of editors who realized that this candidate not only should not be elected, but should have less influence in the community.
Is this you? What do other editors think about your running? Are you asking just your "wikifriends" who can't bear to call you out on your bad behavior?
About the elections
- Voters: the primary question that you should be asking yourself, when you vote, is "can they do the job"? People who vote based on "I like/don't like them" or "I'm blocked on their homewiki, so I will oppose" do the community a huge disservice, and make the elections political. We are here to promote free knowledge, not to be a MMORPG.
- Voting: about half the votes come in within the first 48 hours. Occasionally there will be one or two borderline candidates hovering just above or just below 80% (in 2014, the last 2-3 votes changed the outcome), but usually you will know whether you will pass or fail by the end of the first few days.
- Onboarding: If it looks like your candidacy will succeed, towards the end of the election, you will be contacted by a steward, and your email address and IRC information will be collected. Once the elections close, there will be a 1-3 hour wait while the stewards certify the results. Once the results are certified, an announcement will be posted on m:SN, and you will be given access to the IRC channels, mailing lists, wikis, and the global and local groups almost instantaneously and be ready to receive requests.
Confirmations for existing stewards are a different process. The purpose is to ensure that existing stewards retain the trust of the community.
Officially, there is no set percentage that is considered "passing", but towards the 50% range is usually considered failing. A group of stewards reviews the responses and determines what the consensus is. "Votes" with no rationale, that are nonsensical, have incorrect reasoning or are because of some personal issue are discounted. I personally believe that the bar for reconfirming a steward is too low, but I won't expand on that in this essay.
But here is my opinion on how to make the most of the process.
If we opposed every steward who made a significant mistake, we would be removing every steward, including myself. This is probably not a good idea.
What is better is to evaluate the points brought up. Obviously, very serious mistakes such as those violating privacy should be treated without mercy. But as far as the rest, are they isolated mistakes, or do they reflect a pattern? Did the steward apologize and show they understood their mistake? Or are similar significant mistakes likely to happen in the future?
Complaints about teamwork and collaboration should not be ignored, especially from fellow stewards (who would know the most about how stewards behave on list and in private channels). A steward who cannot work with the rest of the team can significantly limit the effectiveness of how the entire team functions.
As far as activity: sometimes real life happens, and stewards go offline for a few months at a time. If they're active at most other times, and have been more active in past years, that's probably okay. If they are known for little activity in previous years, were really inactive (barely above the 10 actions per year part of the steward policy), or this is their first year (when stewards are supposed to be the most energized), then they will most likely not be active next year, and should be removed. It's always unfortunate when this is the case, because sometimes nice people get removed, but considering that the steward role is particularly sensitive and handles private information (including a full subscription to checkuser-l and access to everything oversighted), the fewer people that have this access, the better. Sometimes stewards help in other areas (such as usurpations, declining requests, off-wiki work, CU actions on loginwiki, providing input on the mailing lists, administering the lists, maintaining the IRC bots, editing closed wikis, and global sysop actions) that aren't reflected in Meta logs; I usually take it on a case-by-case basis (it might help in borderline cases, but the purpose of the steward role is to actually use the steward tools, not just those that are available in other roles such as GR or GS).