User:Slowking4/Leadership in Wikimedia

From Meta, a Wikimedia project coordination wiki
(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.

"decline will continue based on organizational culture, until an investment is made to shift the culture."



  1. "Quality (and not to be under-estimated, the perception of quality or of its lack) drives Wikimedia's reputation, which in turn drives readership and participation."
  2. "We want to encourage the creation and improvement of high-quality content. Our strategy on developing high-quality content centers largely around improving participation."
  3. "There are some signals that we are not as effective as we could be in helping new editors become active contributors to replace those who leave the projects. There are also signals that more experienced editors and administrators are becoming discouraged due to an increase in hostile behavior."
  4. "Today, the Global South is under-represented in the Wikimedia movement." i.e. systemic bias
  1. "the solution for Wikimedia is to slowly and steadily replace the current trend with a new trend that increases the number of users again and in the red point"
  2. there is a fundamental culture problem: for the CS major with a hammer, all problems are a nail. the trend is currently a death spiral; see also: aol, prodigy, yahoo, Encyclopedia Britannica.
  3. there is a thrashing suboptimal use of resources, lacking leadership. We see great efforts at bots and filters, and little effort at quality measurement, quality circles. with the notable exception, in efforts at recruiting and mentoring new editors, at the Teahouse.

Prevailing attitudes[edit]

  • patroling recent changes will fix the problem
  • send a message to vandals that their work will be deleted
  • tag, order, and bait the editors; and enforce rules, nuance is too difficult
  • the less editing, the less patrolling necessary.
  • delete it, it's not my job to fix it.
  • prove to me it has a source or delete it
  • mass deletion is within admin discretion, if they can't keep up with my work rate, too bad
  • oh my finger is in the dike, why won't somebody help me?
  • all the notable articles have been written, let's delete the non-notable ones
  • raise the drawbridge, then patrolling will be easier, quality improve
  • we need to create drama to force editors to work on my agenda


Leadership is required to improve the quality of wikipedia. The Principles of quality control apply to software projects such as wikipedia. Quality Circles need to be organized to address problems in a proactive way, without drama. Statistical Quality Control measures are needed to be developed and used, rather than fixing problems.

We need a new standard of professionalism. Like the librarians and archivists, we need to treat the public with respect, especially when they are wrong. When we treat people with disrespect, then we hurt the project.

"We owe politeness to the people who try to work with us in good faith, whether or not what they do is finally accepted. We owe more than politeness: we owe an effort to understand. Our typical way of dealing with unsatisfactory articles is very abrupt, with notices worded as if they were written by the lower levels of a particularly burocrat-ridden country. We've all encountered institutions which try to avoid being questioned--in the RW, they tend to do it with a bland imperviousness, but we use a crude bluster; it's certainly more forthright, but it probably gets most people even angrier, or at least more willing to express the ir anger. The first step is to delete every template notice we use for deletion,and require people to write personal directed explanations that show they have made an effort to understand. True, some of our people cannot adequately do this--they should leave communicating with new editors to the ones who can, or can learn. We may be a collection of anti-social nerds, and satisfied with our own primitive ways of dealing with each other, but the rest of the world isn't, and it's the rest of the world we have to deal with. (I make no claim to being much better than the average here; I think I do try harder, but there is so much promotionalism and utter trivia to deal with that I keep desperately trying to catch up with it, usually resorting to semiautomatic tools and prebuilt notices.) A policy page is exactly where to discuss it. (anyway, all complaints should be heard, telling people to make them elsewhere is the classic way of rejecting outsiders.) The goal of policy is to make an encyclopedia , and the key requisite for making an encyclopedia is to keep attracting new editors. We're not acting like teachers, we're acting like bullies--like children who get to be teacher for a day and can be as arbitrary as they please within our confines. They're usually much more arbitrary than the actual teachers. Our principle is , after all, that everyone can edit; some have limitations, and the goal is then to teach them. What they can't do well enough today they can perhaps do tomorrow." [2]

Action plan[edit]

  1. . find people interested in the change and get them excited
  2. . meet with decision-makers (top-cover)
  3. . continue grass-roots evangelism while you get approval
  4. . enable adaptation -- support grass roots by becoming part of the group
  5. . touch bright spots and get the SHARING. tie those bright spots in with the boss [3]
  • Learn Quality control; learn human resource management; organize quality plan; implement quality control plan.
  • insurgency to take back adult supervision of wikipedia.
  1. civility test for new admins, i.e. block voting against admins who do not have a history of behaving civilly per DGG standard.
  2. mandatory admin training on human resource management
  3. enforce civility among existing editors by wikilove, hug them to death method
  4. grognard cabal of veteran editors to support their projects; work
  5. ignore trolls, drama queens, whose toxic behavior is counterproductive
  6. focus on process improvement, coaching other editors
  7. spread the quality circle method training among veteran editors.

Notes to veteran editors[edit]

15 behaviors of admin privilege:

  1. perfectionism
  2. urgency
  3. defensiveness and/or denial
  4. quantity over quality
  5. worship of the written word
  6. belief in one "right" way
  7. paternalism
  8. either/or binary thinking
  9. power hoarding
  10. fear of open conflict
  11. individualism
  12. progress defined as more
  13. right to profit
  14. objectivity
  15. right to comfort [4]

therefore, suggested actions

  1. do the good enough not the perfect;
  2. take your time;
  3. debating with admins, will be taken as a threat, adopt a therapeutic, coach role, applaud non-defensive comments;
    1. admins enforce rules, they do not provide solutions;
  4. show the quality improvement;
  5. show the images, sound, and video, that are better than "critical commentary";
  6. show the many ways of editing that work, be open to anyone's way of working;
  7. do not talk back, rather talk with people who can team;
  8. show the flexibility of holistic thinking;
  9. the power of the wiki is distributed, route around power blocking of improvement, as an obstruction;
  10. choose your conflicts wisely, to demonstrate a principle, when you are in a no lose position;
  11. show the power of the group; do work with like minded teams, and projects to work on large problems;
  12. show the better quality of the small is beautiful;
  13. show the non-profit improvement, not the publishable;
  14. show the improvement of subjective quality measure;
  15. afflict the comfortable

bullet form[edit]

  • What are we doing here?
  • Who are we writing for?
  • What is the experience of the reader; editor?
  • More civility enforcement; less tit for tat
  • More coaching; less scapegoating
  • More system fixing; less error fixing
  • More teams; less lone wolves
  • More concierges; fewer gatekeepers
  • Process improvement
    • analysis - cause and effect diagram
  • Quality circles
    • form task forces to solve process problems
    • plan > do > check > act
  • Institute training
    • training new editors ambassadors' tutorials; teahouse; mentors
    • training admins on human resource management