Wikipedia:WikiProject Reforming Wikipedia/Governance

Wikipedia's open editing philosophy and vast collection of content requires a robust system of governance to keep editors in line, content of quality, and the encyclopedia structured and orderly. However, a swelling bureaucracy and a culture of power and leadership have undermined our governance structures and the quality of content (Suh et al. 2009, p. 8). Many new editors experience difficulty navigating our governance systems, and more experienced editors have focused less on content and more on governance, to the detriment of our readers' experience (Suh et al. 2009, pp. 4 & 8).

Our governance structures have contributed to Wikipedia's slowing improvement and its usefulness to readers. It has bred an increasing culture of resistance and power, with administrators using their tools—such as protection, deletion, and blocking—more often and more controversially. Occasional editors with no stake in our governance are finding editing increasingly difficult, with increased resistance between administrators and non-administrators on issues of policy. New editors, especially academics, are now a scarce resource thanks to our overly controlling and oftentimes hostile bureaucracy.

Wikipedia's governance is in need of major reform: it must be respecting of newcomers and occasional editors, responsive to the needs of articles and the editors editing these articles, and maintained as a flat structure, independent of a culture of power.

ArbCom

Remove and distribute ArbCom responsibilities that are either irrelevant to dispute resolution or inappropriate for untrained volunteers.

A primary example of the former is ArbCom's handling of functionary selection and abuse allegations—selection could simply be handled in a community election, and abuse allegations are better handled by the specialized Ombudsman commission. Additionally, volunteer arbitrators simply have no business being involved in cases involving privacy or legal matters (personal information, harassment, child protection, etc.). Such cases are enormously stressful for a mere volunteer, and they have real-world effects—in many of these cases, people's personal, off-wiki lives are involved. Therefore, it is wholly unsuitable to place these matters in the hands of encyclopedia volunteers with zero training in handling these situations. We should turn these matters over to WMF professionals.

Streamline ArbCom case procedures, perhaps by:

implementing firm case deadlines;
focusing more on the specific dispute of the parties instead of allowing every case to turn into a massive, free-for-all community war;
only addressing the parties' assertions in the proposed decision, without the arbitrators introducing any new "facts" that were never mentioned by the parties.

Reduce the use of outright site bans by ArbCom. When the problems are plainly centered around specific editors or article topics, topic/interaction bans should be used. Site bans can be used in the event of broad behavioral issues (such as a general tendency to attack anyone who disagrees), or when lesser, more targeted bans are not respected.
Encourage fair enforcement by mandating that all AE requests be left open for a reasonable amount of time. Currently, AE blocks can be instantly and unilaterally imposed by any admin without discussion, and such blocks are practically irreversible. It is important that all accused editors have the opportunity to make a statement and be heard by multiple uninvolved admins. A minimum time requirement would promote these fairer procedures.

General

Deletion: Reduce overall use of deletion by simply "draftifying" (moving to the the Draft namespace) most articles that we would currently delete. Basically, new articles could still be speedily deleted under the general criteria, but for the most part the criteria for articles would no longer apply. Any article that would ordinarily fall under that criteria would simply be moved to draft space. The same would apply for the majority of articles that are rapidly deleted via AfD under the current system—for instance, mere lack of notability would become a reason to draftify, but not delete. If the draft is abandoned, it can simply be deleted under the G13 criteria and restored upon request (perhaps a bot could be created to delete G13-eligible articles).

If implemented, this reform would be a massive improvement of the editing environment for new users—in the vast majority of what are currently deletion cases, this would give new users time to improve their articles rather than being immediately bitten by large red tags and swift, complete removal of their article by people citing cryptic rules they were unaware of. It would likely be a step toward increased editor retention, with which Wikipedia has a known problem.

Templates: Improve templates by reducing, to the greatest extent possible, their currently impersonal and intimidating nature. With the exception of high-level vandalism warnings, we should eliminate hostile and frightening elements (red/orange icons or colors, bolded warnings, etc.) and introduce a more friendly tone to the text itself. The templates in their current state are rather harmful to editor retention—oftentimes, new users who are simply trying to help in good faith find themselves hit with bright templates that threaten deletions and blocks. Of course, personal messages are always best, but it is obvious that not everyone can or will take the time to manually write a personal message for everyone. Therefore, we should try our best to improve templates.
Orderly noticeboard discussion: It is widely acknowledged that Wikipedia noticeboards (the infamous ANI, in particular) are dysfunctional. For examples, look at the many different redirects to ANI and certain pages on the issue. The problem is that although almost everyone acknowledges the problem, there has been no real effort to fix it. One critical step in that direction is incorporating more structured discussion formats into noticeboards to keep discussion orderly and on-topic. The current practice of permitting free-flowing, infinitely threaded discussion can easily cause convolution and stress. The near-total lack of structure and order in noticeboard discussions is a major reason why they so often spin out of control and become free-fire zones.

One of the main problems with AN/ANI is that participants there often propose the blocking or banning of an editor, which is then agreed to in a pile-on vote (commonly known as "the mob"). But unfortunately, it is not uncommon for the alleged wrongdoer to hardly have a chance to defend themselves before the pile-on begins (or if they do comment, it is drowned out by those bringing the allegations). Therefore, participants do not always hear both sides of the case. If the process is to be truly fair, however, it is critical that the accused are guaranteed the opportunity to fully respond to the allegations against them. Therefore, it should be mandated that before the commencement of any vote to sanction, all accused editors are to be given a certain amount of time (24/48 hours, perhaps) to post a full response in its own section. Perhaps a group like the now-defunct Association of Member's Advocates should be revived. The simple but unfortunate fact is that in hostile environments like ANI, the statements of new or simply unpopular users are sometimes disregarded, no matter what they say.

Blocking: Keep blocking preventative, not punitive. Tighten the blocking policy so that admins can unilaterally block an editor only in uncontroversial cases of obvious policy violation. Require that blocks be subject to general agreement if the circumstances are less clear and the need for blocking is more questionable. Create new alternatives to blocking.
Protection: Use "hard protection" (e.g., protection that completely blocks editing) much more sparingly:

Use pending changes protection in place of semi-protection.
Another possible alternative to protection for vandalism is to allow ClueBot NG to make more false positives (by decreasing the threshold) and improve its vandalism detection when reverting edits on frequently-vandalized articles.

Notes and references

Notes

References

Butler, Brian; Joyce, Elisabeth; Pike, Jacqueline (2008). Don't look now, but we've created a bureaucracy: the nature and roles of policies and rules in wikipedia. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. pp. 1101–1110. doi:10.1145/1357054.1357227.
Suh, Bongwun; Convertino, Gregorio; Chi, Ed H.; Pirolli, Peter (2009). The singularity is not near: slowing growth of Wikipedia. Proceedings of the 5th International Symposium on Wikis and Open Collaboration. Article 8. doi:10.1145/1641309.1641322.