Wiki studies/Wiki disease
This is an essay by User:Abd. It is open for editing, but editing should reflect either the opinion of Abd (as shown by Abd's consent), or consensus. Other opinion may be added if attributed, and, as usual, comment on the Talk page is fully welcome. --Abd (discuss • contribs) 14:49, 12 August 2015 (UTC)
I have generalized the name to "Wiki disease" because it is my hypothesis that this is a general wiki affliction, though enhanced by Wikipedia conditions (flat mainspace, one article per subject, keep/delete decisions with an excluded middle, encyclopedia project with "sum of human knowledge" being interpreted as "summary," which then introduces point of view intrinsically, etc. What is important ("notable") to one is not important to another. While Wikipedia attempts to deal with this by creating relatively objective standards for notability, in practice, they are applied with high variability, unreliability, and outcomes are often not readily predictable from application of policies, etc.)
The term Wikipedia disease comes from the writing of User:JWSchmidt, a major founder of Wikiversity, who was highly distressed and offended by what he saw as Wikipedian thinking that afflicted Wikiversity. He complained about it, with extensive posts, that were considered disruptive. WMF wikis respond to disruption by warning and blocking, so he was blocked, thus demonstrating the operation of "Wiki disease" here. This is not a claim that JWS was not disruptive. He was, as is common for people who are upset. However, it is possible to channel such disruption into constructive work. This was not skilfully done. To some extent, as it was attempted, JWS resisted it. That is yet another symptom of "Wiki disease." Wiki disease creates and maintains conflict.
Wikipedia also deletes pages considered inappropriate, whereas Wikiversity, as envisioned by JWS, was highly inclusive. Associated with JWS's upset were out-of-process page deletions. As well, he was originally extremely popular here. Overwhelmed by an influx of "Wikipedians," bringing in Wikipedian ideas, his upset boiled over and he was then emergency desysopped, banned for a time, though that was overturned, and then indef blocked by someone who should not, from history of involvement, have touched him with a ten-foot pole. I'd be upset too. Come to think of it, I was blocked by the same custodian....
or ... a collection of symptoms.
- Ad hoc administration based on the opinion of the administrator as distinct from clear application of policy. (This was enshrined in w:WP:IAR, which is the common-law principle of "public policy," judicial discretion, and a lot else, but which, when combined with a fusion of judicial and executive responsibility in wiki administrators, encouraged unpredictable and very personal administration.)
- Loss of functional administrative responsibility by the community. (The community may be happy to have someone else handle administrative issues.)
- Administrators preferring to support each other in conflicts with ordinary users, "circling the wagons."
- Administrative control of policy, blocking the development of consensus that might regulate administrative action.
- Participation bias, because policy pages and other central process will be preferentially watchlisted by administrators.
- Warning and blocking based on dissent.
- Deletion of critiques.
- Community bans dominated by administrative or factional votes. Guidelines on community process require decision by arguments, not votes, but closing often are clearly based on votes, and the process guidelines also suggest that the discussion is among uninvolved users, but I have never seen a close where this was considered, where factional participation was obvious.
- Loss of general community participation in central decision-making, which becomes dominated by administrators and a tight circle of associated users, perhaps sycophants, sometimes wanna-be administrators, sometimes simply factionally aligned by POV.
- Difficulty of entry into a closed circle of administrators. Minority views are excluded, and any dissenter from the dominant group can never be elected, so the administrative "cabal," as Jimbo called it when he established it on Wikipedia, does not represent the diversity of the community.
- Because an active administrator in a system like this will make enemies, it is understood that administrators may not be reaffirmed by "consensus" if they are again subject to vote, so admin-for-life becomes the standard. Those who would want to be administrators, then, wait to act according to their own ideas until after they are elected.
- Appeal process from administrative decisions is dominated by administrators.
- Administrative access may be granted based on technical competence rather than vision and service to the community, or, for example, skill at negotiating consensus. Governance becomes technocratic.
- As central process becomes an administrative responsibility rather than a full collective one, the burden of site maintenance in the face of spam and vandalism falls increasingly on administrators.
- As administrative burden increases, administrators become impatient. Original guidelines of graduated warnings, followed by graduated blocks, become, increasingly, immediate indef blocks without discussion, treated as bans unless overturned, which becomes increasingly rare.
- Hostility develops toward anyone seen as critical of administration.
- Minority elements in the community simultaneously develop hostility toward administration, and call error "abuse."
- There is a general loss of the assumption of good faith, which requires patience and tolerance.
- Drama increases, and is blamed on outsiders and "disruptive users." The community does not take responsibility for what it creates by action and inaction.
- Those defined as outsiders sometimes resist, becoming block evaders, "long-term abusers," and vandals, which then increases the need for enforcement labor, increasing the perception of the project as beset by enemies, needing vigorous defense. Which also leads to burnout.
- The wiki, often highly innovative in origin, becomes highly resistant to change, even when changes are purely experimental and optional. The actions of anyone working for reform are scrutinized, looking for reasons to disempower and sanction them.
These are wiki breakdowns. Each of the above can be documented. They are not evident when wikis are small with a monocultural community. These are manifestations of general human organizational process, they are normal structural phenomena. They are not caused by "bad users." They are the result of structure that fosters the development of these symptoms, or the lack of structure that creates alternatives.
To some extent, "wiki" itself encourages this. "Wiki" means "quick" in Hawaiian, and wikis were designed for quick and easy editing and growth. However, this very advantage then creates a bias against anything that requires depth, study, reflection, extensive discussion before decision, etc., i.e., that requires more traditional deliberative process. That process is not directly usable with a large community, for well-known reasons. (This is called the "problem of scale in democracy," an actual book title of an academic study.)
The classic solution is representative democracy, and the best forms of that use proportional representation, not the approval-at-large method used for Wikipedia Arbitration Committee elections, which is the closest entity to a representative, deliberative body in the WikiMedia Foundation wiki system. Without proportional representation of some kind (or proxy systems, even better), the deliberations do not fully represent the community, so it typically cannot resolve factional disputes. Instead it commonly decides to ban one or another of disputing parties, deciding they were "wrong," instead of seeking collaborative solutions. That's "quick," but never resolves disputes, it just sweeps them under the carpet, entrenching them. It is common for the AC to determine a new policy or principle, and to sanction editors for violating it, as if they could be expected to know in advance. Alternative solutions are ignored, considered impossible, banning gets the "problem users" out of their hair.
The same phenomena arise on small wikis without an arbitration committee, where administrators make such decisions ad-hoc. It has been evident on Wikiversity, merely more rare.
Wiki communities, then, naturally become "abusive," if this is not recognized as a hazard and handled. Not understanding the natural character of this "abuse," users then routinely attribute the phenomena to "abusive administrators," Which simply amplifies the effect, as those who have frequently worked very hard for the improvement of the wiki, as they see "improvement," defend themselves and other administrators. (And elements within the community also defend them; the community becomes divided; dissenters either drift away or are blocked and banned.)
There is little or no proven treatment for Wiki disease. There are ideas that sometimes have had some limited effect, they have never been implemented on a large wiki. The basic problem is created by a large community that is not involved in governance, that leaves this sometimes tedious task to a core group. That, in turn, is probably necessary for most users. We don't have time, and structures that would allow efficient participation are missing.
Strong-leader governance and oligarchical governing structures evolved to handle the problem of scale. However, these do not tend to generate true consensus and have other negative side-effects. Hence democracies evolved, and representative government is the closest yet to government by social consensus, it can approach it. In the twentieth century, consensus organizations arose that attempted to use full-out consensus, which was often exciting because the full consensus generated created high organizational unity. However, this was often not sustainable, because full consensus takes very substantial discussion, and most participants, after a time, don't have the time for it. It was not efficient.
The classic solution is representative governance, but most people are not familiar with election systems that generate full representation. w:Single transferable vote, used for proportional representation, can come close, but often depends on a party system. There is a technique developed originally in the 19th century by a major architext of that voting method, w:Lewis Carroll, now called w:Asset voting. As is not surprising, mention of this on en.wikipedia has been suppressed (but not entirely! as I notice today). (The termy "Asset voting" is a neologism, but the concept has been covered in academic publications, to some degree.) A related system was proposed for en.wiki, as an experiment: w:PRX. It was immediately attacked, with surprising vehemence, see the MfD. It was just an experiment with a file system, totally voluntary, no effect on policy.
An aspect of wiki disease is extreme, reactive conservatism, that will oppose even attempts to study what actually happens, and to experiment with alternatives. See also w:WP:Esperanza, another experiment that was crushed as soon as it appeared vulnerable. This was ironic. "Hope" was crushed.
We have an experiment here on en.wikiversity, the WV:Assembly. It had a very moderate success. If these experiments depend on one user (i.e, me,) success will remain limited. Leaders are needed, and not just one.
"We have met the enemy and it is us." (Walt Kelley)