Wikipedia:Arbitration Committee/June 2008 announcements/Consensus seeking processes

Following a June announcement by the Arbitration Committee, the Arbitration Committee presents some views on consensus seeking processes, and urges users in the community to consider the proposals presented to address inertia, stasis, and problems currently seen in difficult dispute resolution cases, that can be traced back in part to problems of consensus-seeking approaches.

Consensus
Consensus is central to the community, the Wiki ethos, and our project. Wikipedia is a "grass roots" organization whose strength derives directly from the ability of its volunteer editors to work well together, and where within reason, any user who edits well and gains a modicum of repute, has equal say on content matters, and is encouraged to conduct themselves to a high standard as an administrator, or to seek RFA if they wish it.

Any entrenched systemic problem that seriously affects the fundamentals of consensus seeking is therefore worthy of most serious attention.

The following is our analysis, and (following reflection) our suggestion to the community.

Scenarios worthy of particular note
There are two main areas identified where the community seems to commonly have difficulty with our consensus-seeking approaches. Both are extremely frustrating and wasting of volunteers' time and their goodwill:


 * 1) One is, a number of proposals and discussions routinely take place which are not closed and therefore it is unclear whether consensus is reached, and if so what it is. Without this, there is little clear authority for deciding "what this all means", and reluctance to act or not act. The matter drifts. Or else, a user enacts it only to have their interpretation of consensus reverted later, or challenged as "acting without consensus" (note - silence, apathy, or "a lengthy debate that peters out" as a conclusion is by its nature rarely easy to interpret)
 * 2) The other is significantly large and contentious cases, often where endless straw polls and discussion have routinely devolved into wasted energy, divisive disputes and "yet another attempt", perhaps ending up a while later at Arbitration, or perhaps just never being decided and drifting.

These two classes of consensus problems seem structural in nature, involving to "how we operate consensus", rather than the actions of specific (namable) disruptive users.

Discussion
In many matters, consensus is taken care of by principles such as "Be Bold", or Bold, revert, discuss, which encourage users (correctly) to evaluate their own views and consider others' views as well, and then if appropriate, act thereafter. The principle is, Wikipedia is a wiki - if it is undesirable, then others will revert or seek discussion, following which an amended decision is made. This breaks down or is difficult at times, and the above two scenarios can be common problems.

Whilst some may feel these are by nature insuperable, we note that actually, neither is especially difficult (although they may require work - no surprise). Both of these are routinely solved on the wiki, and have been on more than one occasion in the past. So it is possible; the know-how is there.


 * The first category of problem


 * To the first of these issues, we note that our deletion processes consider highly contentious deletions routinely, where many users present many views and there is heated discussion. We do not have a problem with "any administrator" (or even sometimes any experienced user) closing such debates to a decision, and concluding what the consensus or "best result" appears to have been. Even the most difficult DRV gets closed routinely this way, and these can be very high profile issues (high profile BLPs, BJAODN, Ezparanza, etc).


 * So we do not believe there is any basis for a claim that discussions on a simple matter cannot similarly be drawn to a close if needed, by an uninvolved administrator or experienced user who assesses whether consensus was found and if so what the result should be. Indeed, such closure is essential for consensus-seeking decision-making. (See Arbitration decision: Effective consensus decision making: Closure.)


 * The second category of problem


 * To the second, a number of highly divisive and major disputes matters have come to Requests for Arbitration over the course of time. In some of these, the remedy given was to hold a proper consensus-seeking exercise. Which took considerable effort, but routinely with serious effort from all sides, did actually prove to work, where disputing and past efforts (often "much talk, not enough methodological structure") had previously failed. These cases often were solved via multi-stage decision making that checked all issues as they progressed - what the matter was, who would supervise for fairness, what steps would be followed, and so on. At the same time if there was disruptive activity (non-consensus moves/deletes/tags) these were restricted for the duration. This combination was what had not been done, beforehand. When done properly, it has proven a good approach, but as said, needs careful set-up and oversight.


 * Example cases include: Arbitration/Highways (link to resulting consensus-seeking process)... Gdansk naming (discussion/setup, vote)... and (currently) the approach being taken by some users to look at the long standing issue of RFA reform (RfA Review). Each of these has the feature in common that setting up the debate in a structured manner agreeable by consensus was as essential or more so than actually broaching the issue at hand. There are probably a range of other debates from time to time that would benefit from this type of approach too.


 * Ultimately this approach does seem able to work, but it takes effort and goodwill to set up, is costly in terms of time, and is probably efficient mainly for significantly entrenched cases where other efforts have not helped and a good resolution is sought (although its approaches may be adapted and scaled down for lesser disputes).

Suggestion to community
We recommend the community that some users may consider proposing some kind of process, similar in style (and perhaps with assistance from) the Mediation Committee and Cabal members, to assist with consensus-seeking processes on the wiki, a form of "consensus commission" similar to some electoral commissions, that assist in ensuring the process works well and is overseen by users with experience in helping consensus form.

This would save cases coming to Arbitration, which we can often do little with except remove or restrict blatant disruptive editors and say to all concerned, "now go away and talk properly" (which they could have done anyway in the first place).

A possible draft approach follows as a starting point.

Suggestion detail
A "Consensus Commission" (or other name) would be created. Its role would be to act as a resource and help-point, for wiki-issues where help is needed to reach a consensus, or for an uninvolved user to review and close a discussion (in a manner similar to WP:SSP, WP:AFD and WP:DRV closes), and determine whether a consensus appears to have been reached. An appeal method might be needed, perhaps similar to  and with anti-abuse provisions, but that's something we have on most processes.

Like the Mediation Committee, the consensus commission would have no "formal" authority to bind parties. It would have a noticeboard where users may link to sections (project or talk page) where 1/ a discussion has taken place and needs review and a closure, or 2/ a problem is stated in neutral terms needing a neutral independent consensus seeking process to be facilitated. (An alternate possibility is the use of a tag such as  linking into a patrollable category, which any user may post to the same effect.) It would probably have a core of dedicated users (as MedCom and MedCab do) and a number of users who wished to patrol independently at their will.

If the inquiry is a simple request to close, then closure would be easy by anyone who wished, and the function of the Commission and its noticeboard or category would simply be to publicize and promote discussions needing clarity and closure. It would also be a resource, if agreement was reached to engage editors with consensus-seeking experience, on complex disputes. In these cases a member (or members) would work with parties as a kind of guide, mentor, advisor, or mediator, to
 * 1) understand the problem and why consensus was hard
 * 2) help formulate as mediators with experience in consensus seeking discussions, a possible consensus-seeking process that all sides could agree with. This might be single stage, or multi-stage, as needed.
 * 3) if needed, supervize, guide, or assist in running the polls, or interpreting their results, or arranging for others to do so, to ensure smooth operation of the resulting process decided upon.

Resources

 * Requests for arbitration/Naming Conventions - the main Arbitration case where we considered consensus seeking processes in depth.
 * See also Consensus decision making and Consensus for ideas.

Action
As Arbitrators, we do not propose to create this process, but we do feel it would be an exceptionally positive and worthwhile endeavor for any users wishing to. Our approach is one suggestion, but it's open to modification. Whatever the outcome, some way to handle these cases, and help others needing resources, advice, and help with them, would be good for the project.

FT2 (Talk 14:29, 27 June 2008 (UTC)

For and on behalf of the Arbitration Committee