User:Aaron Schulz/Consensus

Preamble

This is a short explanation of the basics of what represents consensus.

We have seen a lot of differing ideas on how to achieve "consensus", anything from open votes, technocratic votes, rotating committees with that overlook discussions and what not. I think that before proposing any kind of consensus measuring procedure, one must first understand the basics of what a consensus, as a whole, really is.

Factors of Consensus

Most people around here tend to factor in some combination of the following factors when measuring "consensus". Ideally, if possible, I'd like to weigh all of them.

General factors for the discussion

Number strength

This factor simply consist of measuring how many people I agree on what on the page in question. In a simple vote, this is typically the only thing people consider.

Homogeneity

One issue that determines the strength of consensus is how monolithic the given rationales are. Are most people in agreement or are there many various different reasons floating around.

The whole community

In the case of Wikipedia, the community is the "whole". Any page may have a few votes/comments that seem to suggest a local consensus in favor of one thing, but that does not guarantee that it represents the community as a whole.

In some cases, an apparent consensus on a small poll or page may suggest an idea contradictory to a much broader consensus elsewhere or a long-standing policy. In this case, we need to weigh in the whole community, so it may be best to void the small consensus. The discussion could even be redirected to that of the contradictory ideas at the proper mediums, perhaps reviving old discussions.

Last minute information

Is there very important information that came in right before the time to close? This is a serious issue especially if at applies directly to some of the rationales given or if similar information has historically impacted such discussions. This could be grounds for a discussion/vote extension.

User opinion/vote specific factors

Community membership

How long a user has been part of the community and how much they have been involved determine their commitment to the project. Additionally, after being here for a while, users tend to build at least some basic understanding of our policies and objectives, one that newer editors lack.

Established members of the community are given more weight than inexperienced very new users. Any user that has been here for around a week or so and has done some editing is effectively an established member.

Rationality/truthfulness

Another issue is whether or not to check for simple factual correctness or basic, at least, surface-level attempts at logical reasoning when consider whether not a set of opinions is a consensus. This can get to be tricky and subjective if one injects there own reasoning into this too much on deep levels, on the other hand, we ALL know that if someone says "X deleted Y", when that clearly is an incorrect statement or "A-->B, B is true, therefore A" that is is wrong even on the surface. Going any deeper usually is just too subjective.

That said, it is sometimes useful only to consider the "consensus of people holding rational/factual views". It is important to note that "rational" does NOT mean "right", but it is merely a way of saying that the person attempted to make a logical/rational connection between facts, whether you agree or not. "Factual" only references surface level facts, not conclusions or ideas.

Another thing to consider is which points where verbally addressed and which ones were not answered.

Foundation issues

See m:Foundation issues, these cannot be overridden by a local wiki consensus that is still under WikiMedia, whether it is that of a small group of users or a larger one. Other things like WP:V are up there.

Opinions/votes

So now lets get down to the idea of how to "weight" an opinion or vote. In normal voting, each vote has a weight of 1 no matter what, as long as it is not a clear sockpuppet, otherwise it is 0. Consensus discussions are not so binary though.

However, most votes/opinions only directly represent a small tiny fraction of the whole community. Whether or not the apparent level of agreement or local consensus represents the Wikipedia consensus depends greatly. It is hard to get the "right answer" or the "exact number", but it is often possible for a trusted user, or set of users, who are very open and level minded, to be able to reasonably approximate these things, usually for issues they themselves are not particularly involved in.

That said, there are three main factors to consider when weighing a vote/opinion:

• How strong of a community member is the voter/commenter? (see discussion above)
• Was this vote/comment based on an convincing attempt to be rational?
• Was this vote/comment based off any obvious factual error?

Additionally, we have the issue of:

• Was this comment made in ignorance of an extremely important facts that were NOT presented before the time of the vote/comment? (such as a really serious diff coming up right before a vote closes)

The issue of "ignorance of an important fact" can be tricky, as what is "important" can drift into subjectivity. I would say that if, by past patterns, you could predict that a large number of people would have opposed and/or votes/comments would have been different in light on these facts, then it is "important". Your own personal view is not enough to consider the new facts as "important".