User:Jayron32/Precedent makes bad legislation

Frequently, in discussions at Wikipedia, people look to past discussions as a means to decide what the rules at Wikipedia should be. Which is to say, once a decision is made via a consensus-building discussion, that somehow that establishes some kind of guidance for how all similar decisions should always be made in the future.

For example, let's say that I create an article about my pet cat, Spot. Let's also say that we hold an AFD discussion where the article about my pet cat, Spot, is deleted. There are many who then try to argue that establishes a precedent, so that we, as an encyclopedia, now know that we should always delete all articles about cats. After all, we deleted the article about my cat Spot, that must mean that Wikipedia should never have any articles about any cats. Right? That makes sense. After all, we want to be consistent; without consistency, it's just confusing and unfair. If some cats have articles which get deleted, and other cats have articles which get kept, how can we know what we're supposed to do with articles about cats? The system needs some logic, consistency, and fairness, right? Otherwise, it's all chaos!

Discussions are a feature, NOT a bug, at Wikipedia
Many people will use precedent as a means to not have the same discussion over and over again. After all, it can be counterproductive to keep talking around in circles. Once we make a decision, we shouldn't be dragged into discussing it ad nauseum and going around in circles. That's a reasonable idea, until you realize the flaws that go into it. First of all, Wikipedia really does require consensus building, and that always takes time. The intention is to talk everything out. When we don't talk about things, we get into more conflicts than we do. We should never try to short-circuit the discussion process. We aren't in a rush and it is more important we get things done right than to get them done now.

Policies and guidelines are rules. Individual discussions are not
Wikipedia already has a quasi-legislative process. It's the writing and approval of policies and guidelines. There are clearly established process in place to enact encyclopedia-wide rules on how to handle various specific situations. Back to the original example: If Wikipedia should have a "No cat articles" policy, then start that discussion as a policy discussion at WP:VPP. Propose that we enact that rule. Get widespread agreement. If people don't agree, then maybe that shouldn't be a rule. Anything that isn't covered by policies or guidelines, is INTENDED to be discussed and decided on a case-by-case basis, and is NOT supposed to be decided on an encyclopedia-wide basis.

We are not bound to make the same mistakes in perpetuity
Consensus-building discussions are the best tool we have at Wikipedia to make decisions; but they are not infallible. Sometimes, a consensus will have turned out to be a bad idea. Sometimes, we'll decide something, and then we'll see "Oh, that looked like a good idea at the time, but whoa boy, were we wrong". By claiming that decisions of the past are precedent, we're saying they now have the power of rules, and that we are bound to follow those precedents. What if we screwed up? Do we say "Oh, well, we screwed up, now we have to screw up forever!" No, of course not, the entire point of having a Wiki is that everything is always fixable, and changeable.

Foolish consistency
"A foolish consistency is the hobgoblin of little minds." Not all consistency is bad. Wikipedia has consistent rules, which are supposed to be enforced consistently, unless you have a really good reason not to. We do value consistency. What we don't value is foolish consistency, which is to say consistency which is not itself based on firm footing. Take the cat example from the beginning of this essay. The desired consistency (that we treat all articles about cats the same way; if we delete one we should delete them all) is based on a false premise: that all articles about cats are consistent in all meaningful ways themselves, which presumes that all cats are consistent in all meaningful ways themselves. That is a foolish belief. Some cats may have things about them that merit encyclopedia articles. Others may not. Just the mere coincidence that they are both cats doesn't mean that all things about them are the same. My pet cat Spot is not identical in all ways to, say, Gladstone. They just aren't. We can learn all about Gladstone because people have written many things about Gladstone and published those writings in reliable sources. The fact that something has already been written about extensively in reliable sources is a far more important fact about something than the fact that it is a cat. The fact that Wikipedia tends almost always to delete articles about subjects for which no meaningful source texts exist about that thing is a good consistency. Wikipedia articles need source texts, so we consistently require that articles should be based on reliable, high-quality source texts. Consistently applying that standard is not foolish, it's a good idea. But it's also based on policy (WP:V) and guidelines (WP:N), which are rules that the community has discussed and worked on and decided that are good to apply as rules. They did not happen accidentally because one time, a deletion discussion deleted a particular article. It was established as an encyclopedia-wide rule through Wikipedia's normal legislative processes.

In summation
If Wikipedia needs to have some consistently applied rule, there are processes in place to make it so; start a proposal at the appropriate location, like WP:VPP or WP:VPR or something, seek widespread input, allow time for consensus building to occur, and see if the community generally sees the rule as a good idea. If that process establishes it as a rule that we should follow, it will be enshrined as a policy or guideline, and we can begin expecting it of ourselves and of others. If a lone discussion on a single, specific topic or event, whether it is at WP:ANI, WP:RSN, WP:ITNC, WP:AFD, or really anywhere, reaches a particular conclusion on how to resolve that one issue, it cannot, should not, and should never be used as a justification for resolving any other situation no matter how similar you may think the situations are. Every such discussion should be presumed to be a problem unto itself, and justifications for any one !vote or point you might make should never be based on "precedent" established in a similar such discussion, but should instead be based on broad principles such as WP:PAG and the like.