The value of precedent

One of the most important principles in organizing a just society is the equally applicable rule of law. Worldbuilding is, while perhaps seeming trivial compared to a nation, simply another society that we all hope is organized in a just manner. No one doubts that there is plenty of injustice in the real world, but one of the most important tools for ensuring what fair treatment there is before the law is judicial precedence.

That means that a judge deciding a case must look to the precedent set by earlier cases to make determinations as to whether something is on or off topic. This ensures not just fair treatment through time, but also fair treatment to people to people in front of different judges, for all judges have the same obligation to look at precedent.

The applicability of precedent

The use of precedent is something that we should observe on this site. If a question was asked before that is similar in scope, then there is no reason to close a question that is here for review. The standards don't change over time.

The question linked was closed, while the two similar questions that I posted in comments received 18k views (one of them, possibly both, seem to have hit the 'Hot Network' bar), 24 upvotes, and 27 answers. Clearly, these are legitimate questions for this site. By precedent, no one should be voting to close the new question, unless they do not think it is similar to the old question. I would argue that it is not valid to retroactively say that the old questions, so well upvoted and answered, were invalid in the first place.

Experience is a valued companion to precedent

Expecting every close voter to do the due diligence of an appelate court judge when voting on questions is silly. But expecting them to respect precedent when it is brought up to them is not. I am calling for all close voters on this site to respect the decision made by others before you.

In fact, I am calling for fewer 'snap judgements' and more time spent looking through site history. Experience is valid here. I was able to bring up the other questions because I had seen them before and remembered them. This prompted me to take a look through history, where I read about 10 poison related questions and picked two that seemed similar, but not similar enough to close as a duplicate.

Now, not to be accusational, but four of the give users that voted to close were not able to cast such votes when either of the two linked questions were posted. These users simply did not have the experience to know that such a question was in fact on topic, as shown by precendent.

Now, to be accusational, they voted to close anyways.


Of course, the fifth user that voted to close has been on the site much longer than I, and reviewed many more questions than I, and still voted to close. That is, of course, how our democratic voting system works. We will not all agree on what should be closed and for what reasons. And there is no reason to assume that the newer users' reasoning is defective on this particular vote.

But I would urge people, especially those that have not been on the site that long, to take some time to do research before voting on whether or not to close questions. There is a long long history of questions on this site, and a historical review can tell you both whether a question is on topic and whether it is duplicate, all at the same time.

  • 2
    $\begingroup$ Those other questions should be closed too. One of them in fact already has 4 close votes on it. Remember we should put old questions on hold to set an example for new users. $\endgroup$
    – sphennings
    Dec 7, 2017 at 16:14
  • 1
    $\begingroup$ @sphennings Those four close votes have all occurred after I linked it, and brought attention to it. Hence the need for this post, people are literally doing the exact opposite of respecting precedent. $\endgroup$
    – kingledion
    Dec 7, 2017 at 16:20
  • 9
    $\begingroup$ "The standards don't change over time." Actually, they do. Sometimes subtly, sometimes hugely. What's considered on topic definitely can change over time. Few people actively go looking for old questions, but that's no reason to treat them differently when one comes across them, whether in normal browsing of the site or because someone mentions them in chat, on Meta, or in links from active questions. $\endgroup$
    – user
    Dec 7, 2017 at 16:22
  • 3
    $\begingroup$ That's because the question is off topic. Most people have better things to do than brows through old questions. Once you brought it to our attention it should be closed. $\endgroup$
    – sphennings
    Dec 7, 2017 at 16:22
  • 3
    $\begingroup$ Check this out: worldbuilding.stackexchange.com/questions/18051/…, posted two years ago, garnered 200+ votes, and closed. And this, a recent one, worldbuilding.stackexchange.com/questions/92284/…. Based on your precedent argument, fun question that's old and get many votes shouldn't be closed. $\endgroup$
    – Vylix
    Dec 7, 2017 at 18:28
  • $\begingroup$ @Vylix My argument from precedent mirrors the real world. There are exceptions to prove every rule. Not every precedent is sacrosanct, in the case of something that is obviously wrong, see Plessy v Fergusson for example. However, for smaller cases where are multiple similar precedents, deferring to precedence makes sense. $\endgroup$
    – kingledion
    Dec 7, 2017 at 19:03

2 Answers 2


I think there are three parts to this meta question, beyond what is explicitly asked:

  1. The specific question you've mentioned, What kind of poisons could be used in a hospital setting?
  2. To what degree precedent matters, and how the scope evolves dynamically over time.
  3. Close voting (which you're suggesting some users are doing too often, or irresponsibly) without knowing the scope very well.

1. The case at hand.

The question itself asks for a poison that would kill a person while keeping them conscious throughout the process, with the requirement that the poison be hidden on an initial examination but discoverable upon deeper inspection. In terms of question quality, it doesn't have quite as many details as I'd like, and it doesn't show a huge amount of original research, but it's not terrible, and it does give some criteria (especially the one about keeping the poison hidden, which makes things slightly challenging).

I've read through the comments there, and I found a couple that are relevant:

Please can you explain how this question relates to wolrd building?

Almost any medicine is poison if used inappropriately or in conjunction with certain other medications. This is really story building, though, not world building. Should probably be moved

those questions are not on-topic. They are storybuilding, not worldbuilding questions. Even the setting is in this real world.

As I'm reading things, folks are voting to close for two reasons:

  1. They don't think it's about worldbuilding. I don't see how it's not worldbuilding; the OP explicitly states that it is for a creative work they're making, and they need help on a specific detail - not an overarching idea, concept, or plot arc, but a specific detail.
  2. They think it's idea generation.

Let me address the second one in more detail, because the concept of "idea generation" has a long and - if you'll pardon the pun - storied history on Worldbuilding Stack Exchange. I'll try to get it (mostly) right.

For the purposes of this site, "idea generation/plot building" basically means that someone is asking us to come up with an entire idea for their world. For instance, they could ask us to decide how a knight should best rescue a princess being held captive by a ravenous monster. That's building the plot right there. In general, if a question is about the actions of a character, it has a good chance at being idea generation.

So, idea generation questions have several problems, right off the bat:

  1. They're too broad. To use my example above, the knight could raise an army to save the princess . . . or face certain death and try to fight the monster one-on-one . . . or attempt to bargain with the monster - you get the idea. All of these solutions are, approximately, correct.
  2. They're really just invitations for endless discussion, and we're not a forum. Talking with others and bouncing ideas off them is great, and there plenty of places where you can do that; we're not one of them.
  3. We're not a content-generation engine.

Around the time of this first discussion, we picked "Idea Generation" as a custom close reason, and used it for a while. Monica suggested nixing it about eight months later, mostly because it wasn't doing anything that Too Broad/Primarily Opinion-Based wasn't. It took us a while after that, but we finally got rid of it, especially after some important discussions the following February, where many users expressed concerns that we were allowing too many "What-If" questions (and that's another saga in itself).

. . . But I digress. The point is, idea generation questions are usually overly broad and poorly constrained - or not constrained at all. I will argue that this question is not an idea generation question, for several reasons:

  • It's not asking us to build a plot (not the only criterion for idea generation, but a common one).
  • It does place constraints; it's not just asking "What's the best poison I should use to kill someone?" That is an unconstrained question.
  • It has more than one right answer, probably, but you can objectively tell whether one is better than another. For instance, one poison might show up on an initial autopsy scan, while another might not; the first one is then better. Even more constraints could be added to the question as it is, which would be nice.

I cast the fourth (and binding) reopen vote.

2. Precedent.

Precedent on a Stack Exchange site is a tricky thing. On the one hand, it's great to refer back to, and old questions can tell you what to do with new questions. On the other hand, it does change. I know of plenty of cases on this site where once-good - and even quite popular - questions were closed because people realized that that class of question was not a good fit here.

In some of these cases, we've had (usually minor) scope changes. And that's fine. We live, we ask, we answer, we learn, and we see that some topics work while others don't. I think that that is - and should be - the primary driver behind scope change: experience about what works and what doesn't. This is sometimes why it takes a while to implement, change or remove policy - we don't have a large data set on a specific topic. And you need data - making meta discussions based on just theory is a recipe for disaster.

I don't think it's a problem if we change the scope over time. There are some things that we need to ensure happen if we do - and these are quite hard to implement, at times:

  • We close old questions, too, for consistency and fairness. That way, new users won't ask off-topic questions because they think old ones are still on-topic.
  • We make sure people know about the change; if it's of a high enough magnitude, a meta post may be appropriate to sum up the discussion (alternatively, the discussion itself could be featured).
  • We reach something of a consensus, and as many people as possible get a say. We don't want things happening behind closed doors, and we don't want a schism because half of the site can't agree with the other half of the site.

This is all very hard and messy. I don't like it when we have to go back and close bunches of old questions, because we have to make sure that we don't miss any spots, and it's all very complex. Figuring out what's covered by the new standard is also hard. Fortunately, we haven't historically had many of these large-scale policy discussions in the three or so years this site has been around. Other sites have.

3. Close voting

I will say, first of all, that it's emotionally challenging to see a site grow into something much bigger and almost uncontrollable than where it started out. I missed the private beta, but I joined very shortly after the public beta started, and I've been here for over three years. I've been amazed at how the site's evolved (and succeeded!). I still remember a lot from those days.

I'm so happy that we've done well. I was ecstatic when we graduated; it made my week. But I'm also - totally selfishly - sad to see it flutter out of my hands, because now I can't see how every corner of it works, or how every question evolves, or have the chance to be a part of every discussion. I might be a mod, but I have very little control over the site. It's . . . different.

And so - and I say this gingerly, and delicately - I kinda sympathize with you being annoyed at newer people coming into the sandbox, so to speak, and not really knowing the history of the site. I get it; I've been there - both as a new user (!) and as now something of an old hand - though certainly not the oldest.

Here's what I'll say to those with more experience on the site:

What I would do - in an ideal world, where I have enough time and energy and waking hours - would be to look at many questions in depth, and have a dialogue with those voting to close and/or reopen, and make sure that we all understand why a question is being closed. Everyone who gets the privilege to close and reopen does have a bit of a struggle trying to understand the nuances of the scope, especially on such a weird site as Worldbuilding Stack Exchange. I think it would behoove all of us with experience to help them.

Here's what I'll say to close voters in general:

Let's make it clear, on each question, why we're voting to close or voting to reopen. If we can cite a specific meta discussion, then great. In fact, that's ideal, because it helps the asker formulate better questions in the future. By giving them specifics and details about what questions the site likes and what questions it doesn't, we make it much easier for them to ask good and on-topic questions in the future.

Another relevant thing to think about is Frostfyre's reminder to close-voters. Their standard comment is

Close-voters: Please don't vote to close questions without leaving commentary. The problem cannot be fixed if the OP is not made aware of it.

It doesn't matter whether the votes are cast as off-topic, too broad, primarily opinion-based, or unclear: There should be more explanation of what the OP can do to fix things. We owe that to them.

  • 5
    $\begingroup$ I'd like to tag onto your last paragraph our general Reminder to close-voters thread, if I may. $\endgroup$
    – Frostfyre
    Dec 7, 2017 at 20:52
  • $\begingroup$ @Frostfyre Excellent idea; done. $\endgroup$
    – HDE 226868 Mod
    Dec 7, 2017 at 21:20

There is an explicit rejection of precedent in the existing discourse on both this both for this site in particular and the entirety of stack exchange.

To quote the accepted answer on Meta:

View all question with today's standards. If the question fits as per current standards then leave it open; if it doesn't fit then vote/flag to close it. If we keep the questions which don't fit as per current standards, then people will ask why the questions are still open.

If we don't do this then there will always be hidden gotchas for new users and people newly at 3k rep.

Questions that have been closed that have positive votes or answers are protected from auto-deletion. So there is little risk for well received but closed questions from going away.

  • 1
    $\begingroup$ This is fundamentally Wrong, as in wrong in the sense of cosmic justice. If new users are not bound to precedent, then a group of them can simply decide on a new standard, and retroactively apply those standards, even over the objection of old standards. Of course old questions slipped through the cracks, but ask yourself did that popular question slip through the cracks (with its 12k views) or are you just retroactively deciding on a different standard. $\endgroup$
    – kingledion
    Dec 7, 2017 at 16:24
  • 4
    $\begingroup$ A year ago I wasn't participating in community moderation. I don't see what's wrong with changing the standards. New users are bound to the conventions of the site like anyone else. $\endgroup$
    – sphennings
    Dec 7, 2017 at 16:29
  • 2
    $\begingroup$ Did you know that you yourself have cast 17.5% of all close votes on this site since September 1st? Or that you voted on 88% of the questions that were closed in that time span? Are you sure you have community buy in? Are you sure you are not just driving questions off this site? $\endgroup$
    – kingledion
    Dec 7, 2017 at 16:35
  • 5
    $\begingroup$ @kingledion If you feel that a given question which has been put on hold or closed really is on topic, all you have to do is vote to reopen. That will enter the question into the review queue where people will see it. If you feel that there is precedent for that it should be on topic, you can post a comment to that effect, linking to examples elsewhere on the site. If a comment doesn't allow the space to fully explain your reasoning, then as I am sure you are aware, we sometimes see discussions on Meta whether a particular question should be open or closed. Remember, assume good intentions! $\endgroup$
    – user
    Dec 7, 2017 at 16:41
  • 1
    $\begingroup$ @MichaelKjörling Having reviewed close voting patterns in response to this question, I have evidence of not-so-great intentions. A few people have joined the site in 2017 and, as soon as they are able, began casting massive amounts of close votes. By hook or by crook, I intend to stop this behavior. $\endgroup$
    – kingledion
    Dec 7, 2017 at 16:45
  • $\begingroup$ @MichaelKjörling An appeal to precedent is one way to attempt to divert this behavior in a fair way, without calling people out unnecessarily. $\endgroup$
    – kingledion
    Dec 7, 2017 at 16:47
  • $\begingroup$ @kingledion If you have evidence I'd like see it and your conclusions from it. Sometimes it's necessary to call people out. Or if you can show the behavior without listing usernames so we can discuss what is going on and propose solutions. $\endgroup$
    – sphennings
    Dec 7, 2017 at 16:48
  • 5
    $\begingroup$ @kingledion It's pretty cool that you take the time to analyze the close voting patterns and it looks like you already have some interesting first results from your search. Out of curiosity: how difficult would it be to adapt your research to reopen votes? It might be interesting to take a look at potential differences in the communities current behaviour when it comes to closing and reopening. For example does it look like people think "close everything" and "reopen nothing"? Or is there a tendency for "close fast and reopen fast", which would be the ideal in my eyes? $\endgroup$
    – Secespitus
    Dec 7, 2017 at 16:53
  • $\begingroup$ @sphennings Then, in your case, you are currently #6 on the all time close votes list, since the site started. You have 861 close votes, while the highest has 1120. At your current pace of close voting, you will be #1 by the end of January. You have 4 times as many close votes as MichaelKjorling, who was a moderator in the site Beta, I believe. I think this is evidence that you are voting to close too much. $\endgroup$
    – kingledion
    Dec 7, 2017 at 16:56
  • $\begingroup$ @Secespitus I just figured out how to do that. Post a question in meta. Keep in mind, I'm still trying to figure out how to ask about people who closed questions together, so I can just do the basic counting stats for now. $\endgroup$
    – kingledion
    Dec 7, 2017 at 17:11
  • $\begingroup$ @kingledion I posted a question about reopen voting patterns. Any data would be nice because we don't have anything at the moment as far as I know. Maybe someone could work from your starting stats. Together the close and the reopen data might help to decide as a community in which direction we want to go in the future. Thanks for taking the time to analyze voting patterns! $\endgroup$
    – Secespitus
    Dec 7, 2017 at 17:21
  • 4
    $\begingroup$ @kingledion I was a moderator during the public beta, Oct '14 to Feb '16 if I'm not mistaken. I've also held a diamond since Aug '17. If you look, I believe that you would find that these periods correspond to significant dips in my VTC activity (though not necessarily in the number of closures I participated in, which is a completely different metric), because it's one thing to participate in closing as an ordinary member of the community, and quite another to close by mod hammer. Some of that data would however be skewed by UX details in how some site and network functionality is designed. $\endgroup$
    – user
    Dec 7, 2017 at 20:09
  • 3
    $\begingroup$ One concrete example for the latter would be migrations, which require that the question be reopened and re-closed. (Yes, it's a pain, but there's currently no way to immediately migrate to another site in the network a question that is on hold or closed; it must be reopened first.) If you have data, then please strongly consider posting it, preferably anonymized. You could offer users who speak up to deanonymize their data, but not other users'. In my opinion, it's much easier to discuss actual hard data than it is to discuss vague allegations supported by only two or three examples. $\endgroup$
    – user
    Dec 7, 2017 at 20:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .