-2

However, please be sure that the below answer actually answers OP's question before voting to close! If the question asks about something not covered below (other than the specific fields), the question should be left open until its answers are merged into this canonical question.

In (How) Can I switch from field X to field Y after getting my PhD?

2

4 Answers 4

2

I do not have moderator powers and don't anticipate earning enough rep for them, so I'm hesitant to comment on the precise wording of the preface text.

But as I requested it, maybe some further explanation as to why would be useful.

As a totally anecdotal observation, this Stack closes questions more often compared to some others. I think this is good-- we have relatively low volume, so can afford the time to try to help askers find an answer to their question. The busier Stacks seem to rely more heavily on downvotes in contrast, which can be pretty frustrating for the asker since they don't necessarily get to know what the issue was.

The downside of our predisposition is that borderline questions are more likely to be closed on balance, which is final in that it stops answers from accruing and is still jolting to proactive askers. Once a question is closed as a duplicate, the burden of making an argument convincing enough to reopen it usually falls on the asker. But the close voters are more likely to be familiar with that context in the first place, so clearly should have more responsibility for knowing the answer base.

I suggested some form of warning for this specific canonical question because it seems to me there are a lot of related questions and answers in the genre that are within scope and unlikely to be consulted by general advice answers. The way this concern arose was straightforward: When I tried to think of a simple typology from which to construct answers to a canonical question, I had difficulties.

Cag made a compelling case, however, that we have a history over time of receiving essentially identical questions and producing essentially similar general-advice answers. A preface warning's content may seem obvious, but my thought was it represented a reasonable PSA in a likely-to-be-seen place that succinctly highlights this tension to flaggers and voters.

On a particular point:

We do not need more rules and special cases.

I agree with this and do not think the preface is either a rule or a special case. It is not a special case because all duplicate voters/flaggers should make the suggested consideration in every case. It is not a rule in the same sense that almost every moderation decision on Stack doesn't arise from an explicit rule but from some consensus amongst the users who have moderation powers. In particular, none of the diamond moderators suggest they will user "superpowers" to enforce it.

That being said, I wouldn't be bothering with these replies and suggestions if I didn't think they were useful. I hope Stack members will at least give them a thought if they interface with this particular genre of question.

2
  • Just so you know, voting to close is not limited to moderators, and moderators do not get their powers based on their reputation. Aug 14, 2022 at 21:55
  • 1
    @AnonymousPhysicist I do know! But to clarify my terminology which may have been a somewhat unfortunate choice: I'm using "moderator powers" to mean "powers which have the effect of moderating the site's content" broadly rather than "powers exercised by diamond moderators". So another way to put my first sentence is that I don't think I'll ever accrue 3k+ rep.
    – user137975
    Aug 14, 2022 at 23:50
2

The instructions are stating what should be obvious: don't hammer a question for being asked-and-answered unless it has actually been asked and answered.

It is understandably infuriating for the asker when they ask "how does X affect Y" and it gets closed as a duplicate of "how does Y work?", but the linked question does not address X, does not address any generalization of X, and does not explain why we're not addressing X.

Moreover, this is a problem for the site: if the canonical question should address X, but it doesn't, and we close all questions that ask about X, then there's no way for askers to learn about X or for answerers to realize that their expertise on X might be valuable.

I won't go line-by-line through your points, but I want to address the first two:

It is impractical to determine the contents of a long and changing community wiki for each potential duplicate.

Disagree completely. If you are going to close someone's question as a duplicate of another question, it is your responsibility to verify that it is actually a duplicate. It is not good enough to say "well, they're basically the same, it's too much work to make sure it actually contains an answer." No one is forcing you to exercise your close votes or reopen votes, but casting a duplicate vote means telling a human being "your post should be deleted because it's been asked-and-answered elsewhere." If you're going to do this, you should take the time to make sure that you are correct.

Voting to close a question as a duplicate of another question should be based on the content of the question, not the content of answers.

If the two questions are identical, then there is no issue: close away! The instructions don't state otherwise. As discussed above, the issue is where to draw the line when the questions are close but not identical (most often, the new question asks about a specific case of the canonical question).

[By the way, I did not invent these instructions out of the blue, they were specifically requested here].

Update, 6 mos later: I decided to reword these instructions to make them a bit more concise. The "question closure instructions" have now been edited into the main canonical question intro paragraph. This is just wordsmithing and does not imply any change to our closure procedures.

11
  • "if the canonical question should address X, but it doesn't, and we close all questions that ask about X, then there's no way for askers to learn about X or for answerers to realize that their expertise on X might be valuable." That's just not true. Answers can read the closed questions and respond to the canonical question. That's how it's been working for years. Aug 13, 2022 at 22:15
  • "it is your responsibility to verify that it is actually a duplicate." That only applies to the contents of the question, not the answers. Aug 13, 2022 at 22:16
  • "If the two questions are identical, then there is no issue: close away! The instructions don't state otherwise." The instructions you posted on that question do say otherwise with the phrasing "answer actually answers OP's question". That's one of my objections. Aug 13, 2022 at 22:18
  • 2
    (1) This is how it hasn't been working for years; OPs are frequently frustrated when their questions are closed, and info is rarely added to canonical answers. (2&3): If the questions are truly identical, then it follows logically that the canonical answer will truly answer OP's question. The alternative is assuming that the canonical answer fails to answer the canonical question, which is just nonsensical.
    – cag51 Mod
    Aug 13, 2022 at 23:22
  • We don't have canonical answers. Only canonical questions. Those questions can have multiple answers. A canonical answer would defeat the whole point of the voting system. Aug 13, 2022 at 23:24
  • Here is an obvious example of a question I would consider "canonical" that has lots of answers. academia.stackexchange.com/questions/2219/… Aug 13, 2022 at 23:27
  • 2
    Of course we have canonical answers -- they're the ones where we have one big community wiki and ask people to edit rather than post competing answers. It is true that not every canonical question uses this format (and some questions are just popular but not canonical).
    – cag51 Mod
    Aug 13, 2022 at 23:34
  • I disapprove of the idea that the existence of a community wiki implies alternative answers should not be posted. Wikis are for evolving answers, which is very different from "canonical." meta.stackexchange.com/questions/11740/… Aug 14, 2022 at 1:14
  • I disagree that 2 identical questions will have identical answers. Reasons being: not every question is black and white and depending on who is answering, the shades of grey might differ - especially if alot of time has passed between the two identical answers and the answerers and the circumstances might have shifted.
    – Sursula
    Aug 15, 2022 at 9:03
  • Also: by closing a question as a duplicate, I explicitly do NOT say "Your post it not valuable", I rather say "Your question has merit and should be on the site, but it has already been asked and answered." Questions that have no value are simply closed.
    – Sursula
    Aug 15, 2022 at 9:05
  • 3
    Good point, I edited that line. I’m just trying to emphasize that when someone takes the time to write a valid question, we should be careful before closing it. I’m baffled by the suggestion here (made by others) that the system is working as it should when we tell people their valid question is a duplicate of a different question, the answers to which do not address what OP asked about.
    – cag51 Mod
    Aug 15, 2022 at 14:30
-3

Here's a possible alternative wording:

Note to close voters: please consider this meta discussion before closing a question as a duplicate.

1
  • This page is already linked from the question text; moreover, it does not provide any guidance about which questions should be closed as a duplicate.
    – cag51 Mod
    Aug 13, 2022 at 23:31
-5

These instructions should be removed because:

  • It is impractical to determine the contents of a long and changing community wiki for each potential duplicate.
  • Voting to close a question as a duplicate of another question should be based on the content of the question, not the content of answers.
  • Merging answers is impractical and undermines the answer voting process.
  • Redoing close voting after merging answers is also impractical.
  • If it is unclear if a question is a duplicate, the meta site provides a good mechanism for resolving any uncertainty.
  • We do not need more rules and special cases.
2
  • 1
    "Voting to close a question as a duplicate of another question should be based on the content of the question, not the content of answers." - this is not how duplicate closure (is intended to) function (s) on Stack Exchange. The duplicate closure reason and notice even state this - the question has been closed because it is already answered elsewhere, moreover duplicate closure is not possible if the target does not have an accepted or upvoted answer.
    – Nij
    Aug 19, 2022 at 7:43
  • Duplicate, unanswered questions are useless. Close them. Aug 20, 2022 at 3:23

You must log in to answer this question.

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