If an off topic gets an answer before closing, logic suggests that whatever answers may have been submitted must be off topic as well. The possibility of making a post that has value to the community, which responded to a topic not suited to WB:SE is a paradox. My premise is simple: A post that is not suitable for this site can not create answers which both answer the question and are suitable to this site.

So there is a dilemma concerning improving closed questions and invalidating answers, as answers to invalid questions must be invalid at the outset:

  1. A question that is invalid has an OP intent that is invalid (is it possible to assume an OP wrote what they do not intend?)
  2. Changing an off topic closed question into a valid question will in all cases change the OP intent (you have changed the question topic)
  3. Editing an off topic closed question while maintaining the OP intent, can NOT change the question topic; therefore it can never be opened.
  4. Only by changing the topic of an off topic question can you open it, but to change the topic of any question invalidates answers to that question.

Q: Can the rule against invalidating answers create questions which can not be opened by any means?

If this is the case, this seems like a simple criteria to find candidates for deletion. There is no logical reason to consider the validity of answers to off-topic questions. By definition, all answers to off topic questions are “not an answer.”

  • 1
    $\begingroup$ Are those questions yours? If not, why do you care one way or the other? We are not employees of Stack Exchange Inc. to be concerned with achieving specific KPIs with respect to the good order and maintenance of the catalog of questions. That the results of our collective work are a bit messy is just a reflection of the less than well-structured composition of the community. $\endgroup$
    – AlexP
    Commented Mar 6, 2022 at 9:03
  • $\begingroup$ @AlexP …So, “yes.”. $\endgroup$
    – Vogon Poet
    Commented Mar 6, 2022 at 13:57
  • $\begingroup$ "There is no logical reason to consider the validity of answers to off-topic questions" why would that be ? For instance worldbuilding.stackexchange.com/questions/225304/… this +10 question now has 3 different close reasons. It has 7 answers with a clear preference for LDutch answer.. Which of the close reasons invalidates all 7 answers ? $\endgroup$
    – Goodies
    Commented Mar 8, 2022 at 19:41
  • $\begingroup$ @Goodies -- I think that's a fatal flaw of the argument: no close reason ever invalidates an answer, as they are statements on the viability of the query, not the response. That was a good question, by the way! $\endgroup$
    – elemtilas
    Commented Mar 9, 2022 at 11:41
  • $\begingroup$ @elemtilas if it is true that any response has its value because it solves a problem, then every response that solves a problem unrelated to world building has no value here. Only when a response has value by its own merits does preserving that response make any sense. The 'no-invalidation' exists because responses derive their value from the query. Necessarily; a valuable response to a value-less query is impossible, and should not be "protected." Worse still, "off topic" is not a close reason. "This question doesn’t meet a Worldbuilding Stack Exchange guideline" is. $\endgroup$
    – Vogon Poet
    Commented Mar 9, 2022 at 15:19
  • $\begingroup$ As I said: responses have value based on their own merits. Their status is, in many respects, independent of the status of the question. This is why we see downvoted or closed queries with accepted and highly voted responses. One upvote on an answer means that someone found it useful. It can be argued that an answer with no votes at all means that no one found it actually useless. "No-invalidation" exists simply as a courtesy to respondents who have put in effort to write responses! The custom doesn't mean a question can never be edited or changed in any way; only that the (cont) $\endgroup$
    – elemtilas
    Commented Mar 9, 2022 at 15:51
  • $\begingroup$ (cont) substance and the topic can't be changed. If you ask about a horde of orcs attacking a modern military base, and someone answers that query, the OP can't go back and edit the query to a horde of orcs attacking a medieval city. The answer derives its nature & structure from the query, but otherwise provides its own value and stands on its own. It's incorrect that a valuable response can not exist in the context of a value-less query. We see examples all the time: highly downvoted or closed queries with upvoted and even accepted answers. "Off topic" is literally the verbiage (cont) $\endgroup$
    – elemtilas
    Commented Mar 9, 2022 at 16:00
  • 1
    $\begingroup$ (cont) SE uses in the alert box: Closed. This question is off-topic. It is not currently accepting answers. The query is "off topic" for whatever reasons were chosen. The description of what closure is strongly infers "off topicality" as well: Closing is a democratic voting process where the community identifies questions that duplicate existing content, are unreasonable to answer in their current state, or do not belong on the site. This line of reasoning is getting a little off track. In any event, it's yourself who couched this question in terms of "on topic" and "off topic". $\endgroup$
    – elemtilas
    Commented Mar 9, 2022 at 16:03
  • $\begingroup$ @elemtilas there are a plurality of "Closed" boxes, not all of them result in "off-topic." The concept of "off-topic" is absolutely not from my couch. This question does specifically relate to that concept however. $\endgroup$
    – Vogon Poet
    Commented Mar 9, 2022 at 17:48
  • $\begingroup$ I've put a critical comment, but I now realize my remark was about something else. I'll +1 this question. Very frustrating for the opener to find an edit embargo! An embargo because of existing answers makes it even worse. I'd say always allow edits to get a topic reopened. Wrong answers go down anyway, so a smart answerer will remove it, or edit the answer text. Let it be responsability of the answerer, Answerers don't need protection.. When you're being downvoted, you'll notice soon enough.. you can always check the question text and reconsider. $\endgroup$
    – Goodies
    Commented Mar 9, 2022 at 18:34
  • $\begingroup$ The not-matching-answers edit rule is not always observed or maintained. Yesterday we had a funny example of a question edit, requiring answer editing.. someone put a question in meters and feet mixed. There was a remark about the mix, opener changed meters into feet to have consistent metrics, without changing the amount. There were 2 answers, both assuming meters and the edit was allowed... I adjusted, Willk did not return yet to modify his.. so you see 2 inconsistent answers worldbuilding.stackexchange.com/questions/225349/… $\endgroup$
    – Goodies
    Commented Mar 9, 2022 at 18:47
  • $\begingroup$ @Goodies The DNETIR rule is not recognized by SE (you get no reputation or badge or incentive to do it), it is a community-implemented construct. Changing numbers is considered against the rule $\endgroup$
    – Vogon Poet
    Commented Mar 9, 2022 at 20:37
  • 1
    $\begingroup$ There's nothing wrong with it being a "community-implemented construct". That's how WB functions, by implementing our own rules and respecting our own customs. The answer you link to, and the other question that response links to both speak well to the sensibility of not editing a question once it has answers, and how to deal with it when it happens. $\endgroup$
    – elemtilas
    Commented Mar 9, 2022 at 23:42
  • 1
    $\begingroup$ @elemtilas these never ending issues are not because community self-policing is bad, but because it can be done badly by failing to be implemented consistent with SE policies. Policies we create via META become entrenched when their topic question is pinned and upvoted hundreds of times, so changing community-based policy is much harder than a SE policy. Ergo, we never stop moderating them & they create a vast workload. $\endgroup$
    – Vogon Poet
    Commented Mar 10, 2022 at 0:07
  • $\begingroup$ I never said community moderation was bad. Nor do I find the way this forum operates dysfunctional or problematic. I do agree with you that community moderation can be done badly. $\endgroup$
    – elemtilas
    Commented Mar 10, 2022 at 14:21

2 Answers 2


Yes. And No.

That is all.

Strictly speaking, the narrow answer to the question is Yes, the rule against invalidating answers can, but does not always, create questions which can not be opened by any means.

This is one reason why we respondents have a grave responsibility to thoroughly read and understand a question before answering it. We should always discuss clarifications and air issues with the querent before answering.

We often fail our querents in this regard, and I am every bit as guilty as the next person of answering a question that oughtn't be answered.

The failure of my team leads to exactly the problems you mention: questions that can not be improved (edited) without invalidating the answers. This was the topic of my recent Meta question about reminding eager respondents such as myself not to answer questions that really pretty obviously have some kind of problem that needs to be addressed.

Resolution: While not every question that has been closed after an answer is given results in a question that can not be edited; the optimally less than perfect way to solve the problem is simple: learn from the disaster; write a new question that ties the loose ends; move on with the story.

Other Issues & Dilemmas in Order of Appearance:

Not all answers yield paradoxes. I'd argue your analysis is too polar. Just because someone writes a really bad question (poorly conceived, off topic, whatever) doesn't mean the answers will of necessity be bad as well! This is a false conclusion as these answers can still be useful even if the question is not. You can see this all over the forum with questions that are closed or downvoted to oblivion yet have answers that are highly upvoted and often having earned the Green Checkmark seal of approval.

The reasoning here is simple: answers are not the same thing as questions and, as regards their merits, are not as dependent on the question as you assume. You as querent write your question and it is judged on its own merits as to whether it is on or off topic, well or poorly conceived, how well it fits with SE rules and forum customs. I as respondent write my answer and it is judged on its own merits as to whether it a) answers the specific question and / or b) is deemed to be generally good (useful, entertaining, comprehensive, instructional, educational, etc) by others.

Issue 1: A question that is invalid has an OP intent that is invalid (is it possible to assume an OP wrote what they do not intend?) --- This is possible of course! This is why there is a comments section, so these issues can be brought up and resolved. Preferably before answers start pouring in.

Issue 2: Changing an off topic closed question into a valid question will in all cases change the OP intent (you have changed the question topic) --- Not true. Questions can be "off topic" for a variety of reasons. If it's off topic for lacking focus, then focusing the question does not keep it off topic and will not likely change the OP's intent.

Issue 3: Editing an off topic closed question while maintaining the OP intent, can NOT change the question topic; therefore it can never be opened. --- Convoluted and half true. Again, if a question is off topic for lacking focus, edits will (or should) not change the topic. Thus the question can be edited and reopened. If the topic is changed, the question should not be reopened, the edit should either be reverted or modified and the OP should be encouraged to write a new question.

Issue 4: Only by changing the topic of an off topic question can you open it, but to change the topic of any question invalidates answers to that question. --- Untrue. Again, there are many reasons for a question to be closed as off topic. Questions can and regularly are edited and reopened without change of topic. Answers may or may not be invalidated by a change of topic, but best practice assumes they will be and thus the discouragement of rewriting a question via the editing process. This is why, once a poorly written or off topic question gets answers, we encourage the OP to just write a new question.

Issue 5: If this is the case, this seems like a simple criteria to find candidates for deletion. There is no logical reason to consider the validity of answers to off-topic questions. By definition, all answers to off topic questions are “not an answer.” --- Generally, we only close questions that are deemed off topic. Deletion I think is reserved for spam and so forth. Hence "Vote to Close" rather than "Vote to Delete".

The second part is simply untrue: An answer's validity stems from two sources. First, of course, is the question. The other is the answer itself. Answers can be both "not an answer" and also "an answer" simultaneously. In practice, "not an answer" is generally reserved for posts that are really more suited as comments, or that ask other questions or that simply have nothing to do with anything.

Conclusion: while the answer to this question is a qualified yes; the logical conclusion to be drawn from the situation is never to delete answers or questions. In WB, we judge questions by one set of criteria and we judge answers by a completely different set of criteria. This means that a question can be closed, but answers can not. Answers can be broadly useful, and even answer a bad question that is now closed, and yet itself not be an "off topic" answer.

The best options are:

  1. Avoid this situation in the first place by writing a solid query (onus is on the querent) and by reading the question thoroughly for understanding and addressing any issues (onus is on the respondent);
  2. Rely on your intuition and VTC a question that looks like it could be a) a good question a/o b) looks like it could be problematic; explain your reasons for closing and offer the OP some assistance, especially newer ones;
  3. Bring a shovel so you can cover up the mess -- on occasion, a querent might just have to give up on a question the rest of us loused up and start over again.
  • $\begingroup$ Am I correct in assuming that the “duplicate question” close reason can not attach your query to a closed query? This would multiply the flaw and create vast seas of unaskable on-topic question. $\endgroup$
    – Vogon Poet
    Commented Mar 6, 2022 at 17:52
  • $\begingroup$ I miss the dilemma here. "and I am every bit as guilty as the next person of answering a question that oughtn't be answered." We all have the privilege of deleting our own answers, for which sacrifice I would humbly award the "model citizen" badge of honor. $\endgroup$
    – Vogon Poet
    Commented Mar 6, 2022 at 18:07
  • $\begingroup$ @VogonPoet -- 1) I'm sorry: what do you mean by "can not attach your query to a closed query"? If you mean question merging, then I know it's possible for queries to be merged. I think our moderators don't do this enough; and frankly I think we as a community don't encourage them enough to clear up duplicates in this way. $\endgroup$
    – elemtilas
    Commented Mar 6, 2022 at 18:36
  • $\begingroup$ @VogonPoet -- 2) Oh, I have deleted answers in the past, so you can send the badges my way! I also accept cookies. I'm not one to suffer from a dilemma from this situation, though. If I wrote what I consider to be a good answer and it's well received by the community regardless of the reception of the question, I am under no obligation to delete it. The reasoning is simply because the utility or fitness of my answer does lie solely upon the foundation of the question. $\endgroup$
    – elemtilas
    Commented Mar 6, 2022 at 18:40
  • $\begingroup$ No, I mean does the site allow members to "close as duplicate" of a closed question. Not merging questions. $\endgroup$
    – Vogon Poet
    Commented Mar 6, 2022 at 18:40
  • $\begingroup$ @VogonPoet -- I see. Yes, we can close a question for being a duplicate of a closed question. I've seen many examples of this. Sometimes you can even a short chain of these. I know I've seen several chains of three in the past; and I've also seen webs where a query can be closed as duplicate of both a closed and an open question and may be answerable by one or more other queries. This situation will only get worse as time passes. (cont) $\endgroup$
    – elemtilas
    Commented Mar 6, 2022 at 18:44
  • $\begingroup$ The site may be enabling bad answers by policy. Does a member suffer reputation loss when their up-voted answer gets deleted? That would be bad. As a whole, deciding suitability based on vote count (democracy) will fit poorly with any site of limited scope. You can not pretend to control content and also hand the content controls to a fair and equal democratic vote. Well, I'm sure I'm the first to think of this. :P $\endgroup$
    – Vogon Poet
    Commented Mar 6, 2022 at 18:45
  • $\begingroup$ (concl) -- The best solution, of course, is for the mods to merge all these questions into one. The easy solution is to leave them be, and practically speaking, this is not necessarily a bad way to go as none of us are actually paid to do any of this cleanup work! $\endgroup$
    – elemtilas
    Commented Mar 6, 2022 at 18:46
  • $\begingroup$ @VogonPoet -- I concur! First of all, SE is not designed to handle our kinds of Q&A. By definition, most of our queries are SE-off-topic for being opinion based. But mostly, it's our own laziness as a community that enables bad questions or answers. Answers are only deleted for being spammy or extremely low quality. An upvoted answer should not be deleted by anyone other than the respondent. I think the respondent would lose the corresponding points, but this is not a problem after one has about 250 points or so. We both have thousands of points. I wouldn't even notice if we lost 100 points! $\endgroup$
    – elemtilas
    Commented Mar 6, 2022 at 18:53
  • $\begingroup$ @VogonPoet -- On "suitability", any kind of vote or point system is both a compromise and to some extent a failure to reveal anything useful. An answer that was upvoted years ago may now be out of date, and my new answer that provides better information may never receive more than one or two votes and probably won't become "accepted". That's just the way these kinds of forums work. I'm just glad that we have vote transparency. If you go over to Medium, the only response you can give is a "clap", basically an upvote. Lots of forums support the false notion of self esteem by eliminating (cont) $\endgroup$
    – elemtilas
    Commented Mar 6, 2022 at 18:58
  • $\begingroup$ (cont) "negative" feedback. As far as "fair and equal democratic vote" goes, I don't even pretend we have that here. There are probably 20 or so very active everyday members and maybe another 20 or so frequently active members. I'm pretty sure the vast majority of the comments and Meta questions are generated by these Elite Two Score and Two, and I'm pretty sure most of the voting comes from these active people as well. WB is, and has always been, an oligarchy of the active. $\endgroup$
    – elemtilas
    Commented Mar 6, 2022 at 19:03
  • $\begingroup$ The premise that “off topic for lacking focus” seems a paradox as these are discreet measures of fitness. One can violate both but to be objectively off-topic requires a topic that is known. Lack of focus includes the set of questions which “may contain on topic content” $\endgroup$
    – Vogon Poet
    Commented Mar 8, 2022 at 20:24
  • $\begingroup$ contd: The paradox arises from a couple policies however chiefly, reviewers have no mentoring and no checks and balances. That leads to many fuzzy issues, but re: this issue, it means an arbitrary set of reviewers will see some off-topic content and VTC (leading to the case you describe), and another set who use the Off-topic vote only when none of the query is on topic. A focus VTC seems to incorporate the former case and should dominate a query with these two issues. $\endgroup$
    – Vogon Poet
    Commented Mar 8, 2022 at 20:50
  • $\begingroup$ @VogonPoet -- Lack of formal mentoring is, either by design or by happenstance, the nature of the online world. For all its problems, SE is one of the best online communities going. I'd argue that that is because the community decides what the community will accept and informally mentors newer members. As for "off topic for lack of focus", I think that's more a function of question coherence and narrowness of scope more than it is of nature of content. A content appropriate question can be "off topic" when not focused; a content inappropriate question is always "off topic" even if it is (cont) $\endgroup$
    – elemtilas
    Commented Mar 8, 2022 at 21:52
  • $\begingroup$ (cont) perfectly composed and narrowly focused. As for paradoxes, that's just a design feature of this platform. A trade off if you will. Community moderation by "experts" who each have their own sets of objective and subjective measuring sticks is bound to lead to fuzzy. I don't see the situation as paradoxical, but it certainly can be messy at times! VTC rationales often suffer from the application of these varying measuring sticks. But again, that's part of the design: the community can decide to close, and the community can decide to reopen. And maybe close again! $\endgroup$
    – elemtilas
    Commented Mar 8, 2022 at 21:56

The rule against invalidating answers is not a Stack Exchange rule

I'm not suggesting you didn't understand that. I'm simply using it as a lead-in. It's important to remember because Stack Exchange encourages people to improve questions through editing:

Editing is important for keeping questions and answers clear, relevant, and up-to-date. If you are not comfortable with the idea of your contributions being collaboratively edited by other trusted users, this may not be the site for you.

Any time you see a post that needs improvement and are inclined to suggest an edit, you are welcome to do so.

It helps to remember that the page I just linked was developed long before Worldbuilding existed and at a time when most users of Stack Exchange were using it for incredibly objective purposes (e.g., asking questions about programming). Unfortunately, Worldbuilding is imaginative and creative enough that arbitrary editing for reasons other than formatting, spelling, and grammar is pretty much verboten.

So, why did the don't-invalidate-answers rule come to be?

A lot of questions have been asked about or referencing the issue of invalidating answers. But one of the answers to the earliest question about it makes a valuable point:

Once a question has an answer that is considered valuable by either the community (through upvotes) or the asker (through acceptance), no one, including the asker, ought to edit the question in a way that changes its meaning sufficiently to invalidate the existing answer. If people want to get answers to a different question, they should ask that question separately.

Otherwise, the edit invalidates valuable content and/or requires more work from answerers or the community to update the answers to keep up with the updated question.

In cases where a question is ambiguous, such that there are multiple possible interpretations of what it is asking for, it should be closed until it's edited to be sufficiently precise. The earlier this happens in the question's life-cycle, the better. If the asker notices that there are answers that either don't address the asker's intent or assume multiple interpretations of that intent, the asker should edit the question as soon as possible to be more precise and should comment to the answerers accordingly.

That quote actually came from another Stack, Mi Yodeya, and it's the best explanation I know of. Your issue, therefore, boils down to this:

  • If a question remains open, then any edit to the question that invalidates answers is verboten. Think of it this way: the question isn't so bad that it has been closed, therefore all the answers were provided in good faith. It's requisite that the OP and any future question editors also keep that faith, even if the posted answers lead them to realize they didn't ask the question they thought they did.

  • If a question has been closed, all answers are irrelevant. This is because there cannot be a "valid answer"1 to a closed question. Any argument that editing the closed question cannot invalidate the answers is simply an argument to keep the question closed because at least six people can't agree to the credibility of the question sufficiently to ascertain the validity of the answers.

Which brings us to a reality of Stack Exchange that simply exists and cannot be materially changed

There is an "unwritten rule" that simply exists: caveat responsor (let the respondent beware).

There are people who will answer anything as quickly as they can.

  • Some simply enjoy the Q&A format.
  • Some are rep hunting.
  • Some haven't learned the rules/etiquette of the site.
  • Some aren't paying attention.
  • Some don't care.
  • etc.

But the simple truth of the matter is caveat responsor. A person who answers a badly asked question can't complain when the repairs to that question invalidate their answer. The question was closed because a mod, a qualified user, or five peers sincerely believe the question does not meet one or more of the rules of this Stack and/or Stack Exchange and thus is not a valid question.


The don't-invalidate-answers "rule" only applies to open questions. It does not apply to closed questions.

Any other exercise in logic to better understand or regulate the "rule" is straining at a gnat.

Which means the simple answer to your question is "no."

1There is a HUGE difference between a "valid answer" and a "useful answer." An answer posted before a question was closed may be useful to somebody — but it's invalid by definition because the question was closed.

  • $\begingroup$ This question exists because the invalidation rule is enforced on closed questions. The canned moderator message you will get after editing a closed question states, "Doing this edit had the effect of invalidating the answers already posted, and required that a moderator roll back your changes." Moderators are required to roll back edits which invalidate answers to any question. Even closed questions. Also, an answer's utility derives from either the question or the votes; it can not be both. $\endgroup$
    – Vogon Poet
    Commented Mar 9, 2022 at 14:20
  • $\begingroup$ You point out the paradox plainly with @MiYodeya 's post. "a question has an answer that is considered valuable by either the community (through upvotes) or the asker (through acceptance), no one... ought to... invalidate the existing answer... [because] the edit invalidates valuable content." This logic is convoluted. It means an answer's value is simultaneously communal (per blind votes), proprietary (by acceptance), and contingent (upon the question). Any contingent value exists only while the question has value (is open). Pick only one mode to assign value in the WB community. $\endgroup$
    – Vogon Poet
    Commented Mar 9, 2022 at 14:37
  • $\begingroup$ @VogonPoet In response to your first comment - I do not believe the invalidation rule is applicable to closed questions (just because people do it doesn't mean that's how it should be done). In response to your second comment - you're quoting a statement about open questions to support your concern about closed questions. That statement from Mi Yodeya is dead-on logical. $\endgroup$
    – JBH
    Commented Mar 14, 2022 at 4:45

You must log in to answer this question.

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