-7

Problem Definition

I see a ton of specific questions getting answered along with non-essential stuff not actually asked. From a data perspective, this creates false mappings where one question is asked and what you really get is an answer to 2 or more questions. This method seems to be rewarded when it conflicts with the instructions for answering a question:

Answer the question

Read the question carefully. What, specifically, is the question asking for? Make sure your answer provides that – OR a viable alternative. The answer can be “don’t do that”, but it should also include “try this instead”. Any answer that gets the asker going in the right direction is helpful, but do try to mention any limitations, assumptions or simplifications in your answer. Brevity is acceptable, but fuller explanations are better.

As seen above, the instructions say to answer the question OR provide a viable alternative. Note however that it does not say to answer the question AND provide a viable alternative. It follows that offering a viable alternative as a "try this instead" should only be done when the question can't be directly answered. While a "fuller explanation is better", it is in reference to a fuller explanation of the question actually asked, not the one that wasn't asked.

An Example

'Question A' is asked and answered, but 'Question B' is answered in 'Answer A' without ever being specifically asked.

Failures of this approach

  1. The poster had an opportunity to expand the question base but failed to do so because they didn't create their own question.
  2. The poster undermines the search feature. While searches obviously search answers, the ideal scenario when searching is to find the exact question you're looking for. You don't want to have to do more work by looking for 'Answer B' in 'Question A' because it points to 'Answer A+B'.
  3. The poster creates clutter that slows or stops a future reader from getting an answer to the actual question.
  4. If another user is searching for 'Answer B' and fails to find it because it's buried in 'Question A', the user is likely to create another question that will wrongfully be seen as a duplicate.
  5. The poster thinks they followed the heuristic that "fuller explanations are better", when they really only provided a fuller explanation to another question.

An Alternative Example

To address the need to present more information regarding a question not specifically asked, this additional information should come in the form of 'Question/Answer B'. Then follow up with a concise answer to 'Question A' along with a link to 'Question B'.

Benefits of the alternative

  1. The poster adds more content to the site.
  2. The new question is now easily findable when searching.
  3. Future readers can find an answer without having to wade through non-essential information.
  4. Duplicate questions should go down since finding the original question would be easier.
  5. Posters are not rewarded for the wrong reason and therefore not encouraged to repeat the same act.
  6. More rep is generated overall.

The Question

Given the above information, how could it still be considered correct to answer a similar question in addition to the question asked, when there is a better alternative?

  • 7
    If an answer to question A incidentally answers a question B, it is not our policy to close question B as a duplicate of A. Do you have specific examples of questions from the site? I'm not sure I understand your concerns. – Raven Dreamer Mar 6 '13 at 19:24
  • 3
    Yes, please, it's obvious you have examples in mind so it'd help us if you could let us see them. – badp Mar 6 '13 at 19:30
  • 1
    I believe the specific concern here is related to an answer on this question: gaming.stackexchange.com/questions/86024/… - 10k's can see it, but the jist is that kraftydevil doesn't feel that the answers on there are as good at answering the question as a short "No, there is not" response. – agent86 Mar 6 '13 at 19:35
  • I remember that answer. I didn't flag it (much to my surprise), but it wasn't very good. Your solution here would outlaw quite a few of my excellent answers that go above and beyond. And that's just for Fire Emblem! – Frank Mar 6 '13 at 19:47
  • I didn't bring this up in my answer, but do you have any evidence that your claimed benefits would take place, or that they're even benefits to begin with? I find all of them either dubious, debatable, or inconsequential. – MBraedley Mar 6 '13 at 22:56
  • @agent86 - my previously deleted answer is not the prime example of answering method that I proposed. It differs in that I failed to offer an alternative via a link. What I should have done is then created the question, "Are there any weapons that have infinite ammo?", answered it, and linked my original response to it. Is that a poor method? – kraftydevil Mar 6 '13 at 23:48
  • 3
    I think that is a poor method, yes. The entire point of the question was to find a weapon for which ammo was not a problem. We do not need a separate question for each variation of this goal. All that does is fragment information and make the site clunkier. When information is this closely related, it should be in the same place. – user9983 Mar 7 '13 at 0:34
  • 1
    @kraftydevil: I would have closed that as dupe. "Is there a gun with infinite ammo?" is effectively the same as "Is there a way to regenerate ammo?" The reason we saw the later is because it was a game mechanic in the original. The reason we didn't see the former is because it wasn't. In that case, context is important. Additionally, the context moves us from the the fact finding as worded to the problem solving as implied: "How can I avoid having to buy ammo?", which is the problem that both fact finding questions are trying to solve. – MBraedley Mar 7 '13 at 2:03
  • 1
    I flagged the original answer, explaining that the information provided was already in the other answers. I tried to explain: As a community, we prefer "No, but try this" over just "No", and I linked to the How to Answer page as a response to the point raised on the question, and again here on meta by kraftydevil, which is where the quoted text above comes from – Robotnik Mar 7 '13 at 5:40
  • @MBraedley: I'll work on gathering evidence. – kraftydevil Mar 8 '13 at 8:28
  • @RavenDreamer Don't remember details, but this has certainly happened with Diablo 3 - reason I recall it at all, is one of the questions merged was not actually a duplicate of what it got merged into, tried flagging it but that did not help either. Though considering its a really rare occurrence, not really worth bothering about imho. – Alok Sep 16 '13 at 20:34
-7

Your alternative isn't better.

  • I like this answer a lot more than the others haha. disagree or not. – kraftydevil Mar 7 '13 at 22:26
  • 1
    (to say any more might be to answer a similar question in addition to the one you asked) – MrGreen Mar 7 '13 at 22:26
  • thank you for actually reading the question. Something most people can't do evidently. – kraftydevil Mar 7 '13 at 22:28
  • 5
    NO! This is a textbook case of a terrible answer that is technically correct. The two of you are completely missing the point that agent86, badp, and myself are trying to make, which is to say it is not sufficient to simply answer a question. Simply answering a question gets you (as a rep earner) and the questioner (as the one with a problem) nowhere! You have to make sound arguments to support your answer, as well as answer the implied questions that the questioner didn't think to ask. – MBraedley Mar 7 '13 at 23:04
  • 1
    @MBraedley: look up... that is the joke, flying over your head. krafty apparently did get it, and un-checked my answer. – MrGreen Mar 7 '13 at 23:11
  • who needs rep anyways? Is that the point of the site or is it to provide information? – kraftydevil Mar 7 '13 at 23:15
  • @MBraedley - Would you prefer your answer to be correct or not? I only unchecked your answer because it appears that association with me and anything I do on this site is subjective and I figured the only humane thing to do would be to disassociate us so your answer wouldn't be downvoted. But now that I hear your answer is intended as a joke, I have to make it the correct answer to justify your intention. – kraftydevil Mar 7 '13 at 23:20
  • 2
    @MrGreen: I didn't take it as a joke because you can't impart tone in a piece of text that's void of any emotion. You're encouraging more bad behaviour with bad behaviour, and that's simply not welcome in meta. If you're going to do that again (don't do that again), then at least make it incredibly obvious. – MBraedley Mar 8 '13 at 0:12
  • the only bad behavior is when others try to prevent people from spreading knowledge, which is what is going on at this site. – kraftydevil Mar 8 '13 at 5:32
  • 3
    @kraftydevil - Isn't that the exact point you're trying to make? That we shouldn't be spreading excess information? – Robotnik Mar 9 '13 at 2:54
  • @MBraedley: Please do not give me orders – MrGreen Mar 11 '13 at 5:16
15

It's OK to answer a question and provide additional information to the asker that they didn't specifically ask for. Actually, I'd consider this to be good form and highly encouraged.

Most people asking questions are people who don't know enough yet about the topic to ask the best question possible. It's kind of a catch-22 in that if you could ask a very specific, very excellent, targeted question, you are likely to either know the answer or be in a situation where you are a good candidate (sometimes, the best) to figure it out. That's not to say it's not possible to ask really good, specific, detailed questions, but the point I'm trying to make is that people who come and ask questions here are often not experts.

Often times a sub-par question elicits a short, to-the-point answer which answers the question as asked, but doesn't provide a solution to the problem the asker is facing. It's possible to ask a better question (see here for my thoughts on that subject), but it's also possible to look past the specific thing being asked - the non-expert question - and find the underlying problem.

When it's possible to provide an answer that answers the question, but also looks past the specifics of the question being asked, and solves the problem that the asker has, I think it is worthwhile to do so. In my opinion, that's going above the call, and is something beneficial to the asker and people in a similar situation.

Questions can be linked together, and you'll see linked questions in the right sidebar whenever this happens. Sometimes part/all of an answer is important to multiple questions, and that's OK too. Even fully duplicated questions are not deleted - they are kept so that everyone can benefit from an additional way of searching for a given question.

While it would sometimes be nice to have a cohesive, formal, organizational structure (like wikis do) where everything is nicely cataloged and deduplicated, this site's focus is on solving people's problems, whatever form they may take while on the site. That often leads to situations where we sacrifice the organizational needs in order to be better at the "customer service" aspects.

9

No.

Making it a policy that questions need to answer what the question text says, exactly what the question text says and nothing else is risky at best and counterproductive at worst.

I assume your example question is Are there guns which regenerate ammo?, where your answer was deleted. It simply read:

There are currently no known guns that replenish ammo.

But what if the answer was positive? Would you consider this a full answer?

There are a few guns that replenish ammo.

After all, the asker in question A didn't ask which guns replenished ammo. They just wanted to know if there were some, right? Clearly the list of guns that do replenish ammo is best left to a new question with a more accurate title. We'd get:

  • more content (???) on the site!
  • a new question with a title that more closely resembles the answer (why don't edit the original question to begin with?)
  • Future readers can find an answer without having to wade through non-essential information (too bad the essential information is absolutely useless)
  • Duplicate questions should go down since finding the original question would be easier. (Actually, the more duplicate questions we have, the merrier!)
  • Posters are not rewarded for the wrong reason and therefore not encouraged to repeat the same act. (Wrong, they're just rewarded once instead of thrice)
  • More rep is generated overall. (??? and this is good how ???)

This is a basic example of why going above and beyond what the asker asked in your answer is generally a good idea. Your proposal would generate a really unfriendly and hostile community where you'd read people's questions as strictly and unhelpfully as possible in order to make up and self-answer your own question B and grab reputation from three posts instead of one.

Thanks, but no, thanks.


The answer can be “don’t do that”, but it should also include “try this instead”. Any answer that gets the asker going in the right direction is helpful, but do try to mention any limitations, assumptions or simplifications in your answer.

It follows that offering a viable alternative as a "try this instead" should only be done when the question can't be directly answered.

No, you are misreading this sentence - or this sentence may be written unclearly. It is intended to reference this meta discussion. The idea is that if you ask "What is the best way to open a door, smash through it or unbolt it?" the answer shouldn't be "unbolt it", it should be "don't do that, use the doorknob instead, if the door does have one."

2

Here at Arqade, we have two primary types of questions that we allow: Fact Finding and Problem Solving, as well as a large number of questions that fit into both categories. For the later, our primary motivation in answering is to solve a problem! If we don't provide the full solution, which could very well exceed the scope of the original question, then how is the question asker going to feel?

"Well here's the solution to the problem you posed, but after you complete these steps, you're still going to have problems, and I'm not going to tell you how to solve those problems until you ask another question because I'm a jerk and a rep whore." Not only is that condescending and inconsiderate, but it also leaves the user that asked the original question frustrated and aggravated and ultimately unfulfilled. This is not a good situation!

So to address your listed failures:

  1. The answerer can expand the question if appropriate through editing. You are unable at this time to freely edit posts, but you can still suggest edits that reasonably expand the scope of a question. Some times these will be rejected, sometimes they won't, that's on a case by case basis.
  2. WRONG! Search is not undermined, primarily because search results show whether a post is a question or an answer. You don't search for the exact question, you search for keywords.
  3. I disagree with this, especially in the context of problem solving. A user reading the answer later should still be able to parse the necessary tidbits that will help them, and may in fact solve problems the reader didn't even know they had.
  4. See 2 above. Also, a new question may or may not be seen as a dupe, which in and of itself could be good or bad. If the original question was too narrow in scope, having a second question that essentially points this out is a Good Thing (TM) and will give experienced users the chance to fix it.
  5. More info = good thing. 'Nough said.

If you think that an answer is providing too much information, you're probably wrong, and really it's the question that is too narrow in scope. This isn't always true, but consider it to be a good starting point. If the answer is really about two completely different questions, then yes, we probably have a problem, but I rarely see that happen in practice.

  • The situation you provided in italics is not what I'm talking about. By your response I know that I wasn't very clear, but I did mention that providing a link to an alternative was a good practice. I would like you to consider that it might be a good thing under the premise that it teaches people how to ask the right questions -- by giving them a solution along with the right question. That's got to be more valuable and than worrying about repercussions of the rep gained by doing so. What difference does it make if someone gets more rep because they took the time to do more work? – kraftydevil Mar 8 '13 at 8:12
  • Editing the question sounds like a convenient way of telling someone what their question is so you can make up your own answer. Furthermore, this is after you told me it was ok to provide an answer to which they didn't ask. Making up answers and questions should really be your concern about rep. 2) We should think bigger than what you have described. We should all be working toward the day when we can find the exact question we want to know the answer for. Ubiquity. – kraftydevil Mar 8 '13 at 8:14
  • 3.1) This site is not about problem solving or learning how to problem solve. If it was, then the question asker would be required to demonstrate their newly given understanding. If it was, the question answerer would be required to fill out a more complex answer form that took the shape of steps. 3.2) In what science is it considered optimal to embed the essential information in other information? The user may be able to parse later, but at the cost of speed and the possibility of not finding their answer buried among others. – kraftydevil Mar 8 '13 at 8:14
  • 4) Right, it may or may not be a duplicate. When you say experienced users would have a chance to fix the first question after a second clarifying question has been posted, then it follows that the second question will then become a duplicate, effectively penalizing the person who originally took the time to outline a similar but different problem. – kraftydevil Mar 8 '13 at 8:15
  • 5) I don't really know what you mean by a "Good Thing". A "Good Thing" is what's optimal. The least amount of information required plus a quick link or links to viable alternatives is optimal. I define 'optimal' as whatever leads question askers to the answer faster. This applies to at least two classes of question askers that I can identify. Question askers who write questions (some because they can't find an answer at all or some can't find an answer that satisfies them) and question askers who search for questions. How do you define a "Good Thing"? – kraftydevil Mar 8 '13 at 8:15

You must log in to answer this question.

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