I have a question which originates on previous assumptions and informations. I formulate this question in a way that contains (from my point of view) enough background information for someone to see its intent. However, there are answers/comments incoming which suggest that the people answering are missing background information or have misunderstood something. Those answers/comments seem to make it necessary to improve the initial question to either mitigate misunderstanding or include more background information

What tools/ways does SE provide to handle such a situation in an agile way?

I have tried:

  • Editing the question to include more background information.

    This is bad for complex questions. I want to have the question-answer pair to be easy digestible for someone having the same question as me.

  • Splitting the question into different sub questions.

    Bad as well. Then the sub questions are possibly answered but I will be the one answering my own complex question. This reduces credibility.

  • Clarifying those missing background informations or misunderstandings in comments

    Works to some extent. Issue is that most people seem to not read the comments on answers, where the explanation is.

Basically I am looking for a way to handle a complex question like a program in SVN.

  1. State it (write the initial code)
  2. See that it's too complex for SE (some functions are too complex and need special handling)
  3. Branch the question (branch the source code)
  4. Merge and consolidate

The goal at the end is that there is a distinct question-answer pair which is super easy to digest for someone having the same question.

Any recommendations?

  • 14
    Yes, use chat or prepare off-line. SE isn't well suited for collaboration on content before it is considered to be of enough quality to be posted on a site. – rene Mar 7 '20 at 21:17
  • @rene Thank you for your confirmation. I already felt like it might not be possible. Would you be able to suggest any site which allows interdisciplinary open source collaboration in the way i described? – user718637 Mar 7 '20 at 21:24
  • 5
    I have no idea but a lot of users that complain about SE find their home on reddit, Quora or a friendly discord server. – rene Mar 7 '20 at 21:27
  • 8
    I think asking on the local Meta site is a good approach as well - if you anticipate problems, or have difficulty formulating the question, you can ask how to ask that question in the local Meta site. – V2Blast Mar 7 '20 at 21:29

Maybe the real answer is: if it takes you hours to write up the question, then it is probably too complicated.

Earlier on, there was the close reason of "too localized, implying that a question might not be in general interest.

Meaning: if your problem is big, you have to dissect it. If that isn't possible, then maybe an online community isn't a good fit.

You see, if the answer would boil down to writing a whole short bachelor thesis (because problem so big), then chances are that neither the question nor the answer are a good fit for a place like this.

  • Actually i think a crowd based solution (online community) is perfect for complex questions. If the discussion and the thought chains of all contributing parties are easily trace and trackable, such that anyone joining the conversation can just look into answers which he immadiately has when joining ("Basic" questions) and has a easily digestibe metric to decide which answer is the right one, high quality solutions are created. – user718637 Mar 8 '20 at 18:32
  • Why? Complex questions are, as said, often very specialized. Most likely very few people actually care about the exact same "Themenkomplex" ("cluster of topics") that a complex question is framed around. Or how many people are out there who worked on the exact same topic for their bachelor/master/phd thesis?! – GhostCat Mar 8 '20 at 19:10
  • That they are very specialized is the actual reason why they are good for crowd based solutions. If you have a spezialized Question and you would be able to ask it to all of mankind, the propability of finding someone who already has a solution to it and can answer it immediately rises immensely. Of course, effort needs to put into ways to highlight those Expert answers and filter out answers of People who haven't completely understood the question, or the Topics thereof. – user718637 Mar 8 '20 at 19:24
  • That is an opinion, not a fact. Sure, the probability to receive helpful answers depends on the number of people having access to your content. On the other hand, the more energy you spend, the higher the chance ... that as said, nobody else really cares about the exact thing you want to talk about. And the other answer is right, on SE, the idea is to have a minimal complete example with the question. There purpose of the communities on this network is to provide answers to questions that matter to a large general public. – GhostCat Mar 8 '20 at 19:31
  • Please do not read that as offensive. Yes you are right my statement basically is an Opinion (Idea), but so is yours. As we can not know which one of the statements is true, we should consider them both as equally likely. And base our knowledge of which one is true solely on measureable evidence and keep our conversation on a professional level. – user718637 Mar 8 '20 at 19:41
  • Basically science: "Hypothesis" -> "Theorie" -> "Experimental Data" -> "Reflection" -> "New Theorie" -> "Experimental Data" -> "Knowledge" – user718637 Mar 8 '20 at 19:42
  • 2
    I hate to argue from a point of authority ... but then: I invite you to compare my SO profile with yours. Check out how many reviews I did, edits I did, votes I did. So unless you have another account going with similar activity ... somehow I think that I know a bit more about how that community works ... compared to someone who is only a bit of active here for a few days. This is not meant to belittle, just about what I wrote elsewhere before: this place has 10+ years of history, and thousands of users talking to each other the whole time ... and all of that tells me: complex questions do .. – GhostCat Mar 8 '20 at 19:47
  • 2
    ... not work here. The essence of this network is to focus on problems that are dissected to their essential core. – GhostCat Mar 8 '20 at 19:48
  • "Complex questions do not work here" How does SE know what is a complex question and what's not? What's the measureable metric for that? As SE itself is complex, aren't by default all questions questioning SE itself complex as well? How can they be possibly discussed in SE if SE is not suitable for complex questions? – user718637 Mar 8 '20 at 20:09
  • All i see and read right now suggests that SE is suitable for questions which are already well understood by mankind and answers to those questions are readily available in other resources. This shouldn't be the goal of SE. – user718637 Mar 8 '20 at 20:15
  • Ah, i might have noticed why we don't understand each other. You wrote "Why? Complex questions are, as said, often very specialized." They are not specialized. We have a different understanding what complex means. – user718637 Mar 8 '20 at 20:16
  • For someone in a specialized field, Questions arriving in this field are not so Complex as for someone being out of this field. Furthermore if two people of the same specialized field talk to each other and one has a question he thinks is Complex, but the other one has already truly understood the Question and Intend, it won't be Complex for him. So percieved Complexity depends on what the People think is Complex and is not necessarily related to specialization. – user718637 Mar 8 '20 at 20:23
  • You probably mean complicated, not complex. Complexity basically means: not accessible to science and engineering practices. – GhostCat Mar 9 '20 at 10:11
  • I haven't thought about the differences between complicated and complex yet. But your definition might be right. Maybe i should discuss the differences between those two words in the Philosophy or Linguistics Subsites. But i have a feeling that the difference between those two is complicated as well and will lead to the same level of misunderstanding i have perceived here. Again, as i got told SE is not suitable for complicated questions. – user718637 Mar 9 '20 at 13:43
  • Well, as a starting point, look into the books by Nassim Taleb, namely "Black Swan" ... and just to be precise: many people consider "his" work on complexity science, but there are also plenty of folks who consider it a bit obscure resp. over-simplified ... – GhostCat Mar 9 '20 at 13:47

In theory, complex or detailed Q&A can be formatted like this -- FAQ for Stack Exchange sites


  • Not on the main site
  • Not on the Meta site except with community endorsement.

Still that is what is supported (all that is supported) by the software --and if you did license and operate your own Teams instance then you could presumably do whatever you liked with it.

The other way to show that various "subquestions" are related e.g. to the same project would be to give them all the same "tag", e.g. foo -- and possibly subtags like foo-ui and foo-db.

If you're talking about the public SE/SO sites, it sounds to me like you may be trying to offload too much onto SE. If I develop a project I expect to keep most of it and most of the (person-to-person) conversations about it elsewhere (not on SO), and only use SO for very narrow/detailed questions.

SO explicitly requires minimal questions -- see How to create a Minimal, Reproducible Example

You must log in to answer this question.