21

The help centre’s What types of questions should I avoid asking? is linked in many prominent places guiding users on how to ask questions. However, with respect to too broad question contains only the following line:

Your questions should be reasonably scoped. If you can imagine an entire book that answers your question, you’re asking too much.

This fails to capture the case that a question asks multiple questions (that are sufficiently independent to be asked separately) — which is also bad and explicitly closeworthy but does not necessarily mean that an entire book is needed to answer the questions.

As this is a common problem, I propose that a respective sentence is added to address this issue, e.g., something along the lines of:

Also do not ask multiple questions at once unless you expect their answers to overlap considerably.

  • 2
    This largely fits with meta.stackexchange.com/a/299893/215590 – PolyGeo Sep 2 '17 at 19:52
  • Isn't the too broad close reason enough for this? – NVZ Sep 3 '17 at 3:07
  • 3
    @NVZ Too Broad is used after the event whereas help/don't-ask is pre-warning potential askers about what is expected before they decide to ask. – PolyGeo Sep 3 '17 at 4:53
  • Ah, prevention is better than cure. I see. Voting it up. – NVZ Sep 3 '17 at 4:55
  • 3
    Possible duplicate of Add the "one question per post" rule to the Help center – Rory Alsop Sep 11 '17 at 19:02
  • 1
    @RoryAlsop: This one has a better answer and a higher score, so I deliberately voted them in reverse chronological order. – Nathan Tuggy Sep 11 '17 at 20:44
7

The wording that we use in the Tour of the GIS Stack Exchange is:

Your most important question is important to us

Asking one, and only one, important question within your Question helps attract prompt and clear Answers.

Your other questions are just as easy to research/ask separately!

That forewarning then makes it quite easy to place multi-part questions On Hold as Too Broad with an optional comment like:

As per the Tour there should be only one question asked per question.

The GIS SE Meta Q&A that led to the above was:

Where does it state "one question per post" in the GIS Stack Exchange instructions?

A logical guideline that flows from this is that a question like:

How do I do X in products A, B, C or D?

is also Too Broad because it is effectively four questions:

  • How do I do X in product A?
  • How do I do X in product B?
  • How do I do X in product C?
  • How do I do X in product D?
  • Isn't the too broad close reason enough for this? – NVZ Sep 3 '17 at 3:07
  • @NVZ Too Broad is used after the event whereas the Tour is pre-warning potential askers about what is expected before they decide to ask. – PolyGeo Sep 3 '17 at 4:08
  • 1
    Ah, prevention is better than cure. I see. Voting it up. – NVZ Sep 3 '17 at 4:55
3

I surveyed the site for reasons why we should avoid asking multi-question questions at meta.Islam.SE. I've been trying to get people there to ask one question per question (e.g. by commenting, by editing away all but one question, by closing as too broad, and by making the above meta post), but some users (new users in particular) don't like it.

While I feel it would be great to be explicit about this in the off-topic page, the line

Also do not ask multiple questions at once unless you expect their answers to overlap considerably.

might not be enough. If you give this inch, users will take a mile: "but my sub-questions overlap considerably". (My impression is that "overlap considerably" should mean "if they were asked separately, one would be closed as a duplicate of the other".)

If this is going to be implemented, I'd be consistent to re-use the too broad close reason description:

Avoid asking multiple distinct questions at once.

  • I see the same intransigence as you but I think it is a horse that we just need to keep trying to lead to water. One day I think they'll find opening a granular Q&A to find an answer "instant-like" is far easier than reading an omnibus Q&A with 15 answers to 3 questions in one, and then they'll get it. – PolyGeo Sep 12 '17 at 7:48
  • I also started my phrasing with distinct questions, but then I realised that it is not very clear – in particular to a new user – what makes two questions distinct … and arrived at my criterion of overlapping answers (which is more or less your duplicate criterion explained to a new user). Also, the help centre mostly exist to offer help to those who are actually open for it and I wouldn’t focus on catering people who want to misunderstand things (consciously or subconsciously). Anyway, those can misunderstand your phrasing as well. – Wrzlprmft Sep 12 '17 at 8:12
  • @Wrzlprmft If we used simply "Avoid asking multiple questions at once." then I think any definition of a what makes a question is unnecessary because in English it ends in a question mark. – PolyGeo Sep 12 '17 at 8:47
  • 1
    @PolyGeo: Question marks are not a good indicator for actual distinct SE questions. Consider the following examples: ➊ “I wonder how this works.” – no question marks, one SE question ➋ “Does this work? Why? Perhaps with discombobulators?” – three question marks, one SE question (in most cases) ➌ “How do X and Y work?” – one question mark, two SE questions (in most cases) ➍ “My situation sucks, doesn’t it? What is your experience with this?” – two question marks, zero SE questions. – Wrzlprmft Sep 15 '17 at 7:57
  • 1
    @Wrzlprmft I think you are reinforcing my point - in most of those examples carefully placing question marks would remove the ambiguity if they were used and read correctly. – PolyGeo Sep 15 '17 at 9:44
  • @PolyGeo: Which ambiguity? – Wrzlprmft Sep 15 '17 at 9:47
  • @Wrzlprmft 1. "I wonder how this works." would lose ambiguity between whether it is intended as a statement or a question by replacing the fullstop with a question mark. It would be even better to simply write "How does this work?" – PolyGeo Sep 15 '17 at 21:32
  • @Wrzlprmft 2. "Does this work? Why? Perhaps with discombobulators?" can be re-written as something like "How might discombobulators work if that does not?" – PolyGeo Sep 15 '17 at 23:36
  • @Wrzlprmft 3. "“How do X and Y work?" is two questions disguised as one and easily unmasked as "How does X work?" and "How does Y work?" making it Too Broad for focused Q&A. – PolyGeo Sep 15 '17 at 23:39
  • @Wrzlprmft 4. “My situation sucks, doesn’t it? What is your experience with this?” is asking two open-ended questions - the first seeking an opinion/confirmation so should be simply removed from the question - the second can have as many answers as there are viewers of the question so question is Too Broad unless that can be removed and a focused question spotted somewhere else within the question. – PolyGeo Sep 15 '17 at 23:42
  • @Wrzlprmft Question marks are a great indicator for where to look for the focused question we seek in every SE question but there being one does not always make it a focused question. Carefully (re-)placing question marks along with editing is a powerful tool for removing any ambiguity from SE questions but I think it is best if the askers learn to use question marks correctly so that they always ask a single focused question. – PolyGeo Sep 15 '17 at 23:48
  • @Rebecca If you've not seen gis.meta.stackexchange.com/a/3353/115 I suspect that it will resonate with you. – PolyGeo Sep 15 '17 at 23:53
  • @PolyGeo: 1) Your changes do not address ambiguities (not that there are any to begin with); at best they highlight the question. 2) "Does this work? Why? Perhaps with discombobulators?" can be re-written as something like "How might discombobulators work if that does not?" – I think you misunderstood my example. I would rather rewrite this as “How does this work, if at all? My best guess is that it uses discombobulators.” – Wrzlprmft Sep 16 '17 at 9:48
  • 3) Your guidance is difficult to apply, because it requires that askers already know which questions they should split and combine. For example, how should the uninitiated asker know that “How do X and Y work?” needs to be split, while it’s okay not to split “Does this work and if yes, how?”. Moreover, the entire process of rewriting the question to find out how many question marks you need seems overkill. Finally, do you really expect the asker to deduce all of this from “Avoid asking multiple questions at once.”? – Wrzlprmft Sep 16 '17 at 9:48
  • @Wrzlprmft Guideline: "Avoid asking multiple questions at once." Helping users understand guideline: Step 1: Assist with what makes a focused question - not everyone gets it initially so re-wording/re-arranging examples like you gave may be needed Step 2. Make sure there is only one focused question per question. I would expect a user to start to understand it soon after they start using our sites. I don't use a count of question marks as anything other than a guideline. I find it quite easy to distil questions from users to either one focused question or a question that is too broad. – PolyGeo Sep 16 '17 at 10:43

You must log in to answer this question.

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