# What are your thoughts on users who ask too many questions?

It seems that some folks are asking large amounts of questions. I remember seeing one fellow ask over 10 questions in an hour!

This is getting outrageous, should we introduce a limit on the maximum number of questions per time interval?

• You can only ask 50 questions per month on certain high use sites. You can review the Rate Limiting Guide for more limits if interested. – Erik Mar 25 '16 at 17:37
• @Erik Thanks for these links; the Blankman's meta thread is hilarious. Checking the Guide, the limit is 50 Qs per month and 6 Qs per day. – amoeba Mar 25 '16 at 21:07
• @amoeba yeah those are the limits but the Guide implies that those limits might not be in effect here. The Blankman thing is amazing and hilarious. They have >55k rep and less than 20 answers (on SO).... That ratio is baffling to me. I'd think after +1k questions they'd learn enough to give a bit more back to the community in the form of answers. – Erik Mar 25 '16 at 21:11
• @Erik Yes, 55k rep earned exclusively by asking questions, most of them apparently pretty dumb, is, well, weird. But then StackOverflow is kind of crazy: Blankman has 1200+ upvotes for a trivial one-line question with a trivial one-line answer (that has 1800+ upvotes itself). – amoeba Mar 25 '16 at 21:16
• @Erik Asking (decently good) questions is a great contribution to the community. – Franck Dernoncourt Mar 29 '16 at 18:40
• @amoeba One-line questions are sometimes very useful to ask. The question you gave as example has almost 1 million views so far. Doesn't it show it was useful to ask? – Franck Dernoncourt Mar 29 '16 at 18:42
• @FranckDernoncourt I agree. Questions are clearly essential, but answers are essential to bringing in more questions. Which leads to a chicken and the egg relationship to the two. All that being said I personally feel that people should strive to do more than ask questions – Erik Mar 29 '16 at 18:46
• @Franck: Yes, it is useful (in some sense). But my "kind of crazy" referred to the amount of upvotes, not to the amount of views. The fact that one can get ~20k reputation on StackOverflow for answering one single trivial question with (more or less) one line of code, is in my opinion quite crazy. 20k rep unlocks all privileges accessible to non-mods and in general appears very respectable and hard-earned (at least on CV). – amoeba Mar 29 '16 at 18:59
• @Erik I recall Quora used to limit the number of question users could ask by using some credit system, but they have removed it. The seeder vs leecher is a tricky question indeed. My view is one learns more by seeding than leeching, so up to the users to do what they feel. – Franck Dernoncourt Mar 29 '16 at 19:02
• @amoeba sometimes I wish there was an option to hide all the SE gamification, just focusing on content :) well, at least, one would hope the highly voted answer to work ok. – Franck Dernoncourt Mar 29 '16 at 19:02

I presume this is a tongue-in-cheek self-reference, but the question is a good one.

Let's take it as a general question -- so let us imagine that we face some case where someone posts a lot of questions. I don't think that some high number of questions a day is automatically a problem in itself, as long as the usual conditions for a question to be suitable for the site are met.

So if the questions are:

1. clear, well-formed, answerable questions (i.e. not in the form of a thesis) that show some basic research (awareness of definitions, familiarity with basic explanations of concepts from a textbook or an on-line resource like Wikipedia, awareness of one or two related questions already on site)

2. also relevant/on topic, not duplicates, not too broad, not opinion based etc and generally follow the various guidelines for asking a good question not already mentioned above

then the raw number of questions is generally not of itself an issue. We'd get what -- 100 questions a day or more on site? (This is not based on checking the figures, but one can count how many pages of new questions posts appear in the last day or so readily enough - and I sometimes do that, so it's not hard to get in the ballpark)

Good questions are the lifeblood of a question-and-answer site, and good questions are actually hard to write (a reasonably good question is harder to formulate, I think, than a reasonably good answer*) -- if someone who posts a lot of questions is taking the trouble to formulate their questions well, they're contributing.

* From my own experience at least, good questions are very difficult. While I am satisfied with a good fraction of my answers, I don't know that I am really satisfied with any of my questions. As a result, when someone can write a good question, they're a treasure.

To remove any hint of the personal from this -- at least as might be seen as personally relating to you -- let me discuss another user who in the past posted a lot of questions. As might be expected, the danger with posting more than a few questions a day was that there was a tendency not to do enough research and show a lack of thinking about comments on the question (which might often lead to suitable improvements) but simply to comment back almost instantaneously, often leading to a string of new questions in comments, each itself showing a lack of research.

Such an eventuality can be very frustrating and does lead to issues: you can for example end up with one user absorbing a lot of the effort of the people who are trying to edit, clarify or otherwise make questions answerable on site, at the expense of other people with questions -- not just because of the volume of the questions, but the lack of care in formulating them.

Correspondingly, it would also be necessary to avoid, as far as possible, holding discussions in comments (which SE network policy already asks us to avoid, but which become more crucial if you want to post many questions) -- respond to comments by improving (i.e. editing) the question as far as possible, rather than generating further questions in comments, and wait long enough to understand the answers before posting any related questions$^\dagger$). Only respond in comments where you need to clarify the comment or to directly answer a question that (for some reason) could not be edited into the question in some way (this would be very rare).

$^\dagger$ Since answers may take a few days to appear, this would typically mean that a string of related questions should normally be spaced out over time. [Multiple unrelated questions could still appear on the same day, if the above considerations are followed.]

With a high volume of questions and a lot of posted comments, there's a strong risk of annoying users, so one must strive as far as possible to work "above and beyond" what would constitute an average question quality, and to make sure there's no risk of comments seeming to be trivial or thoughtless.

As a result I think it's incumbent on a user that wants to post more than a a few questions a day to put in additional effort to make the questions they do post require less-than-average effort from others to make the questions answerable (so they're not 'hogging resources'), rather than posting low-effort low-research questions and failing to address issues with the questions when they're pointed out (which are obvious risks with a high volume of posts).

So an acceptable but high question volume is possible, but quite difficult, since the level of effort required to make the questions suitable will also be high. In particular, it is doubtful someone could post more than about two questions in an hour and still be doing a reasonable job of it, unless they were already so expert that they hardly needed the answers. [I doubt I could post more than two adequate questions in a day myself.]

Conversely, a limit on questions per day would be no guarantee of additional effort; better I think to encourage users to post better questions (via the usual tools --- comments, placing questions on hold, downvotes, closure/deletion etc) than to restrict the volume -- however, if someone posts enough bad questions, downvotes and closure will eventually limit their posting rate (the system does it automatically), but it tends to be fairly unusual to encounter it if one takes even a little care to post some good questions or improve old ones.

--

There's a corresponding danger with people posting a lot of answers per day, of course -- they too will tend to risk being glib, poorly explained, same-ish, and not especially useful. One way for answerers to keep an eye on whether their volume is getting too high is to keep an eye on their acceptance rate (and on upvotes/downvotes -- though that sometimes tends to be a fairly noisy indicator, if you regularly get more than a tiny fraction of downvotes you're probably better to answer fewer questions with the same total effort).

I do answer a lot of questions myself -- I will probably pass 3000 answers on site in the next 24 hours (edit: I did) -- so this is an issue that I try to keep in mind as I go along -- but my acceptance rate is pretty good (close to 1400 of my answers have been accepted) and my downvote proportion is very low (0.2% of the votes I get are downvotes), so at least by those crude measures I'm probably pretty safe. We have had users in the past with very high answer rates but pretty low rates of acceptance and much higher rates of downvotes; and in those cases there would be a reason to think that more work on the individual answers (with correspondingly fewer answers posted) is called for.

there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns -- the ones we don't know we don't know. -- D. Rumsfield

The problem with using Q&A sites like CrossValidated for learning new stuff is that you can only ask about known unknowns -- things you know that you don't know. Sites like this will never be a substitute to going through a handbook, attending lectures, or even participating in one of the multiple available online courses. Questions you ask and the answers you get cover the topics only partially. So asking multiple questions may lead to illusory and fragmentary knowledge if used instead of studying some topic by yourself. So I would say that there is also a risk for the person who asks the multiple questions of gaining false sense of knowledge.

• i agree. we should make sure users know this – user46925 Apr 1 '16 at 12:27
• Good point - I'd add actually performing data analysis. – Scortchi - Reinstate Monica Apr 1 '16 at 13:30
• Do you mean that data analysis can only be learned by practice? – user46925 Apr 1 '16 at 13:51
• I meant just to put data analysis on a par with reading, lectures, bookwork, &c, as important things this site can't substitute for. All the same, I'd be happy to say that. – Scortchi - Reinstate Monica Apr 1 '16 at 14:10

Users who ask too many questions are welcome, as long as the questions are decently good. Questions are very useful contributions to a QA community.

(How good a question is is a bit subjective, just vote. If a user asks too many downvoted or closed questions, they will be banned from asking questions anyway.)

• Yes, but it is not humanly possible to ask 10 good questions per hour. The specific fellow whom OP has in mind has probably one half of their questions downvoted or closed (and another quarter has zero upvotes). Some occasional questions are quite good, to be fair. – amoeba Mar 29 '16 at 18:52
• @amoeba Then indeed -> banned. It's hard to set a threshold though on number of questions per hour (e.g., the user may a draft many questions throughout a day, then dump all at once during his SE time), but easier on when to (temporarily) ban. In other words, question throughput is question different from question quality. – Franck Dernoncourt Mar 29 '16 at 18:57
• The wording "too many questions" is already an evaluation, and raises the question of who decides and on what grounds. (No criticism there as these are the words in the title.) As with everyone else who has commented, I am in the odd situation of being totally aware of why this question arises now, but do not want to personalise discussion. The bigger question is what happens when there are many downvotes, many questions closed or on hold, and many flags, all triggered by one member. – Nick Cox Mar 29 '16 at 19:32
• @NickCox Then the OP gets a post ban, right? – Franck Dernoncourt Mar 29 '16 at 20:10
• Yes, at some point behaviour deemed antisocial triggers a ban; it's just that most of us don't know where that point is, apart from moderators using their powers. – Nick Cox Mar 29 '16 at 20:12
• @NickCox The algorithm to decide the ban is indeed not public. meta.stackexchange.com/a/60294/178179 No idea what purpose it serves to keep it private, sounds like this gives room for Stack Exchange employee's arbitrary decisions. – Franck Dernoncourt Mar 29 '16 at 20:17
• I suspect a much, much better reason is to stop some people making calculated decisions on what bad behaviour they can get away with. Statistical people should not be surprised about outliers among the members here; it's just hard to know how to deal with them. – Nick Cox Mar 29 '16 at 20:32
• @Franck It's valid to have a suspicion like that. However, the bar appears to be an exceptionally low one. The automatic ban is rarely triggered on this site, and never for very active people no matter how bad their posts have been (perhaps because they always manage to get a few upvotes here and there in the same way the roomful of monkeys will occasionally type a line of Shakespeare). That leaves the onus on us to make such decisions as a community. Your moderators tend to act conservatively, taking action only when its need is blatantly obvious. That's intentional. – whuber Apr 1 '16 at 19:03
• @whuber if the bar is too low, how about setting a CV-specific bar? – Franck Dernoncourt Apr 1 '16 at 19:05
• I don't think we have that power, Franck, but the present circumstances have me thinking about starting a conversation with the SE team about it. I just don't know what to request: exactly what criteria would we want to apply? In the meantime, a little judicious downvoting by just a few more people might trigger existing criteria. That needs to be a personal decision. It takes extreme circumstances before I would publicly recommend downvoting any particular post and I would never recommend focusing one's downvoting on an individual: treat the posts on their own merits. – whuber Apr 1 '16 at 19:15
• @whuber I noticed that OP's account has been "temporarily suspended". Can you say if if it finally happened automatically or via some moderator action? – amoeba Apr 6 '16 at 8:09
• @amoeba The suspension was a moderator action unrelated to the issue raised here. – whuber Apr 6 '16 at 13:44
• @whubber please do not disclose personal information. this is against the rules and regulations. – user46925 May 27 '16 at 12:39

I believe that it is indeed a great idea to ask questions. However, one should be careful not to offend with lame questions.

• Do you really think this is a better answer than every other answer provided to the thread? – Sycorax Apr 1 '16 at 13:42
• yes, it is much more precise and condensed – user46925 Apr 1 '16 at 13:49
• why the -1? i gave a perfectly good answer – user46925 Apr 1 '16 at 13:54
• The tooltips to the downvote arrows provide clues in the case of downvoting without comment. You can see the tooltips by hovering your mouse over the icon. On meta, the tooltip reads "This answer is not useful." – Sycorax Apr 1 '16 at 13:58
• I agree with the sentiment of this answer--but it appears in the present instance that it is more honored in the breach than the observance. Many community members are aware of that (and have begun expressing their annoyance) and accordingly view both the question and this reply as disingenuous or worse. Whether this guess is correct or not, it still might be constructive to interpret the downvotes in this light and act accordingly in the future. – whuber Apr 1 '16 at 18:56
• =(, I can't believe we live in a world where freedom of speech gets public tormentation. – user46925 May 27 '16 at 12:54
• @Phil xkcd.com/1357 – Sycorax Jun 15 '16 at 13:54