# Question self-destruction: why don't experts ask more questions?

I've been kicking around this site for a little while now and I've realized that I'm very rarely even tempted to post a question. I'm an astronomy grad student with a background in physics, so naturally as I go about my life and work questions relevant to physics occur to me; it's one of my main areas of interest, after all. However, I feel like that same background/interests prevents me from actually wanting to post questions here. There is a very limited set of questions that (i) I think of and (ii) I can't reason out a plausible answer to, this is after all what my training as a physicist is supposed to make me capable of... or (iii) I can't find an answer to with a bit of research, another skill I think any self-respecting physicist possesses.

It seems I'm in good company. Most of the top answer authors on this site seem to have a ratio of questions to answers in the range of 1:100 or 1:1000. I wondered if this might be particular to Physics.SE (and other SE's where the topic is one that will naturally attract problem-solvers), and got sort of mixed results. SO's top answer authors barely ask anything, and likewise on Math.SE. Photography.SE, which I'd describe as a non-problem solving based topic, is similar as well. Bicycles.SE has somewhat more even Q:A ratios amongst its top users. Top users from Gaming.SE and RPG.SE ask a lot of questions, and I think I'd describe gamers as keen problem solvers. So hmm, not a lot of support for that hypothesis. It's probably a lot simpler; it's a lot easier to answer questions than to ask (good) questions, especially in the volume required to end up at the top of the rep scale.

Great questions can come from anywhere, and sometimes a simple question that anyone can ask can turn out to be very interesting and complex (my favourite example here is A mirror flips left and right, but not up and down). But a piece of old wisdom goes that the more things you know about, the more things you find that you don't know about. An expert (I use the term loosely, what I really mean in the context of Physics.SE is anyone with about the equivalent knowledge required for a bachelor's degree in physics) should have some advantages when putting together a question, though. They should understand the basic physics underlying the topic, so they can really zero in on the concept they want to ask about. They know where to look for reliable information and can understand and evaluate the validity of whatever background material they come across. And I'm sure there are more reasons. Of course there are also potential drawbacks to being an expert trying to write a question, which I sort of outlined above.

So my question is: Why do you think many clearly knowledgeable users don't ask more questions? Can we/should we/how can we encourage them/help them to ask more? I feel like there may be an untapped wealth of great questions lurking out there.

• Some users (Qmechanic for example) have 0 questions – Manishearth Jul 15 '13 at 17:06
• I noticed that. Couldn't decide whether to call that Q:A ratio $\infty$ or undefined, so I just left it out heh. – Kyle Oman Jul 15 '13 at 17:08
• Similar question from the mother meta: meta.stackexchange.com/q/13611 – dmckee --- ex-moderator kitten Jul 15 '13 at 17:46
• @Kyle: Q:A ratio would be 0. – Abhimanyu Pallavi Sudhir Jul 15 '13 at 18:46
• @Dimension10: Ehhhh you know what I meant, the inverse of that ratio. – Kyle Oman Jul 15 '13 at 18:48
• @Manishearth Qmechanic has zero questions because Qmechanic knows everything. This is common knowledge... – twistor59 Jul 16 '13 at 18:09
• We should not encourage experts to ask questions. If an expert has a question he can't answer then it belongs to physics research... aka in the realm of grants and minions. – Dale Jul 16 '13 at 18:36
• I am also one of the guilty ones. The answer is that 1) easy questions I clear up through google or my old textbooks 2) if it is a controversial question I look it up in Lubos' blog 3) if it is bordering on metaphysics, (due to my age :) ) it does not belong here. Anyway high reputation depends on the length of being on this site. Ron Maimon is still gaining reputation after more than six months of voluntary inactivity. – anna v Jul 18 '13 at 4:01
• What exactly has this to do with "self-destruction" ? . – Abhimanyu Pallavi Sudhir Sep 5 '13 at 16:10
• this is very likely a system-wide property of stackexchange, have observed same pattern on generally all sites visited. or it may even be a deeper aspect of human sociology/psychology. there is a catch22 concept between being an expert and asking questions. some people do not know answers because they havent read existing literature, others are well versed and dont know because the answer is not known. the questions that experts ask tend to be more active-research questions or "open questions" in the field. also, there is the social inhibition of not wanting to appear not-knowledgable... etc – vzn Dec 10 '13 at 17:30
• I personally don't because most of the questions I would want to ask would be closed as off topic "engineering". I guess take that for whatever it's worth. – DanielSank Dec 23 '14 at 3:02
• Relevant on MSO: Do the top answerers have secondary accounts to post questions? – Emilio Pisanty Aug 8 '16 at 11:28

Can we/should we/how can we encourage them/help them to ask more? I feel like there may be an untapped wealth of great questions lurking out there.

Yes, we should encourage expert users to ask more questions in their area of expertise. I agree with you that the proportion of good questions would be greater if the experts shared more.

Here are five suggestions for experts considering contributing a question:

2) Write questions for an audience of physicists trained outside your specialty.

3) Share the big, significant, important questions in your field.

4) Elaborate on the scientific news that does manage to get attention in the media.

5) Ask questions with answers that actually matter to the quality of life for people on our planet.

6) Be an example of reason, scientific methodology, and professionalism.

Reasons to contribute

1) We become what we write.

2) There are too few examples of scientific thinking available to people outside a university.

3) Good posts have a feedback effect by encouraging others – making physics.se more fun.

• I like #4 especially, there are always people with questions when science ends up in the news, and someone with a background in the field probably has a good chance of identifying the key points in a question (or parts that the media didn't capture in the press release). – Kyle Oman Jul 18 '13 at 18:39
• The feedback effect is true in a very verifiable sense. In some questions I've outright linked to previous questions that make me think of the idea. For some sets, you could nearly draw a tree of influence. – Alan Rominger Jul 19 '13 at 18:06
• It is always a touchy game to talk about one's own work. I tried to distanciate as much as I could when I had the opportunity. It is difficult not to support one's own work even though there may be dissenting views. Talking to specialists is different, they know you are supporting a specific view and take it into account. It is their job to be able to do it. No so for other people. – babou Jul 22 '13 at 18:05
• I like number 1 the most, thats the point of being able to asnwer your own questions. maybe if everytime the OP had a question that they answer themselves and they post it here we would have the advantage of his expertise for future use – RhysW Jul 26 '13 at 11:14
• I recently posted a self-answer question that was a great interest to at least myself, and it was downvoted almost as much as it was upvoted. Seems like someone around here doesn't like self-answered posts, and as an "expert" user I find that rather discouraging. – DanielSank Feb 11 '17 at 20:25

For me, if I have a question, I usually do the research myself and since I'm in grad school, I can use my library to get any paper I want to read.

But the other side of the coin is that I'm paid to ask questions and answer them. That's what research is about. And in my field, the competition for money is fierce and we have to be very careful to protect not just our results but the questions we are thinking about so we don't get sniped by our competition. So I can't ask my questions here.

• True, but I doubt that every interesting question that occurs to you is directly related to your research. Still, I get that it's hard to find an interesting question not related to your work and that you can't solve yourself quickly... I get the same thing. – Kyle Oman Jul 16 '13 at 15:43
• @Kyle You greatly over-estimate my curiosity ;) In all honesty though, I don't actually come up with questions that are outside my area, at least not ones that would go on Physics.SE. I think I'm just kind of boring? – tpg2114 Jul 17 '13 at 2:49
• This makes a good point. Responding to @Kyle : Also, a decent number of the interesting questions that occur to me are inspired by others' research in my field, where I am privy to their unpublished results. In those cases I am even more worried about spilling the beans than with my own personal research. – user10851 Jul 17 '13 at 5:54
• I imagine this is the correct explanation, at least for practitioners. I'm saving all of my good questions for PRL. ;) – wsc Jul 26 '13 at 3:00
• I was a grad student until recently, and I have to say I don't understand this. Most of the hard part of research and understanding stuff that others have already figured out. I have no qualms about asking for help with that stuff here so that I can get to the interesting parts of research more quickly. – DanielSank Feb 11 '17 at 20:26

I believe I am in the minority of high rep users who ask a lot of questions. I've asked 65 questions, and this is high compared to the next 5 users ahead of me, who've asked 3, 2, 20, 29, and 2 questions. When someone writes 200 answers and asks 2 questions, then after all the time they've spent on the site, it's pretty obvious to call it a pattern.

So why do so many advanced users ask few, if any, questions? And why am I an exception? Easy - I'm an engineer, not a physicist. Ok, I'm sure we have plenty of engineers and otherwise non-professional physicists here, and among the ranks of the high rep users. Myself, I've just always had pent up angst against physics, since I started out excelling in it, but specialized in more lucrative directions.

Here's an example of a random thought experiment I had.

Why aren't gas planets and stars fuzzy?

In retrospect, it's fairly trivial. But yet, it's a highly upvoted question. If you share a curiosity with a genuine technical basis, other people will find it interesting. Being extremely difficult to answer isn't a requirement for questions here. If you're not willing to occasionally throw your hands up and say "why not, I'll do a question", then you'll rarely ever post one.

Experts would ask more questions if it was more "banter" than questions because their thoughts can be expanded on by all parties. Of course that has issues of scope. Stack Exchange Q&A format is a way of enforcing good structure and scope, and even so, we go outside that scope all the time. New questions are often answered by an answer on an old question that went off topic... and thus answered the new question. Seriously, it can get to be a problem with close votes.

• As one of the 2's in {3, 2, 20, 29, 2} I will say I've admired your commitment to asking - more of us high-rep users could stand to ask imperfect questions for the good of the site. I think you've hit the nail on the head, though (typical engineer...), as it's just so easy to get distracted formulating a great question so as to never end up asking it. – user10851 Jul 16 '13 at 1:31
• I guess as a high rep user in general you get too nitpicky about what questions you ask. I sort of touched in this in m answer, but this is more thorough. – Manishearth Jul 16 '13 at 4:51
• I've come up with lots of questions over the last few years, but usually I've been able, through a significant amount of work trawling the internet, to find the answers. This has taken, in many cases, a serious amount of time, and I've ended up generating lots of notes on the topics in question. It would probably have been quicker to ask the questions here, but I'd have missed out on the journey. Maybe I'm not alone in this? – twistor59 Jul 16 '13 at 18:24
• @twistor59 I think motivations can be varied. One reason to ask a question that you can answer yourself is that others will approach it with a completely unexpected method. I learned about 4-vectors for momentum balance on Physics SE this way. That was a new method for a problem I could have otherwise solved clumsily. Questions also give me a URL to use later to evidence a claim, so what I learned will never be lost, like notebooks... notebooks are always getting lost. – Alan Rominger Jul 16 '13 at 19:22

Because writing a good question is hard. New users post questions that aren't so good -- not as much effort is put into it, etc etc. Veteran users are more careful about what they ask because they know hat it's like to be on the other side and will want to ask a question well-tailored to get answered.

While I'm no expert, I certainly have my times hen I get a physics question and I'm like "I'd better post this on SE". But I first sit and mull over how I'd phrase it. And then I make sure that I'v tried all I can to solve it by myself. In doing so, I usually solve the question on my own.

What would be beneficial is if experts self-answer tough, relevant conceptual confusions that they encounter and solve on their own. Instead of encouraging them to ask questions, which isn't exactly feasible due to what I mentioned above, we can encourage them to make more use of the "answer own question" feature.

I almost did that with this answer. I and my friends had been discussing the apparent paradox, and later on I worked it out on my own. I was thinking of self-answering this, except my friend (to whom I had recommended the site), posted the question first.

Besides this, there's always the uncertainty that a question won't get an answer (See @DavidZaslavsky's answer). This doesn't apply to me as much as it applies to the experts here -- I'm reasonably certain that any questions I ask here would get a good answer.

• As a 2:196 Q:A person myself, I personally have considered the "answer own question" route - but, well, that just seems like a lot of work to do oneself. Also, if I'm just answering, someone else at least walks away with a better understanding. If I'm just asking, then someone else gets the fun of thinking about a good question. If I do both, it becomes less clear that anyone other than me is benefiting. – user10851 Jul 15 '13 at 18:06
• And yes, I understand SE has the goal of making a broadly useful repository of knowledge, but if I ask and answer a question, then even if a bunch of upvotes come in, I'll suspect they're all from people who knew the answer but appreciated the effort I put in. It's hard to put in the effort without so much as a specific (albeit likely anonymous) username that I know walked away from the post with a better understanding/appreciation for some aspect of physics. – user10851 Jul 15 '13 at 18:08
• @ChrisWhite There is a slight benefit here. Codifying what you already understand into words that others can understand makes it much clearer. There are many times when I think I understand something only to realize, while writing an answer, that I don't. – Manishearth Jul 15 '13 at 19:33
• You're partially right about that, though. Expert questions are of interest to a smaller community, ergo your answer helps less people. That can't be helped, I guess. – Manishearth Jul 15 '13 at 19:34
• I would argue that the need to make sure that I've tried all I can to solve it by myself is self-imposed. If you recognize a question as valuable, why not post it? If you later figure out the answer you can still post that. – Emilio Pisanty Jul 19 '13 at 17:44
• @EmilioPisanty Yep, and I've mentioned that in the next para :) – Manishearth Jul 19 '13 at 20:48

I think you identified the reason already:

There is a very limited set of questions that (i) I think of and (ii) I can't reason out a plausible answer to, this is after all what my training as a physicist is supposed to make me capable of... or (iii) I can't find an answer to with a bit of research, another skill I think any self-respecting physicist possesses.

There is a "middle class" of questions where the person asking doesn't have the expertise to answer them, but somebody else does, but in order to make such questions "profitable" to ask on the site, we need to have a wide variety of topical experts, many many more than we do now.

• 1. But why would many topical experts stick around if there's nothing for them to answer? 2. You only answered (the easy?) half (of) the question :) – Kyle Oman Jul 15 '13 at 17:33
• Yeah, I don't have any great ideas for the hard part. Of course you're not the first one to have thought of this. – David Z Jul 15 '13 at 17:51

I have a PhD in physics and teach the subject at a community college, so I guess I would count as an expert. However, my expertise is concentrated in a couple of areas. My Q:A ratio is 29:262. Sometimes I ask questions about subjects I don't know as much about, e.g., field theory. Sometimes I ask questions and get zero answers, which is discouraging. I recently offered a small bounty on a question and still got zero answers. Other times I'm very, very happy with the answers I get, e.g.: Radio antennas that are much shorter than the wavelength and What determines the angle of the cushion on a pool table? . (In the latter, the answer was given in a comment -- this seems to happen more with nontrivial questions, maybe because people are less sure their answers are right.)

• It also has to do with the commenter feeling they don't have the time to do the answer justice. I often will say "You need to look up X" as a comment when I know the answer but don't have the time to write a full answer... – tpg2114 Jul 18 '13 at 0:04
• Your questions are great; usually, they're also damn hard :-) I forgot about that pool table question. I guess I could still turn it into a proper answer, although I don't have much to add. – Pulsar Jul 18 '13 at 2:53

I recently asked this question precisely in the spirit to have more "expert" questions, and I'd like to share this thought while it is still fresh in my mind:

Reading a paper related to my research prompted a tangential question that became my Question. It took remarkably little time to form and emerged very clear and crisp in its full form. Essentially: "Oh, that's curious! It's not important here, but surely one can do better experiments now. I wonder what one would see then? What's the theory behind that, anyway?".

However, actually posting the question on the site was a pretty laborious effort of phrasing it correctly, furnishing it with the appropriate context for it to be answerable, and the mechanics of digging for the images and links and references. In this instance that effort was probably "cheaper" than trying to figure out an answer (which would involve some digging for the state-of-the-art on that measurement, at least, and I'm not in the mood for more papers), but this is probably rare. I also have rather faint hopes of that question getting an answer here, as I feel there are few molecular physicists around.

I really like this kind of question. I will not really try to answer it, since I consider Mark Rovetta already answered it perfectly somewhere else on this page. Instead I will give my feeling and experience, as a user of this website from a research-like perspective.

I would like first to say that I already elaborated somewhere else that the quality of the questions and answers sounds to me to depend on the topic. I pretty much like the condensed matter questions and answers on this website, as well as the quantum information ones for instance. And that's great because that's my research topics !

I have a background in superconductivity and quantum optics. So, as far as I can, I've tried to answer -- since less than one year I'm using SE -- the questions on superconductivity which were unanswered, and more generally on condensed matter when I know the answer. It appeared to me that the quantum optics questions are most of the time correctly answered, so I've nothing to add about them.

Since one year, I'm trying to learn -- as a post-doc -- more about topological issues in both condensed matter and quantum information topics. Clearly, I will not ask question here like

which would be ridiculous. But at the same time, it's pretty hard to understand the problems at the edge of two topics.

How I'm using this site is as follow:

• Sometimes I just want to test some ideas with other communities people (especially high-energy physicists for instance) to compare their point of view with my poor-condensed-matter vision (like this one, or this one).
• Sometimes I want easy access to a stream of articles, for which I would spend a lot of time finding relevant ones (like this one), and which are related to my current problems.
• Sometimes I would like to understand in more details a point I've no time to stop on at the moment (like this one), since it's just an aside of what I'm thinking at the moment.
• Sometimes I want to point out something I consider not well discussed in literature (like this question, or this one).
• Sometimes the questions are just something I consider funny when I think about them, and which could be interesting for someone else (like this one), but I've neither time nor interest to keep on thinking of.
• And yes, sometimes I'm asking purely fantasy, bulls... questions which are completely stupid (like this one).

Most of the time, I try to make useful questions, not clever ones. I mean, I'm trying to ask questions which could be useful for others, not specifically dedicated to my problems of the moment. I'm usually writing on a piece of paper some questions I would like to answer, just inspired by the time I'm thinking about, most of the time after reading some papers. I feel some of them could be interesting for the SE community, so I try to publish them on this website. I do not think more than that !

The funny part is that two colleagues at my lab are using SE, too. We sometimes discuss the questions/answers together, to try to improve our understanding about specific staff for instance. What I want to point out here is that a researcher needs to discuss a lot in order to (un)validate her/his ideas, especially at the early stage of the elaboration of a new concept. To try to collect as much as possible opinions is the key I believe. This website is really useful for me in that respect. But still it's highly time consuming. As anything you have to type, you need to think about, read, write, clarify, read, write, think about ...

I'm trying to promote SE as much as I can around me (in the real life I mean). The most important reason why researchers do not use this website more often or as a working tool is simple I believe: they don't have time !

A view from the other side: why did I ask a question here? As with others, I am/was an "expert" (PhD in astrophysics, although that was 10 years ago and I left academia directly after it). However, that doesn't mean I know everything about physics.

Therefore when an interesting (to me, anyway) debate was occurring on Photography.SE about classical vs quantum optics, the obvious thing to do was to send it over here, where people who are better at optics than I am would be able to help out. Sure, I probably could have done the research myself (and I had a strong suspicion as to the answer), but if there are friendly, clever folk here who will help me to confirm my thoughts, why not take advantage of that.

At this point, it probably helps that I both have a more than passing knowledge of the subject matter and am also an experienced user across other bits of Stack Exchange, so writing a "good" question was relatively easy for me. However, I'm not going to be unique in this - there are going to be other physicists who want answers to a question from a different area of physics.

Disclaimer: this is not a complaint, just a thought and an observation ... ;-)

Maybe experts prefer helping others by answering questions and do not ask much here, because Physics SE is not an "expert-level" site. The number of people who could answer their questions at the same level might be too small.

To better explain what I mean, when Lubos Motl reads in interesting paper concerning his own subject, thinks about it and comes up with some interesting thoughts and questions he rather posts an article on his blog as he did for example here to obtain a better response than he would get when posting a corresponding question here. And think about it, who should answer, if people like Qmechanic have a question about something they do not understand concerning their own research subject? There are simply not enough physicists of this calliber here who could appreciate or even answer such questions.

In my opinion, if one wants to have questions from experts (about their own subject at least), the size of the community of fellow researchers who can appreciate and answer such questions must not fall below a certain critical value. For example on sites with a higher number of researchers like TP.SE experts did, and on MathOverflow they still do ask more questions themself. With tpg2114's objections concerning the TP.SE example I disagree. This was a nice small extremely high quality site and it only died because the SE network is not appropriate to support such small communities and maybe they could have lowered the bar to ask questions a little bit as I said here.

So in summary I would say, if we want to have more questions from experts about physics issues they are working on (?), we should first of all try to attract more experts and researchers to the site and make them feel at home here.

• Might I suggest observation bias in your statement? For one, TP.SE didn't survive so it's hard to use that as a model for good behavior. Second, looking at the top 10 users on MathOverflow, the Q:A counts are: 39:869, 62:792, 3:408, 58:447, 21:413, 52:425, 0:504, 144:459, 259:444, 94:858 – tpg2114 Jul 19 '13 at 23:53
• So with the exception of 2 of the top 10 users on MO, they really aren't any better than any other site at top users asking instead of answering questions. – tpg2114 Jul 19 '13 at 23:54
• @tpg2114 I have extended my answer a bit to clarify what I mean. – Dilaton Jul 20 '13 at 10:37
• @tpg2114: Isn't that high . ? – Abhimanyu Pallavi Sudhir Aug 4 '13 at 11:31

I had a longer answer, but I will limit it to this, for what it is worth.

I do write many more answers than questions. I answer only relatively simple questions, as my math is no longer what it used to be.

I am no expert in physics, but I am a scientist, and I can often find my own answers. I also learn by answering questions (typically what I did about the working of ABS brakes, on which I knew practically nothing).

I did ask a very decent question of physics, for which I was downvoted (-1) for no reason I could identify (52 views). A few comments, but no answer. I provided an answer myself (read apparently by no one). The purpose of the question was to understand how people make assumptions, as the answer to that question contradicted assumptions made in a previous one that was referenced. I still believe it was a good question, even though the purpose of it was (barely) hidden. I think understanding methods of reasonning is also part of science.

This is not encouraging.

I do have questions to ask. But I let them ripen. I feel that I will not (should not ?) get many chances of asking, and I better ask the right questions.