# How can non-experts best contribute?

Hi all,

I'm a fairly recent arrival to math.se. I've asked a good amount of questions (and learned a lot from the answers). But I haven't had the chance to answer many questions.

I've been perusing some of the user profiles for the top users, and I see a variety of university faculty, very strong undergraduates, and hobbyists who know their field very well.

I try to answer others' questions when I can, but I know up to undergraduate-level mathematics, and since more people know how to do those techniques, those questions are answered very quickly.

So: are there other ways that I (and people like me) can help contribute to the site in addition to asking interesting questions?

Thanks.

• @Jasper Hear, hear! As I've mentioned on a couple occasions, if you browse through the lowest-voted posts of experts you will find many gems - often much deeper and more insightful than their highest-voted posts. Don't stop reading these posts just because you encounter a few words past your knowledge level. They can still plant germs of ideas that lead to Aha! realizations later in your studies. I learned many deep things this way from papers and lectures of masters. Apr 26 '11 at 2:13
• @jasper ironically that comment would be more effective as an answer so I could vote it up :) Apr 26 '11 at 9:57
• I have a question (which I can post separately, if that's more appropriate): When voting on answers, particularly answers to "elementary" questions, there are often two competing factors to consider: the intrinsic quality/depth of the answer, mathematically speaking, and the quality of an answer, taking into account the appropriateness of the solution given the background (or lack thereof) of the questioner. I realize that background info is often not available, but to a certain degree, it can be gleaned from the "level" of the question... Apr 27 '11 at 2:11
• If what I'm asking is unclear, I'll elaborate here or in another thread. If "appropriateness" is a key consideration, that may help motivate users to learn how Apr 27 '11 at 2:18
• oops...learn how to "come down a couple of notches". I found the "related" link on the right to be relevant to what I'm trying to get at: link. Apr 27 '11 at 2:23
• @Amy: you should probably ask this as a separate question. Apr 28 '11 at 1:05
• @Amy Personally, I upvote all the answers that contribute to a complete set of answers for different readers. I think that the answer that best fits the questioner should be the accepted answer. But the upvotes express what I find useful for myself and other readers, not only the questioner. Apr 30 '11 at 12:17
• @user9325 I like your approach, and you "upvote"/"voting" philosophy/criteria. Thanks for commenting Apr 30 '11 at 12:55
• This question is really nice! I would like to suggest put some point of the views in the answers into FAQ.
– user9464
Aug 3 '11 at 5:50

One way to contribute is by helping to improve answers by experts. One of the difficulties faced by experts answering questions by non-experts is "forgetting" expert knowledge, so that one can answer questions at the appropriate level of the questioner. Once one's mind has settled onto a "local maximum" viewpoint of something, it is often difficult if not impossible to shift to another (less-optimal) viewpoint. It goes against years of mental training to reason optimally in one's area of expertise - using the most general and powerful tools available. As a result, for example, an algebraic geometer presenting a purely algebraic proof here might mistakenly omit a step that is obvious geometrically, but nontrivial algebraically. You can help by pointing out these gaps in exposition. This not only helps improve the answers, but also gives valuable feedback to the answerer - feedback which can be incorporated to help improve future answers. Such feedback loops are essential to the success of the site.

• I wish I can upvote this more than once! There are at least two ways to "improve" other's answers. One is to directly edit them (fix typos, add clarifications), the other way, as Bill and Qiaochu mentioned, is to ask really good questions in the comments. Apr 26 '11 at 17:19
• Another pitfall of the "expert knowledge" is that sometimes the experts end up answering the "wrong question" by over-focusing on the subtleties, especially in the case when the OP has not identified the main problem he or she is having. How you can help in this case is left as an exercise (hint: apply J.M.'s "skin a cat" proposition). Apr 26 '11 at 17:29
• So for example, in the spirit of helping to improve answers by experts, should "it goes against years of mental training to reason optimally" have "suboptimally"? Otherwise I don't understand. Apr 29 '11 at 22:56
• @Harry In the example I gave, the "optimal/advanced" proof for the algebraic geometer employs geometric intuition. Apr 29 '11 at 23:20
• After coming back and reading it, I believe I misparsed the sentence. I read it as saying "reasoning optimally goes against years of mental training". It seems that it was just me, but I think it could be less ambiguous by moving around "mental training to reason optimally". Apr 30 '11 at 1:31
• @Bill: I had the same problem understanding as Harry did. May 29 '11 at 7:17
• This is the reverse Dunning Kruger effect and sadly I experience many cases on MSE of excellent mathematicians underestimating the ineptitude of lessers and then getting frustrated and even abusive thinking it is some bad attitude, deliberate awkwardness or lack of effort. The same goes for quickly closing trivial questions; when MSE is for mathematicians OF ALL LEVELS Apr 5 '18 at 8:29

In no particular order:

1. Vote.
2. Comment. Even if it's just to say "wow, this answer was really helpful, thanks!" But especially if it's to say something like "I'm not sure I understand what you did in this step," since if you didn't understand it, probably others didn't either and would appreciate the clarification.
3. Retag questions when appropriate.
4. Edit tag wikis.
5. Suggest tag merges and synonyms.
• I'd add something like: 6. Don't be too shy to add your own answer if you think you have something to say or clarify. Any complementary point of view will be appreciated by all the people involved.
– t.b.
Apr 26 '11 at 12:28
• If there's more than one way to skin a cat, and one finds that his/her "way" has yet to be posted, s/he ought to feel free to post it. Apr 26 '11 at 12:34
• The "thanks" comment seems to be discouraged at SO; show your appreciation by voting instead. Apr 28 '11 at 2:11
• Yeah, but no credit is given to users who work behind the scenes voting, reviewing, flagging, reviewing, voting, reviewing. The mods here, take such users for granted, and most users have no way of knowing about these contributions, and have no clue how much they're, and this site, is benefiting from such effort. Apr 3 '17 at 20:09

Some ideas:
1) You can read the answers and see if they contain errors. Some answers look fine at the first impression, but they have errors.
2) You can specialize in some area- you can analyze, read, try to solve problems from some specific subject, for example, everything about series or determinants. In this way your expertise increases.
3) A problem may be solved in many ways. Approach the problems from a different angle than the given solutions.

• I like all your suggestions, 2) in particular. May 13 '11 at 14:27

Michael, as a non expert, I really liked your question. My math.SE agenda would be something like this:

1. Vote - a lot. I mostly run out of votes in the day. This helps others that make good questions, and make other people eager to work on their answers.
2. Criticize constructively - from your (mine) point of view, you will find some answers expert users consider straightforward not so evident. This will clear other users' doubts too, and help the expert improve his pedagogical skills.
3. Browse my interests - I personally browse ,,, and to try help others and see what new things I can learn. Many questions in math.SE apply to a broad audience, but there are some questions/problems you don't really find everywhere.
4. Ask questions! (And don't be shy to do so, as long as they aren't off-topic and you follow the rules.)
5. Browse topics I am comfortable with - this will help you polish your knowledge and help other users improve theirs. It's a win-win situation. I personally started browsing some time ago to help users with that.
6. Comment/ down-vote - We like compliments, we like to chat a while, and as any fair "scientist", welcome criticism.
7. Help other users with . Although some questions tend to be rather imperative and impersonal, it works as a source of exercises you might be interested in being able to solve.

If you are interested in the records of that agenda, I have:

• Given $$195$$ answers.
• Asked $$60$$ questions.
• Commented $$802$$ times.
• Upvoted $$1357$$ times.
• Downvoted $$18$$ times. (I usually comment and ask for a correction before downvoting. I think it is better.)
• I agree with everything, but would like to add to your point #$7$, i.e., "Help other users with homework", that you try to avoid answering PSQ's. Soon after joining here, I wasn't as knowledgeable about this site and tried to be helpful by answering a few PSQ's. However, after learning more that this is generally frown upon (and I believe for good reason), plus one answer was down voted, and several were deleted, I learned to try to avoid doing this as it's overall not worthwhile. Feb 18 '19 at 23:38

Besides the many good tips that have been given, I would add this one: be on the look out for questions that are open-ended. The questions requiring a definite answer will either be homework questions or matters for experts. But the open-ended questions sometimes are accessible to non-experts and just require a bit of creative (not necessarily original) thinking.

Of course, even on these questions, experts will have an edge, but in my experience, depending on the kind of question, you can have a good chance of providing a good answer. Not only that, you will also enjoy thinking about them just for the sake of the problems themselves. And if you still get beaten to the finish line by an expert, you'll be able to appreciate the answer even more because you'll have thought about it yourself. And you'll gladly upvote the answer, as well as the question.

Almost $20\%$ of questions remain unanswered. A question is labelled unanswered if it has no upvoted answers. One thing you can do is sift through these questions. If you find a question with an answer, you may consider upvoting it. If there are no answers, you should consider providing an answer yourself. One benefit at looking at such questions is that there is no pressure to answer quickly.

You can find a list of such questions here. I would suggest starting with questions which have a low number of votes as they have probably received the least attention.

• You can also filter the unanswered questions by tag if you feel you have a better chance of contributing something to, say, an algebra-precalculus question than to a complex-analysis question. Dec 29 '14 at 19:11
• 275879 of which 4570 are negatively voted.
– user645636
Jan 21 '20 at 11:58
• where as in the all question queue we have 1196520 questions 18715 with negative votes.
– user645636
Jan 21 '20 at 12:07