I am used to blogging, so when I answer a question I often reference previous answers. I just noticed that that is not a good practice here, since the answers change order, depending on the votes. (I had said I disagree, meaning I disagreed with the previous answers. Then my answer was moved to the top, so it looked like I disagreed with the question somehow. I have edited my answer.)

So I see one non-obvious principle. Don't reference other answers implicitly. (If it's important, can you provide a link to the answer you are referring to?)

In case the question gets edited, the more self-contained your answer the better.

What are some other principles you've learned as you've written things on SE sites, for how to write your answers well?

  • $\begingroup$ Just for the detail on link: this will be a lot more convenient later, then every post has a "share" button next to "edit" and you will get a direct link (I think this is disabled due to private beta, but check some other SE site to see it) For the moment you can get the link like this (perhaps there is a better way): matheducators.stackexchange.com/q/xyz where the xyz is to be replaced by the post number. To find out this number click 'edit' for the relevant answer, it is then the number given between "posts/" and "/edit" in the URL Or, for simplicty no link and UserName's answer. $\endgroup$
    – quid
    Commented Mar 17, 2014 at 23:27
  • $\begingroup$ It's worth noting that an answer should always contain its own content rather than purely responses to other answers; there's a standard Stack Exchange Not an Answer flag reason saying something along the lines of "this does not appear to answer the question but is instead a response to another answer or an additional question. When you have enough rep you can comment". $\endgroup$
    – AndrewC
    Commented May 14, 2014 at 18:28

2 Answers 2


Throughout my college career, I have found it beneficial to outline the question asked at the beginning of an answer. Not only do I have to restate the question to make sure I understand it, I also am able to take up more space on the paper with my answer!

I believe the practice carries over to SE as well, though only for the first reason. When writing a good answer, restating the question so your readers are on the same page as you is beneficial. Many of the questions asked here have multiple parts to them with one underlying theme. If you were to just start answering without giving your reader a frame of reference, it could be quite easy to become lost within an answer.

I'm not sure if referring specifically to another's answer is a good practice within an answer. If you agree with their answer, you upvote it. If you mostly agree, comment. If you disagree, well, that's why you're providing another answer. It could be helpful to state why your answer is different from the other answers, but that can be done without specifically referring to another's answer.

I feel that many of these issues will be resolved simply with maturity of the community and some helpful guidance from the moderators.


Excellent question! In the early days of Stack Overflow (before there were comments!), answers regularly referenced each other. (My top-voted answer still bears the marks of that practice.) But it quickly became obvious that answers work best when they are reasonably self-contained. This is why you might occasionally see complaints about link-only answers. As you say, referencing position makes little sense when answers may be moved around by voting.

If you want to reference another answer, the usual way to do it is with a link. During the private beta, that's a little harder than normal since the share link is removed from the interface. But when the site goes public you will be able to copy a direct link to the answer itself:

An amazing answer, in my opinion.

Then you can use Markdown's link syntax:

Building on [UnCleJack's answer](https://parenting.stackexchange.com/a/11828/4521)...

Depending on how much you want to add, you might be better off suggesting an edit to the answer instead. It's a judgement call, but the more you have to add, the more likely you should write a new answer.

When answering, I have two audiences in mind:

  1. The asker, and
  2. The future readers.

As a rule, the asker will be less and less important as time goes on. Many questions I've asked are no longer interesting to me, but continue to help future searchers. I try to answer starting from specifics and expanding to generalities:

In your situation, X should work well, but you might also consider Y and Z when similar problems arise.

That sort of thing.

Blogging turns out to be good preparation for answering on Stack Exchange. I sometimes pretend that the question was an email that I decided to answer publicly. Other people might be answering too, which is interesting. But my answer should stand on its own. Obviously it's critical to answer the question directly. But there's also room to address the bigger picture. Ultimately, we optimize for pearls:

Incoming questions are a universal constant, all around us in countless billions. But answers—truly brilliant, amazing, correct answers—are as rare as pearls. Thus, questions are merely the sand that produces the pearl.

So be careful not to let other answers distract you from creating timeless, complete, and expansive answers.

  • $\begingroup$ I quite disagree with the idea that "questions are merely the sand that produces the pearl." A few quotations. "“The formulation of a problem is often more essential than its solution, which may be merely a matter of mathematical or experimental skill. To raise new questions, new possibilities, to regard old questions from a new angle, requires creative imagination and marks real advances in science” - Einstein. $\endgroup$ Commented Apr 24, 2014 at 17:03
  • $\begingroup$ / “Often in great discoveries the most important thing is that a certain question is found. Envisaging, putting the productive question is often more important, often a greater achievement than solution of a set question” - Wertheimer / “In re mathematica ars proponendi pluris facienda est quam solvendi” (“In mathematics the art of asking questions is more valuable than solving problems”). - Cantor. [And so forth.] $\endgroup$ Commented Apr 24, 2014 at 17:04
  • $\begingroup$ @Benjamin Dickman: Questions, especially the well-framed variety, are critical to the pursuit of knowledge. I am, by nature, a socratic thinker, which means I love questions. But both questions and answers must serve their purposes: "Once you see that an answer is not serving its question anymore, it should be tossed away". Good questions are humble and don't much mind not being called pearls. ;-) $\endgroup$ Commented Apr 28, 2014 at 15:38

You must log in to answer this question.

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