I have solved a couple of questions myself in the past, and I think some of them are interesting to the public and will most likely appear in the future. One example for this is the question how to enable antialiasing in the Linux frontend, for which there is no native support right now. My question would now be whether posting these as a new question would be appropriate, and then immediately answer it.

3 Answers 3


If it's an interesting question, ask it. But give someone else a chance to answer it first -- you may learn something new.

  • 2
    I did exactly this with my question about adding columns and rows to a matrix. I thought I knew the best route but didn't know about ArrayFlatten.
    – Andy Ross
    Jan 18, 2012 at 22:33
  • @AndyRoss, that one showed up on SO, and some of the older regulars saw it. (I didn't have time to answer, otherwise I would have put it up.) But, it does illustrate a very good reason to post things you know the answer to, you may learn something new.
    – rcollyer
    Jan 19, 2012 at 2:41
  • If I had a question and an answer for it, I wouldn't necessarily want other people to start needlessly worrying about answering it -- I might want to provide the answer right away.
    – Andrew
    Mar 8, 2012 at 23:36

If your questions are genuinely interesting or intriguing problems, then posting self-answered questions is generally considered okay.

My only caution this early in a beta is with regard to asking questions simply to "seed" the site. Stack Exchange sites don't generally need to be seeded for the purpose of adding bulk. Take a look at this blog post:

Asking the First Questions

But if you're question is interesting, certainly… feel free to post.

  • The alternative would be saving the answer in a text file and posting it when it comes up, which doesn't make much difference for me.
    – David
    Jan 18, 2012 at 20:59
  • @David If it is a good question, post it now. Jan 18, 2012 at 21:00

Most of these answers say, "If it is interesting, ask it." But what do we mean by "interesting"? Or more appropriately, interesting to whom?

If we want only Mathematica experts to visit the site then we should probably ask questions only an expert would ask. I'm of the thinking that we probably want some frequently asked beginning and intermediate sorts of questions to draw new users to the site.

I say this for two reasons. 1: we don't want to scare people from asking simple questions. 2: We want to draw people in who are doing a web search.

When I started with Mathematica most of my problems dealt with simple list manipulation and importing data. I also found simple shorthand used in examples (e.g. #, /@, ,__, /., /;) maddeningly confusing.

I wouldn't really be interested in a question about these things now, but I bet many new users would be.

  • If you listen to the SE podcasts, the guys talk about "canonical answers". This is what we need for the simple, new user, type questions. Clear answers to clear questions that will turn up on every search engine when someone is having that specific problem/confusion.
    – Simon
    Jan 19, 2012 at 1:02
  • The question remains, should we ask those questions during the beta phase or wait until a newbie comes along to ask them? In the latter case, we will inevitably have to clean up the question to remove any confusions that the OP has...
    – Simon
    Jan 19, 2012 at 1:03
  • 1
    I would argue for experts writing the question for just the reason you provide, we won't have to clean it up. Also, if the question is phrased carefully, it is likely to generate more hits in searches.
    – Andy Ross
    Jan 19, 2012 at 1:08

You must log in to answer this question.

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