Tal Fishman--in his answer to a previous meta question--suggested that expert users take it upon themselves to post questions.

If others are like me, they may have ideas for questions but don't know enough about the topics to write a good question.

My hope for this thread is that expert users can come together to brainstorm ideas for questions, and those with enough expertise can craft and post a quality question to the main site.

It would probably be best to limit one question idea per post and to edit the post when it has been asked on the main site.

  • $\begingroup$ I agree with the essence of Robert's answer: just post and let the community work on improving your question. $\endgroup$ Sep 18, 2011 at 20:59
  • $\begingroup$ @Tal: I'm hesitant to post a question like, "Can you estimate total return for a bond index using only the yield?" If I did, I bet it would be closed, not improved. $\endgroup$ Sep 19, 2011 at 1:48
  • $\begingroup$ I strongly disagree. I personally would argue against closing such a question, and I can even posit a not-too-simple answer to that exact question. If you truly have only today's yield and NO OTHER INFORMATION, ok, that's kind of an impossible question, but if you have any other information, ratings, yield history, anything, that is a legitimately expert question IMO. $\endgroup$ Sep 19, 2011 at 2:13
  • $\begingroup$ @JoshuaUlrich That example question seems reasonable to me, provided you have some context in the question body to help people understand the motivation. The questions I close are almost universally asked by people outside the industry who want to know what software tools quants use. $\endgroup$ Sep 19, 2011 at 5:45

1 Answer 1


You should remember that this entire site is a collaborative effort: questions, answers, everything. When you have a bit of trouble wording a question, you should should be encouraged to go ahead post the question and and ask for help… on the main site.

There's no need to create a "question factory" in meta under the guise of "get your help here." There's nothing wrong with posting a question as a work in progress. That's a major role of this community; If you make a good faith effort to be as complete and clear as possible, folks here should be happy to help with the editing.

That's the message you should be sending your users.

Stack Exchange has a really strong wiki component to it. That functionality is already built into your site throughout… Use it. Posts should describe the problem and concepts as best you can. If you are having trouble, follow up in the comments to ask to clarification. If you get really stuck (not even knowing where to begin) feel free to ask for help as a meta post.

But don't draw that collaborative activity away from the main site into a meta thread where it will have less of an audience. A "question factory" might even set the wrong tone for that type of activity.

Does "Question Factory" = "Seeding?"

Not really. I know that's not what you are talking about here, but the meta post you linked started talking about asking experts to "take it upon themselves to post at least one question per week." That's not really a sustainable way to operate and grow this site. A question factory also sounds a bit like a "Someone should ask a question about … " thread, and that's not really what you want here.

Let me back up a bit — I have never been a big fan of "seeding the site." When a site starts struggling for questions, suggestions inevitably pop up for schemes to generate more questions just to pump up the numbers. But questions asked "just for the asking" will become boring and uninspired and people will notice that the author doesn’t really care about the answers. That's where the site starts to fall apart. I talk about this in the blog post: Asking the First Questions.

Turning the Site Itself into a "Question Factory"

Your post stumbles into an important issue that should be the end goal of your efforts. What will drive this site to fill with particularly intriguing questions is to remember (and remind others) to use this site in your day-to-day work. Quantitative Finance is a fascinating subject brimming with interesting little puzzles and problems you solve everyday. The tough part is reminding everyone to drop by and ASK those particularly intriguing, long-tailed questions as they occur. That's how this site will grow.

There IS a way to keep this site in that "top of mind awareness" — Use those social bookmarks!


It really is the best way to attract new users… by provoking a "Hey, I always wanted to know that!" response. But more importantly, it's a great way to lead by example. Users will imitate what they see on a site, and when you ask a great question, it will also trigger users to ask more questions like it — "Hey, that reminds me of a similar question!"

That's how you turn the site itself into a "Question Factory."

  • 1
    $\begingroup$ Great advice, thanks for your input. I wasn't suggesting that non-experts post questions to seed the site, I was asking the high rep users, who are presumably experts, to use the site more in their day to day work, exactly as you suggest. I and perhaps a few others have been doing that since my meta post, and I doubt most people would characterize those questions as "boring and uninspired." What's wrong with a little friendly competition to increase the amount of quality content? $\endgroup$ Sep 18, 2011 at 20:56
  • 1
    $\begingroup$ @Tal Fishman: Nothing wrong with a little competition. We love "friendly competition to increase the amount of quality content" as long as the it involves questions you actually have. You commented to "...use the the site more in their day to day work." That's the key point here. $\endgroup$ Sep 18, 2011 at 21:34
  • $\begingroup$ @Tal The key (as Robert put it) is for questions to represent actual problems that (as I always put it) working quants face. If a question doesn't represent an actual problem that a working quant faces, then it's unlikely to be of much use to anybody. $\endgroup$ Sep 19, 2011 at 5:51

You must log in to answer this question.

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