I'm not entirely sure how to put this, but there have been a number of questions that fail for two reasons:

  1. They are written without any appreciation for objectivity and/or have very poor grammar.
  2. They cannot be answered (i.e. they ask purely subjective or unknowable things).

I find the existing finance message boards to be awash in idiocy, and was really hoping that we could maintain very high standards for this site so that it would truly be expert friendly; to the point that we could have known professors contributing to the site in a meaningful way.

As such, I suggest that we propose a set of grammatical and logical standards for questions. Please make one primary point per answer. Our final set of "standards" (based on consensus voting) can then be referenced whenever a poor question arises.

(Of course, another option is to suggest that we just continue to grow organically and let people follow their own personal style; feel free to answer that way if you are so moved.)

Check out the FAQ, especially the section titled "What kind of questions should I not ask here?" This issue has come up before. –  Joel Spolsky Feb 8 '11 at 19:47
@Joel Thanks; I appreciate some of the recent closes. Feel free to close this one too, in that case: quant.stackexchange.com/questions/251/quantlib-in-industry –  Shane Feb 8 '11 at 19:50
add comment

4 Answers

Here are some points which apply to ALL Stack Exchange sites and are already well established principles that every site doesn't have to work out on its own:

  • Grammar and spelling can be fixed trivially... that's why there's an EDIT button. If something seems like a good question but is written in ESL, just hit Edit and fix it.

  • If a question is clearly being asked by someone who is not engaged in Quantitative Finance and doesn't really even quite know what that is, close it as off-topic, but leave behind a note explaining why so that other people can see and learn from your action. That should clear up the "why does my ATM card pay interest" type questions, lickety-split.

  • Also please check out the FAQ, especially the section entitled "What kind of questions should I not ask here?" We already have strict community principles throughout the Stack Exchange network that questions are supposed to be real and answerable in some way, not open-ended conversational bids.

I'm sure there are also going to be specific community norms that apply to the Quantitative Finance community, which you should discuss and we will add to the FAQ, but the meta/logical stuff is already covered and you shouldn't hesitate to edit, close, delete, and move questions that don't conform.

Here's a totally-on-topic question: "Do I still have the first dollar I ever made? Is it still at the bottom of my checking account?" Wondermark –  Joel Spolsky Feb 8 '11 at 20:02
Lickety-split, indeed. –  Shane Feb 8 '11 at 20:48
add comment

We can down-vote bad questions. Writing the prompts in an expert-friendly way will encourage experts.

And I believe it's within mods' power to rewrite for grammar.

It is within ALL our respective powers to re-write for grammar ;o) But I think this matter has been long ago put to rest. I enjoyed @Shane's observation about finding the existing finance message boards to be awash in idiocy and concur. I think that Quant SE has been successful in avoiding that fate. –  Feral Oink Jul 14 '12 at 21:05
add comment

Questions should not ask about things that are simply impossible to know. Questions should start with the possibility of correct answers.

As an example, this question on Quantlib asks about the "street cred" of the Quantlib C++ library. There is no mention for what this would entail, or how one would go about verifying it.

While "street cred" is a vague term, Dirk's answer shows exactly how to give an objective answer to a subjective question. And QuantLib is indeed a software package aimed at professionals, so it's within our realm. So I'm like to keep that question open. (It's the same issue as that sports betting question, where the question was bad only because the person asking didn't understand it, but potential answers could have been very good.) –  chrisaycock Feb 8 '11 at 16:15
@chrisaycock My 2 cents: that's reasonable, but we need to have some sense for guiding questioners toward asking good, well-defined questions. That's my hope with this thread. Maybe closing the question isn't the answer, but surely that question should be revised. Lastly, just because Quantlib is in scope, doesn't mean that any question related to it is acceptable. –  Shane Feb 8 '11 at 16:18
Chris, thanks for the tip of the hat and you raise a good point. The question is not out of scope for the site; the way it is formulated may be. –  Dirk Eddelbuettel Feb 8 '11 at 16:29
Other Stack Exchanges suffer from the "lack of answerability" -- but prior to asking, the question asker doesn't necessarily know if the question is answerable. I think the grammar and intelligibility criteria are fine, as well as specificity prompts. –  isomorphismes Feb 9 '11 at 1:16
I found the answers to "the street cred of QuantLib" very useful, as they contained links to other competing libraries. –  Contango Jun 3 '11 at 15:40
add comment

Could not agree more. That language was the reason I waited a day to reply to the question in question here. Also note that the original poster rapid-fired a few sloppy questions in a short time span.

add comment

You must log in to answer this question.

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