7
$\begingroup$

We have seen some questions where, after an initial struggle, the poser elects to initiate a new thread in the guise of a new question. For example, from today alone, we have

Developing a Statistical Test to Ascertain a Better Fit and “Normalized” standard deviation;

Threshold for Fisher linear classifier and Calculating the error of Bayes classifier analytically.

There are pros and cons for this approach. One argument in favor is that sometimes a large body of eventually irrelevant comments accumulates, so starting off with a clean question might be clearer. An opposing argument is that this can create a complex trail of closely related or essentially identical questions. What are best practices in this regard, how are (or should they be) implemented, and by whom?

| |
$\endgroup$
7
$\begingroup$

To be honest, I'm not really sure (hence the delay in this answer).

I would suggest:

  1. Discourage the opening of new questions.
    • If their are a large number of comments, well don't worry about it. The page only shows the top 5 anyway.
  2. If a new question is opened, then:
    • delete the old question.
    • don't flame the questioner - they're probably new here. A simple comment stating that we discourage opening new questions, when the old question should have just been reworded.
| |
$\endgroup$
  • 1
    $\begingroup$ +1 I agree with this. $\endgroup$ – Shane Nov 29 '10 at 21:20
  • 3
    $\begingroup$ I also agree but I just want to point that there may be some cases where editing the question would change drastically its original meaning so that responses that were already given (and were correct in the context of the original question) would lose their relevance. And I doubt plain readers would take time to track all of these changes. $\endgroup$ – chl Nov 29 '10 at 23:39
  • $\begingroup$ Although I have accepted this answer, @chl makes such an excellent comment that I doubt our conversation about this topic is really closed: it's just suspended until the issue is eventually forced on us again. $\endgroup$ – whuber Dec 1 '10 at 22:31
  • $\begingroup$ @whube I agree. I suppose that any question that falls in this category will be awkward to deal with. $\endgroup$ – csgillespie Dec 2 '10 at 9:10

You must log in to answer this question.

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