Recently, i was stuck with a problem so i ask a question here hoping someone can help me. I was suprised in a cool way thanks to 2 sweets answears. Both of them help me to figure out the theoritical aspects and an easy way to solve my problem.

Thank to the voting system an answear lead the race of visibility but i consider that both answears are as important as each other for a global understanding, theory and application.

Why not distinguish theoritical answear from pratical ones and let the user choose what he prefers to focus on ?

Moreover even if we stick to one domain. Multiple answear can appear given different options to understand/solve each one with a specific trade off.

To solve this i propose to let contributors specify if there answears are more theory/solve/both oriented by adding topics. And let to everybody the ability to reward all answears by specifying which part of his specific problem this answear help to solve.

Then we will be able to cluster answear by topic with a far better granularity in order to allow efficient searchs to all our problems.

Hope it can help !

  • $\begingroup$ 1) Applying subjective labels to posts would create more problems than it'd solve; see also the whole homework discussion. 2) You can always hand out bounties to reward answerers. Well, if you have enough rep. ;) $\endgroup$
    – Raphael Mod
    Jan 31 '17 at 18:05
  • 3
    $\begingroup$ I don't understand this question. What do you mean by “1-NN strategy”? If there are several good answers, you can upvote all of them. Why do you think another mechanism is needed? $\endgroup$ Jan 31 '17 at 22:27
  • 2
    $\begingroup$ If you are only looking for a theoretical answer or a practical answer, you can specify in your question which kind of answer you are looking for. If you are interested in both kinds of answers, you can upvote both answers. I understand that you can only accept one answer, so you'll have to pick which one you found most useful to you. As for searching, my advice is to search for all posts that are relevant to your question, including both theoretical posts and practical posts, so the clustering you mention might not be useful enough to warrant the disadvantages of your proposal. $\endgroup$
    – D.W. Mod
    Jan 31 '17 at 22:36

You must log in to answer this question.

Browse other questions tagged .