-10

When I am looking to new questions related to a given tag, it is a bit tedious to sort questions by reading all titles and have an idea if the question is for me, or if it is too easy (I don't really like to answer basic questions, but I truly think they deserve to be on SO) or too difficult for me.

I though it would be great to have some "question levels", like beginner - intermediate - high intermediate - expert, so we could directly focus on questions we are more likely interested by.

My feature request is near from how-about-a-difficulty-rating-for-questions, but is still different. In this post, the author proposed that the asker chooses the question difficulty. Like many answers, I don't think it would work since non-expert people are unlikely able to correctly evaluate the difficulty of their own questions.

Thus, I propose several more or less realistic systems:

  • A. Let moderators select a level for each questions.
    • Pros: Quick. Proper leveling.
    • Cons: More work for moderators.
  • B. Make a voting system to let SO users decide about the level, and take a kind of average.
    • Pros: The result comes from SO users wisdom.
    • Cons: Would be slow to have a proper classification for new questions.
  • C. Merge A and B: let users vote in order to help moderators' choice.
    • Pros: SO users participate to the classification. Maybe less work for moderators.
    • Cons: Still slow (might be even slower than B).1
  • D. Having a machine learning system to classify questions, according to the technical vocabulary used in the question (a bit like systems used by SPAM filters)
    • Pros: Immediate. Fewer work for moderators.
    • Cons: Efficient? Is it possible to determinate the difficulty of a question by just analyzing the used vocabulary? I don't think so...

What do you think? It is really possible to have a question level system which is both fast to decide a classification, reliable, and not overwhelming moderators?

  • 10
    Difficulty levels are relative for everyone, answerers included. You cannot rate a post on difficulty level because the level depends on the current observer, not on any objective metrics. I find Python questions easy. Don't ask me about Haskell. – Martijn Pieters Mar 12 '14 at 20:01
  • 4
    Letting moderators select a level for each question would be neither quick nor proper. – Bill the Lizard Mar 12 '14 at 20:18
  • 1
    @Bill I've heard moderating stack overflow doesn't take long, I'm sure you've got a couple of minutes to assess all questions :p – Richard Tingle Mar 12 '14 at 20:21
  • I am a bit surprised to see so much negative votes. May be I do not understand to purpose or the proper use of Meta. I though it could be used to share, propose ideas to make SO better. Downvoting does not mean one is not agree with the question, it should mean the question is inappropriate or badly asked. To be honest, I don't have this feeling with my question, and the non-constructive commentaries I get do not help me to understand what is wrong. – Florian Richoux Mar 12 '14 at 20:59
  • This is a common issue, I personally don't like the current system for this reason. Votes on meta are different, the downvotes mean people are voting against this idea not that it is badly written or you were wrong to ask it. Think of it like an referendum rather than usual stack exchange voting – Richard Tingle Mar 12 '14 at 21:01
  • @Richard Tingle Thanks! I understand now, and better accept these downvotes. – Florian Richoux Mar 12 '14 at 21:08
  • 1
    @FlorianRichoux The idea has been brought up quite a number of times before. You linked to one, but the "related" tab is full of others. The idea is not original, you have not done a lot of research on the topic. People get annoyed seeing the same suggestions proposed over and over. – Servy Mar 12 '14 at 22:36
5

The key problem here is that difficulty has two elements:

  • The question
  • The person answering it

Now lets assume you can perfectly gauge the question part, you can't account for the person answering it. To take an extreme example, I consider myself a reasonable Java developer, I am however not a poet. So I would consider the first question much easier than the second

  • How can I create a persistent file that represents objects in a binary form?

  • What is iambic pentameter?

A different person (in fact most people) would probably feel the opposite.

This is clearly an extreme example, but one persons easy question is another person's impossible question. As such such a measure isn't very helpful

  • +1 for making me search for Iambic pentameter. – Frédéric Hamidi Mar 12 '14 at 20:21
  • 1
    I see your point, but I disagree. Questions have an intrinsic difficulty, independent from the asker or ansewers. Let's take an example: [stackoverflow.com/questions/388242/… there is a list of beginner/intermediate/expert books. This ranking is independent from users C++ level; it is a fact. Classifying this or that book into the beginner or intermediate level can be arguable, and that's why a voting system is not a bad idea (to me). – Florian Richoux Mar 12 '14 at 21:04

You must log in to answer this question.

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