What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 128 Stack Exchange communities.

Yesterday I submitted an answer to this question. I'll use it as an example to demonstrate why down-voting requirements should be improved. That question's quality is not the greatest, but I'm going to talk about something other than that anyway.

The question is about something most people don't understand well (or even don't understand at all). I can't say that I understand this all too well either, the answer I gave is something that I found somewhere else. So I want to talk about evolving of votes on the question, and my given answer. Both were going down until another person helped me, then my answer was acceted and upvotes started to flow.

This situation confirmed my idea that in some cases down-vote means that the person who is casting it has no idea about the specific topic. Also, up-votes means something worthless, but for now I'm not concerned with that.


So here I start with my suggestion of what should be added to have the possibility to down-vote a question or answer.

Originally, the question was tagged with: C, C++, algorithm, integer-division. I edited that and retagged that question, so my suggestion is connected with tags.

As every question can be tagged with 5 tags, I suggest that the person who casts a down-vote should have some amount of reputation earned from those tags. At the least this means that down-voter has some knowledge about the topic he's down-voting.

However, I understand that it is not always a good idea, due to certain reasons. Let's look at that question again; it is tagged with C and C++, lots of users have some reputation on these tags. Also, for example, if the question is tagged with unpopular tags, not many users could cast down-votes on very low quality and even dangerous posts.


It is just my opinion and my advice is not always the best. I just wanted to draw attention to this type of problem, and maybe someone will find some better and acceptable solution.

share|improve this question
    
What do you mean by "requirements for down-voting"? –  BoltClock's a Unicorn Sep 10 '13 at 16:01
8  
You seem to be arguing against yourself. While i'm usually a big fan of that, in this case I'm confused as to exactly what you're proposing –  Richard Tingle Sep 10 '13 at 16:02
    
@BoltClock'saUnicorn Now user needs 125 reputation to cast down vote, it is the requirement. I think it should be improved. –  123456789 Sep 10 '13 at 16:03
    
@user2623967 To what? –  Servy Sep 10 '13 at 16:03
    
@RichardTingle I'm not arguing against myself, I posted this because I see some kind of problem in SO, but I don't have good solution, so my idea is just a base and some erson may evolve it in some good solution. –  123456789 Sep 10 '13 at 16:04
18  
As for the whole "only downvoting posts with tags your active in" idea, that gets a big no from me. I recognize drivel even in plenty of tags I don't participate in. Bad questions don't always require me to know about the subject to recognize them. –  Bart Sep 10 '13 at 16:07
6  
not to mention the fact that it will just encourage people to tag questions with more general language-agnostic tags just so that they can be downvoted. –  Servy Sep 10 '13 at 16:08
    
@LBT I understand your point. There is one of problems that I even myself see in my suggestion, so as I said, I posted is as a base idea of something. And maybe someone will advice how my offer should be improved. –  123456789 Sep 10 '13 at 16:10
    
I don't understand the first part. Someone helped you how exactly? Did you edit your question following comments? (A link would help.) –  Mat Sep 10 '13 at 16:11
    
I think you have got it right with the downsides and the current system is as good as its going to get. Seeing problems with a post is often easy and does not require language level knowledge –  Richard Tingle Sep 10 '13 at 16:11
    
@Mat I added one important stuff because of comments, this morning someone deleted all of them, because part of them was useless and slightly offensive, but together deleted useful-ones. –  123456789 Sep 10 '13 at 16:14
3  
Do I understand this right: initially, your answer was, at least, incomplete. And when you fixed that/made it complete, votes changed from negative to positive? If that's the case, I don't see a problem at all. I see things working as they should. –  Mat Sep 10 '13 at 16:16
    
@Mat my answer didn't had strong proof that it is right. The comment, that helped me was something like: "I cannot believe myself, but your way is better" also he added a link, I added into my answer. So my answer wasn't wrong. –  123456789 Sep 10 '13 at 16:18
3  
If it wasn't wrong, but didn't explain or somehow demonstrate (even with only a link to a more complete explanation) that it was right, then it wasn't a very useful answer. People would be inclined to mistrust it if even you don't appear to know why what you said was true and it isn't evident from your answer. –  Mat Sep 10 '13 at 16:22
    
...down-votes on very low quality and even dangerous posts.. How will it be dangerous? Is there a chance it will come and bite me? –  Michael Sep 10 '13 at 21:26

3 Answers 3

This situation confirmed my idea that in some cases down-vote means that person who is casting it has no idea about the specific stuff.

The basis of your argument and subsequent request is flawed. How could this situation possibly have confirmed any idea? Voting is anonymous, so you haven't a clue as to whether those who voted on your question have "sufficient" experience or not. Beyond that, you can't even say, with any credibility, why those users voted the way they did.

As every question can be tagged with 5 tags, I offer that person who casts down-vote should have some amount of reputation earned from those tags, it at-least means that down-voter has some knowledge about the stuff he down-voting.

What would this solve? Just because someone has asked a question with a given tag (and thus, theoretically, earned reputation for that tag) doesn't mean that the person knows more than others on the subject.

The following is very important to keep in mind:

reputation ≠ expertise

Reputation only (very roughly) measures a user's experience with Stack Overflow, not their experience or expertise in a particular technology.

Beyond that, some questions are obviously "bad". I know next to nothing about Python, but I know that asking about a good IDE for Python, or recommendations for books to read, is off-topic and should be downvoted and closed. Just like I know that "answers" to questions suggesting that a new Chanel bag will make me look sexy and for only $17.99 should be downvoted and flagged.


In this particular case, from the comments on the question, it would appear that your answer may have attracted downvotes from knowledgable users who disagreed with your "solution" because it violated some (in their opinions) best practices. Users expressed their disagreement with your approach (though there is no way to say for sure that these are the same users who downvoted your answer):

Others with C++ experience disagreed:

I think you posted a solution which was contentious among other users familiar with C++. Your theory that the downvoters were ignorant doesn't fit the evidence. I've seen cases of voting based on best practices before and documented it. For these kinds of contentious answers, the best way to avoid the downvotes seems to be a disclaimer of some sort. In this case, you could have simply said "In many cases the compiler selects the most efficient algorithm for you. Barring that, you could try this..."

There may have also been some confounding variables involved, including a possible (since-deleted) similar answer which, apparently, did not work:

share|improve this answer

I personally prefer it when you can up-vote or down-vote for whatever reason you see fit.

What likely happened is that the down-votes came first because they came from impatient people who didn't want to bother parsing your question, or people who immediately didn't take your question seriously. (it is a bit awkwardly worded)

The up-votes might have came from people who did take the question seriously, and gave it more consideration, and then took more time to think about the question before voting.

At least that's my theory.

I definitely have seen a trend of down-votes on questions followed by up-votes. It's not always a matter of the question being outside of the depth of the voter.

share|improve this answer
1  
Since he has admitted to improving the question, it seems the downvotes were because the content was poorly explained and not backed up with explanations, proof of correctness, etc. to go along with a code dump. When the post was improved to include an explanation of why the code works, included a proof of it's effectiveness, etc. then the post itself went from something that should be downvoted to something that should be upvoted. The voters didn't do anything wrong. In fact, by being downvoted, they encoruaged the author to improve, thus they were helping. –  Servy Sep 10 '13 at 16:31
    
I wish I got a nickle everytime someone on Meta claimed to see a voting trend. –  JDB Sep 10 '13 at 17:11
    
The OP posted an answer not the question. –  Martin Smith Sep 10 '13 at 17:35
1  
@MartinSmith Right. I meant to say answer. Just pretend I said answer everywhere I said question in my previous comment. –  Servy Sep 10 '13 at 17:36
    
@Servy - That's only one one pretense required. –  Martin Smith Sep 10 '13 at 17:38

People vote based on the content of what they see.

Your answer could be improved. You said:

This way works because 2^32-1 (max 32-bit value) is divisible of 15, however if you take, for example 7, it would look like working, but wouldn't work in all cases.

EDIT: See this, it proves that this solution (on some compilers) is faster then module.

Three things:

  1. You said something would look like it would work but wouldn't work in all cases. But you did not say anything about what the cases were.

  2. Then you posted a link that you said had material that proved your point. You probably should have summarized the important material in your answer, and left the link to show where you got the material.

  3. I think link to be better post showing your point would have helped. Here's one right on our site: Should I use multiplication or division?

I think the question got up-voted later because people liked your answer. The question didn't show any effort by the questioner to try anything or do research.

share|improve this answer

You must log in to answer this question.

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