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.

Possible Duplicate:
How about hiding answer score for first few hours?

I think it would be useful to hide votes on existing answers for the first X minutes of a question's life. Early answers have a higher chance of being upvoted, so if you're ten minutes late to the show, your answer will be the last one displayed even if it's great.

Not everyone will check all answers to a question, especially if the first one already has some upvotes.

Also, there's another twist to it. Seeing an answer with 5+ votes, some will assume it's the best one without even looking at the others, due to herd mentality.

The main upside to this would be that others are not influenced by existing votes when they cast theirs. This doesn't have to occur for everyone (the OP could still see the votes in case he's looking for a quick solution, and answerers could see the votes on their own answers).

And... what if the answer is wrong? Indicating that it's wrong early is very, very good.

Of course it is, but then again, a wrong answer that appears to be right that has several upvotes is very, very damaging (especially if it reaches a score of 10+). From my experience, in C++, this is highly visible with undefined behaviour. An explanation for some piece of code comes along, which may make sense or even apply to a particular compiler, and be taken for granted, when, in fact, the answer is wrong.

share|improve this question

marked as duplicate by Pops, Brad Mace, yhw42, random Aug 11 '12 at 6:57

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

1  
related : meta.stackexchange.com/questions/9731/… –  kbok Aug 10 '12 at 19:28
13  
And... what if the answer is wrong? Indicating that it's wrong early is very, very good. –  animuson Aug 10 '12 at 19:30
4  
So maybe "+" and "-" signs instead of total count? –  Bartek Banachewicz Aug 10 '12 at 19:33
3  
@animuson Not saying I agree with the proposal, but you can still come to that conclusion on your own. You don't necessarily need to see the votes for that. As long as it's indicated to the OP that could suffice. –  Bart Aug 10 '12 at 19:34
4  
It's worth noting that answers that are fundamentally very wrong tend to have comments indicating that, as well as downvotes. Additionally, if this rule were implemented, I'm sure we'd see even more of it. –  Servy Aug 10 '12 at 19:43
    
I kinda liked this idea until I thought about the bad answers like @animuson stated. Bad answers deserve to be displayed as bad answers to prevent people wasting their time trying it out. For now +0 –  PeeHaa Aug 10 '12 at 20:24
    
@animuson see my edit. –  Luchian Grigore Aug 10 '12 at 20:30
    
    
@bemace it's different. This has nothing to do with fast answers, but with fast answers being upvoted or downvoted by a bunch of people just because they already have up/downvotes. –  Luchian Grigore Aug 10 '12 at 20:33
2  
@LuchianGrigore That's one of the underlying problems with FGITW. Someone posts a quick answer, gets an upvote or two, and then it always appears above a later, more comprehensive answer, which never gets read enough to be properly upvoted. While not identical problems, they are certainly [closely] related. –  Servy Aug 10 '12 at 21:01
    
The intent of this question is noble but I think it has been well discussed in the past. Is there really that much of a problem that needs solving? –  slugster Aug 10 '12 at 21:25
3  
This might be specific to me, but: I only upvote if I feel the post deserves more upvotes. If someone posts a correct, one-liner answer that gets ten quick upvotes, I don't add mine. Nor do I add my vote to Redditted posts that have a hundred upvotes. Without being able to see the upvotes, I wouldn't be able to make this distinction. –  Robert Harvey Aug 10 '12 at 22:05
    
@RobertHarvey I do the same. I'm sure most veteran users also. But those are a small minority IMO, and votes weigh the same. –  Luchian Grigore Aug 10 '12 at 22:06
    
In addition to the possible duplicate, this is very similar to Suspend Upvoting/Downvoting Within the Answer's First Five Minutes and Don't show the answers for the question in the first 2 minutes. –  Pops Aug 10 '12 at 22:20
    
@PopularDemand I failed to see a dupe so far. I'm not in favour of either of those two proposals... The scope for this one is completely different. –  Luchian Grigore Aug 10 '12 at 22:28

2 Answers 2

This is a great idea!

While similar proposals have been discussed (e.g., the discussion kbok mentioned), they're missing the crucial feature of always showing the questioner the votes.

StackOverflow is very good at answering questions fast, and this doesn't prevent that (unlike hiding the votes from everyone).

No one else visiting new questions actually needs an answer to that question. They're there to answer the question, vote on the question and answers, edit the question, etc. They don't need to know which answer the community thinks is best.

Temporarily hiding vote counts from all but the asker allows the community to better evaluate answers without herd-mentality problems, but still allows the asker to get their problem solved quickly.


As for quickly marking wrong answers, this shouldn't change anything. If an answer is obviously wrong, the community will downvote it regardless, so the only important case is an answer that's subtly wrong. Depending on downvotes to tip people off to this is a bad system. It only works if the first voters notice the problem. Unless everyone's in the habit of breaking out vote counts, a +3 answer looks the same as a +5/-2 answer. Subtly wrong answers are only effectively identified via comments whether votes are hidden or not.

share|improve this answer
1  
Very good point in the last paragraph. Also, taking into account that new users can't even see the balance between up/downvotes, it will just appear as +3 to them. –  Luchian Grigore Aug 10 '12 at 22:29

This actually started as a comment to blahdiblah but has turned into an answer. I think blahdiblah has missed something crucial in his analysis of why people go to specific questions. I frequently visit questions because they look interesting and I want to read it - I have no intention of answering or editing that question.

Hiding votes removes an important piece of context from the answer. I give answers an up vote if they deserve it - if it was an mediocre answer and was already at +10 I would be unlikely to give another upvote as it just isn't that good. Maybe a badly written but fundamentally correct answer is already at -3 when I read it - I want to see that information because it might be the most correct question there.

When I cast a vote I have my big boy pants on and I'm totally capable of making judgements myself. Please don't hide information from me because you think I need to be saved from it because it might lead me astray. When I write a specification or design I don't do it with some of the information withheld - in fact I dislike information being hidden. I don't follow the herd, I vote according to my own judgement.

One last thing - when I write an answer I want to see the vote count on it, and I want to see it all the time. It's my answer, I have the right to see that don't I? I also give permission for everyone else in Stack Overflow to see it, no matter whether it's positive or negative.

share|improve this answer

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