What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 122 Stack Exchange communities.

I would like to revive this feature request: Sort questions by user reputation

On stack overflow, I like to browse questions tagged "matlab" and be helpful when I can. Unfortunately, this tag seems to attract an inordinate number of "how do I solve this specific homework question" , "here's a page of incomprehensible code -- why doesn't it work" , and "how do I do computer vision -- please advise" for every real question asked. These questions are typically asked by "user12345" types with single digit reputations.

The downvote system doesn't really work well here; it's not worth it to spend time downvoting questions from users who clearly have no investment in their own reputations.

The number of spam questions makes it hard to identify legitimate questions. Would it be possible to set an option to suppress questions from users with a reputation below a certain threshold? I recognize this might be seen as discouraging new members from joining the site and seeking help, but the option would be turned off by default anyone who wants to help out and sort through junk questions to find good ones from brand new users would just leave it off.

EDIT

Alternately, would it be possible to ignore questions from users who have not been members for at least one week. I think it's fairly naive to think you can't at all judge the quality of a question based on the person asking the question.

share|improve this question
8  
What's wrong with the matlab unanswered tab? If there are bad questions at the top of that list, please vote them down. –  Bill the Lizard Feb 10 '12 at 19:05
7  
I would respond to this but the guy who asked it only has 88 rep... –  JNK Feb 10 '12 at 20:38
1  
there should be a badge for 10 downvotes! –  Learner Feb 10 '12 at 20:38
1  
I use this little app I threw together to filter questions in a very similar manner -- I have no scruples, I know :( –  Marvin Pinto Feb 10 '12 at 21:50
    
"The downvote system doesn't really work well here; " really you'd look at a -1 question? –  Some Helpful Commenter Feb 10 '12 at 23:06
    
@JNK's MetAccount - even 80 rep would put you in the top 10% of people asking questions tagged matlab. –  Marc Feb 14 '12 at 15:23
    
@Bill the Lizard-the unanswered tab doesn't show questions that have no accepted answer. –  Marc Feb 14 '12 at 15:25
    
@Marc I'm not sure what you mean. That's what the Unanswered tab is for. It shows questions that have no answers at all and questions that don't have any upvoted/accepted answers. They're ordered by votes, so the best unanswered questions should be at the top. –  Bill the Lizard Feb 14 '12 at 16:13
    
@Bill the Lizard - the sort order appears to be by number of votes, then by whether the question has any answer at all, which means that for all intents and purposes (on the MATLAB tag at least, where most questions get only a few upvotes; 4 is a really good question), these questions get buried in the middle of a heap. Again, I don't think the upvote/downvote system is working very well for the MATLAB tag; there are simply too few people voting, so the noise swamps the signal. –  Marc Feb 14 '12 at 23:23
add comment

2 Answers

Vent your frustration by down voting and leaving comments on such questions. To improve the quality of questions on this community, you have to play your part too.

I don't think that creating a view without questions from low rep or new pple is a good idea as this would encourage high rep users to totally ignore them rather than act on them(close/edit/down vote/leave comment) and that would end up declining the community quality

share|improve this answer
4  
Indeed, we want to avoid any potential of "elitism" or "cliques" in Stack Overflow. If the noise to signal ratio is too high, help lower it (or suggest ways to help automatically lower it). Locally filtering with extreme prejudice isn't the answer. –  David Feb 10 '12 at 19:15
2  
@moguzalp: en.wikipedia.org/wiki/Clique –  David Feb 10 '12 at 19:22
3  
@moguzalp: "These two words explains many issues on stack and meta." - Do you meant to say that you're observing this behavior? If that's the case, we're certainly not intending it. –  David Feb 10 '12 at 19:47
add comment

You became a member of SO on 5 March 2010 @ 16:39, and asked your first question 11 minutes later (@ 16:50). It was answered within two hours.

Would it have been better for you if the person who did answer your question instead never saw it because they'd ignored it via a flag?

share|improve this answer
13  
Ah, the magical effect of the mirror. –  Toon Krijthe Feb 10 '12 at 19:10
1  
+1 Nice answer eheh... Where do you see the exact registration time? –  Alenanno Feb 10 '12 at 19:33
    
Nevermind, I found it... (Hover the mouse on the time showed for "Member for".) –  Alenanno Feb 10 '12 at 19:35
    
Everyone chooses what questions to answer. Sorting options enable you to find questions which you feel you can helpfully answer. No one would be forced to use the option to ignore new users. What's the logical difference between your construct and saying "would it have been better for you if the person who did answer your question never saw it because they'd ignored because it was tagged matlab?" –  Marc Feb 14 '12 at 15:20
    
@Marc - What I'm saying is that, if the answerer of your question had decided they wanted to ignore questions from new and/or low-rep users (i.e. you at the time), you may not have ever received the (very, very excellent) answer that you did. My response was primarily to your "alternative" suggestion (for ignoring new users), but also tangentially applies to sorting by reputation. If the answerer had been sorting by rep, he might not have made it to the bottom of the list (where you would have been at the time). –  Rob Hruska Feb 14 '12 at 15:57
    
@Marc - To answer your "what's the logical difference" comment: Someone can choose to ignore [matlab] tags because they simply don't want to answer [matlab] questions (perhaps they don't have the experience or knowledge, and don't want them cluttering their Interesting Questions tab). That makes sense. But choosing to ignore someone just because they're "new here" is prejudicial. It has everything to do with (what little we know about) them and not the content of the question. Maybe I'm naïve, but I don't think newness is a reliable indicator of question quality. –  Rob Hruska Feb 14 '12 at 16:06
    
@Marc - (to continue blabbering on about this :) - Both you and I were new here once and we both manage to ask clear, competent questions. There may be correlation between new users and poor questions, but that doesn't imply causation. By allowing people to ignore new users, we'd be doing an unjust disservice to the new users who are asking good questions. –  Rob Hruska Feb 14 '12 at 16:13
    
here are examples of questions I would prefer to ignore. stackoverflow.com/questions/9284836/… and stackoverflow.com/questions/9284483/plate-number-recognition . instead, per suggestions here, I have downvoted them. I think that's probably more discouraging to a new user than just not having the question answered. Due to the high volume of spam-like questions on the matlab tag, it's not feasable to downvote and comment on every bad question, so I just downvoted without comment. –  Marc Feb 14 '12 at 23:25
add comment

You must log in to answer this question.

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