Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 158 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

I'm not really sure if there is a solution to this but it strikes me that 90% of the time you are going to get most of your points by answering questions from beginners. These tend to get mass answered and lots of upvotes occur. I see no major issue with that.

However when a complex question is posted you are unlikely to see a large number of up-votes. This means that those that answer easy questions will, on average, get more reputation points (I fully admit that the vast majority of my reputation points come from helping beginners).

Can anyone think of a way round this problem? The only option, really, would be to get more "experts" (I mean no offence to the plenty of experts there already are on here ;)) on board or to, somehow, encourage up-voting of the more niche questions.

After all it strikes me that you can gain a good reputation for helping beginners but the best area to build a reputation is in the more complex niche areas. Using stack overflow to demonstrate that you are a niche expert would, overall, make stack over flow a far more useful tool for things like job hunting.

Can anyone think of a solution to this? Is there, even, a solution to this?

Edit: To clarify, I have no issues with helping beginners and do so myself a fair bit. My point is that it seems to me for the more niche areas the system should be weighted somehow. Otherwise the system encourages only helping beginners and not getting your teeth stuck into the more complex problems. I regularly see questions left unanswered that someone with experience really ought to be able to answer. As much as I can I try to answer these questions and get my 10 rep points + 15 for having my answer accepted. However If I answer a more beginner question I can expect to see significantly more rep points. This bias towards beginner questions then makes advanced users less inclined to ask and answer questions. As a result, the usefulness of stack overflow to the more advanced users is diminished. I would have thought it would be good to keep stack overflow useful for beginners AND make it useful for more advanced users. Otherwise it will end up being seen as a beginner's resource.

share|improve this question
Nearly a duplicate of… , especially the sort of answers that will be coming, even if the question is somewhat differently put. – Mark Hurd May 19 '10 at 10:36
More than differently put. It's related, but not dupe. – D_N May 19 '10 at 10:43
This is an old observation, and it is true, but so what? I don't consider it a bug, per se, just another example of the bike shed effect at work. – dmckee May 19 '10 at 13:33
I didn't suggest it was a bug I just suggest it may drive away some of the people you most want on here ... – Goz May 19 '10 at 16:26

12 Answers 12

Easier questions draw more attention because they are understood by a larger number of users. This is just a fact, and there is nothing necessarily wrong with it. Voting is not a way of measuring intelligence, it is just a way of measuring the usefulness of a question or answer. Answers to broadly appealing questions are, in a way, more useful than answers to narrowly targeted challenges.

share|improve this answer
Accepted. However it does seem to then draw the site towards being an incredibly useful resource for beginners and NOT a useful resource for advanced users (which it should be as well, IMO). – Goz May 19 '10 at 12:30
Add subjective and poll questions to that disproportionate attention group. – Gnome May 19 '10 at 13:42

Isn't the entire point of our profession to help beginners, so they become competent programmers and not fodder for The Daily WTF? The kind who we have to clean up after all the time?

Weren't you a beginner once? I know I was. In my opinion, you have to take the long view here.

That said, we don't tolerate abuse of the system, so feel free to flag posts by so called "beginners" who are putting in no effort whatsoever and failing to improve in any way over time.

share|improve this answer
Well I've edited my original question because I think you misunderstand the point im trying to make. – Goz May 19 '10 at 12:28
I think he's saying that questions where the answer could be found with 5 seconds of Googling get the most points. You're spoon-feeding them, not teaching them to feed themselves. I agree however, that changing the system would drive the noobs to TDWTF producers. – Chris S May 19 '10 at 13:21
At what point do we as a community try to collectively get past merely competent? Is that outside the scope or vision for SO? – SnOrfus May 19 '10 at 18:44
@chris see the last paragraph of my answer. You have a flag link -- use it! – Jeff Atwood May 21 '10 at 8:47

On the one hand, you are obviously correct, helping beginners will generate a lot more reputation than helping out on the niche questions.

On the other hand. The niche questions do not help as many people. This is a community site, so the goal is to help a lot of people. If I have very peculiar and unusual questions, then the person who devotes 60 minutes of their time to help me will probably only be helping me, but helping out the newbie with a stellar answer will help that person PLUS all the other newbies who have the same question and arrive at your answer. So, the reward system is designed towards helping the most people possible.

This makes it sound like there is no incentive to help people who have tough questions, but the incentive Does exist. There are people who are only here for reputation, those people will stick to beginner questions, and will do quite well from it. There are people who are here for challenging questions, and there are people who are here to help. Those last two groups will be willing to provide thoughtful and helpful answers regardless of the 'reputation reward' that they may or may not get, they are only chasing the green check, and the satisfaction of solving something unsolvable.

In those cases, having a niche or difficult question is a plus, because you pique their interest.

share|improve this answer
Very well answered. Also, a point to make is that with so many people answering the "beginner" questions, it can be tough to compete. I rarely find beginner questions with no answers, so I often spend my time looking for more challenging stuff. – The Unhandled Exception May 19 '10 at 12:50
This assumes all beginner questions are general, which is false. From what I've seen, they are, on the whole, just as specific as non-beginner questions, and are often better when they add more details. – Gnome May 19 '10 at 13:40
Most beginner questions are about specific errors that are related to general mistakes. Or general errors that are related to specific lines of problematic code. These are useful to every one who has a "seg fault" in their code, at the very least as things to rule out. The opposite is true for a complicated, intricate, and large (well architected) code-base that has a specific error relating to two obscure interfaces that are failing to interact correctly because of the customized framework they are using. – devinb May 19 '10 at 14:07

I think the best thing you can do is choose your votes carefully. I make a point of saving my upvotes for worthy answers that have had time and effort put into them. It's probably fair to say that all of us receive a good percentage of our votes from questions that can be answered with one line of code or a very short explanation.

How users vote is entirely up to them, most will splash their votes around on the questions that they can understand the answer to. Questions that require a higher level of expertise to answer won't generate as much interest.

One potential solution might be a "deserves more votes" flag of some kind, which flags that question/answer so that it attracts more experts to agree with it but that sounds borderline pointless to me and, given the jokey cruelness expressed by most meta users, asking for extra upvotes (even when it's not your own post) generally backfires with extra downvotes.

share|improve this answer

It's funny, I was thinking about the exact same "issue" yesterday, after I answered two questions. One was a page long answer, complete with code, a full explanation and ascii diagrams which earned me 25 rep. The other was literally a half line answer that earned me fully 100 points more at 125 rep.

I agree with your statement that there's nothing inherently wrong with getting big rep for helping the beginners, but it does seem unfortunate that the big effort or niche answers don't always get the recognition they deserve. However, I also think that its an unsolvable problem since there's no way to tie effort/value to the wisdom of crowds (which is inherently tied to the number of views a question gets), so the only way to reconcile that fact is to accept that rep doesn't reflect effort/skill/knowledge or to just try & ignore rep altogether.

For me personally, I don't particularly care about rep (ok maybe a little) and actually get much more of a kick out of knowing that I've helped someone solve a difficult problem than I get out of knowing that I've helped someone avoid typing their question into google (even though the rep shows the opposite).

Also, one other observation: It has been my experience that while the "easy" answers may give you a quick burst of rep, the more complex/thorough answers seem to attract more rep that slowly trickles in over time as people stumble across them via searches. It doesn't completely balance out, but its probably as good as its going to get.

share|improve this answer
Welcome to the internet (or is it the M-TV culture) of 2010. I tend to prefer writing longer, more thoughtful, more composed answers than your "half line". Some of the most interesting - to me - questions require some research. They almost invariably get zero or one vote for I assume (condescension alert!) that they don't get read. "tl;dr" was not invented for no reason (and yes, I have a really bright kid with the seeming attention span of a gnat, "O brave new world!") – msw Aug 15 '10 at 14:50
@msw - Sorry I only got as far as "M-TV" before I got bored and stopped reading. :P (for the record, I totally agree with you and if you look at my post history, the majority of my answers are on the tl;dr side of things and my half line example is by far the exception to the rule). – Alconja Aug 16 '10 at 1:19

One solution might be to have periods when only questions from people with over 10k rep get displayed on the front page. After all, they're the experts as determined by the community (hah!) and so must be asking the best questions?

OTOH, I can't think of anything more calculated to make new users go away. So it's actually a terrible idea, and I actually suggest that no such thing should ever be done.

share|improve this answer
I'm not suggesting that beginner users go away. Im trying to solicit suggestions on how to stop ADVANCED users going away. – Goz May 19 '10 at 12:28
I agree that you should NOT do this, but what if there was a tab for viewing unanswered questions by high-rep users only? – whybird May 21 '10 at 3:28
@whybird: So long as it is not the default view, that ought to be no problem. – Donal Fellows May 21 '10 at 12:36

Can anyone think of a way round this problem?

Yes: ignore reputation.

The focus of the site should be:

  • Posing good/well-formed/interesting questions, and
  • Providing helpful/complete answers
share|improve this answer
hahaha thats so true. However the site's owners, with their job site, seem to be trying to push the whole reputation thing as a sign of how good you are to potential employers and other users :) – Goz May 19 '10 at 16:26

I think you have it wrong here. This system doesn't favor beginner questions per se, though that may seem to often be the case. This system favors what's popular with the community.

Niche questions aren't automatically difficult. Some questions are simply for an obscure technology or an edge case which might be easy to solve with some specialized knowledge of the situation. On the other hand difficult questions aren't always passed over or ignored. Code golf questions would be a challenge for a beginner but see a fair amount of activity.

This is what you get with a system largely run by the community. However, maybe the community aspect could be utilized to set up some system where people with a certain amount of rep could vote to mark a question as an "expert" question or "difficult" which might net more rep for getting the marked answer. Or maybe the asker themselves could mark it as such but have it cost rep so that not everyone feels free to mark every question as such.

Of course, I'd also love to see questions that go a long time without answers have increased rep given to people who find the question and answer it.

share|improve this answer
To your last point .. yeah that would certainly be interesting but I guess thats what bounties are about. Thouh it does then penalise the person asking the hard question. – Goz May 19 '10 at 18:39

I would think that the Bounty system would at least slightly balance this out - questions that are niche, and/or hard, might be more likely to have a Bounty offered.

share|improve this answer

I agree that in some ways this is a feature, not a bug, but if you did want to address it you could use the (un)popularity of the various tags attached to weight how much an upvote or accept is worth.

Let me be the first to say that at first blush this solution could probably lead to more chances for those so inclined to game the system, and unnecessary complexity.

share|improve this answer
Yeah I agree it could lead to more problems. I am jsut making an observation for discussion tbh :) – Goz May 23 '10 at 19:49

For questions viewed less than the x %tile increase the value of accepted and up voted answers.

share|improve this answer

Let's be honest, points and badges aren't really worth anything in the real world, so whatever scheme you come up with is not going to be a strong motivator for the really experienced people potential answerers who are already paid to do real work. I don't believe that tweaking the reward system is going to change the nature of the site to any significant degree.

share|improve this answer
Au contraire. The site is driven by the points system. The whole community is defined by the system. Changes to it are changes to the community. – D_N May 19 '10 at 23:49
So your assertion is that all people who participate in this community are primarily motivated by earning points and badges? Give me a break. – Christopher Barber May 20 '10 at 13:50
Also, with, points and badges ARE worth something in the real world, at least in some cases. – whybird May 21 '10 at 3:30
Not much. I wouldn't pay much attention to SO badges when hiring someone for a programming job, nor would I be likely to look for jobs through this site. My point is that any point system you come up with is only going to provide motivation for a subset of the people who you want asking and answering questions. – Christopher Barber May 21 '10 at 13:52
Was that my assertion? Hm. – D_N May 23 '10 at 3:15
Your assertion was actually that the "community is defined by the system". Do you really believe that? If you ask me, the community is defined by those who choose to participate. The system only defines the community if you believe that it is the only reason that people choose to come here. If not, then the system is only a partial determinant of the community at best. – Christopher Barber May 23 '10 at 23:59

You must log in to answer this question.

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