On Stack Overflow, I am noticing a trend of there being almost 10 answers to a question, with an accepted answer, and no one voting (not even the asker not voting for the accepted answer). What's happening?
What is meta?
- Any Stack Exchange user can ask a question
- The community provides support, votes on ideas, and reports bugs
- Your voice helps shape the way Stack Exchange operates
I think it is because the questions are not getting the views that they did back then and there's a strong correlation of views to votes.
Personally, with the above in mind, I know that climbing the SO ladder would take a bit of work providing great questions and answers. Although I'm familiar with many topics here, it's hard to compete with other well established users for getting those answers out there in the hopes that there will be enough people around to vote up without the fear of being accused of posting duplicates. Back in the day, that used to be common practice in many cases, now you're likely to get downvoted instead. As far as questions go, it's hard to ask a good/popular question that isn't a duplicate nowadays because many of them were already asked.
As a side-effect, I don't really browse Stack Overflow just to read questions and answers most of the time, I'm generally looking for unanswered questions hoping to be the first to answer. That means questions that already have answers might not get a view from me, a potential answerer or voter. Part of that is my fault admittedly but you can probably understand the dilemma. If I genuinely see a faster (or better) answer that I would have written myself, then I'd upvote it. Perhaps offer come comments to supplement the material if there were some important points not mentioned. But these answers doesn't get the votes unfortunately since there simply isn't enough people reading them.
Because of the solstice lunar eclipse.
Or just because a lot of folks take vacation around the end of the year.
Or maybe because of complicated sunspot patterns...
Anecdotally I have noticed this too, over the last 6 to 8 months, and i don't think it is related solely to the volume of questions. I think Jeff M has the answer:
People have changed their usage pattern. As the easy questions have all been asked a billion times and the questions have got harder and more specific, it is harder to build rep on the site these days, so users are browsing for questions where they have a chance of being singled out as the answer. I would wager that even Jon Skeet isn't getting as many votes on his new answers as he was 9 months ago, simply because users are browsing past the questions.
Maybe the solution is not another badge, but to award a small amount of rep (say 10 or 20) for using up all your votes. I'm sure the guys here can come up with some clever algorithmn for awarding the rep, like the user has to use the votes over a 6 to 12 hour time frame (eliminating from contention all those who use their votes willy nilly in their first hour on the site), and the votes have to be spread across a range of answerers and tags (once again avoiding those who just splurge the votes to get the rep).
Sometimes I invest a lot of time to make good answer - the complex one which describes the problem and possible solution deeply. Even such answers are in my opinion voted very poorly. I think problems are:
Getting Nice answer is probably the best achievement I will ever get on SO. But I'm still happy to answer questions and helping people - and I guess it is the purpose of the site.
Also I think that reputation is making even more reputation.
Maybe we need a nice badge for people who initiate enough 'close by exact duplicate' rituals.
I've noticed the trend too (anecdotal evidence, I know). A related problem is that a lot of questions which deserve to be closed stay at 3 or 4 close votes / idem for questions which ought to be migrated.
I'm wondering if the sheer number of sites that scrape SO data and repost it could be affecting the equation. When I do a google search on something, frequently a "clone" site like "efreedom" shows the answer before the stackoverflow answer in the pagerank.
How many people are just not getting to SO because they stumble across SO information elsewhere?
The hot questions get a lot of views.
With that out of the way, the rest of the questions get hits mostly from diligent SOers who scrounge through the questions or through Google.
Due to the specific nature of questions that get asked(a lot of common questions may have been covered by now on SO) most new questions aren't likely to be widely searched. This lends itself to two problems.
1) new questions may not get enough answers
2) Answered questions may not get enough views.
A question after being asked, stays on the front page for some time(On SO this time can be really short, due to volume of questions being asked or answered), which is enough for the moderators and diligent SOers to answer the question. The answering of the questions brings it back to the top of the homepage. Now if the question is interesting enough or if the answer given has something which can spark a debate, the question tends to get more answers and continues to stay on the homepage. If this goes on for long it gets into the hot questions list(?). Otherwise once the question is answered the tendency to re-answer decreases and the question disappears from the homepage. The same thing happens if no one answers the question.
The traffic which comes from Google is mostly looking for a solution. This user mostly does not have a SO account. The users that come to an unanswered question generally use the back button on the browser. (Rarely they may answer.) Since the user is not likely to have a SO account, the user is unable to upvote the question or any answers(if they land at an answered question).
Some of this Google traffic gets converted to SO users. A lot of questions thus get asked by first time SO users who may or may not register. Since they do not have any reputation, they cannot upvote answers even if they are satisfied.
I guess, that is the reason for fewer upvotes now that SO is maturing. However that is certainly not all bad. Users which find SO through Google and return often to the site, start asking more questions and thus earn reputation and the ability to upvote, which will balance the upvotes issue and other issues in the long run.
In the initial phase SO may have grown very quickly due to common questions. The next growth phase will be slower, but given SO's hold over most programming related search results, this phase will only strengthen it.
(Personal view.Off the top of my head, no reasearch, no sources.)
The problem may be with the sheer volume of answers.
(waits for other answers equally as valuable to get posted)