84

Recently I answered two questions on SO:
The first is a completely trivial answer to a really basic question (kind of "read the manual and follow the one liner example there").
On the other hand, my second is a much more complicated answer that requires much more specialized knowledge of a very specific domain (happens to be related to my PhD research).

Clearly, the first answer could have been given by many SO users - I just happened to be the first to pick up the glove on this question. However, this is not the case for the second answer.

In contrast, when it comes to reputation - the first answer instantly got upvoted (everyone can relate to a simple question/answer), while the second one - I don't believe it will get much upvotes...

So, life isn't fair: for the one answer that really very few people could provide (the second answer) and is much harder to find and get -- is not rewarded, while the trivial answer is rewarded fast and strong...

How can we amend this? How can we encourage rare and specialized answers for difficult and deep questions?

Do we want SO to turn into a pool of trivial and superficial questions?


Update: Thanks to many meta-users who took the time to read through this question and ponder its implication, both answers I used as examples got exposed far beyond their fair-share.
So, despite the fact that my original claim that the specialized answer get less reps is no longer valid, this does not change fact that in general my claim is stil valid: specialized answers requiring expert knowledge will recieve less attention at first.


Bottom line: (following my wish to have some sort of conclusions to meta discussions).
Indeed it sometimes seems like life isn't fair, but this is only in the short term. In the long run high quality answers do get their proper credit and exposure and their fair share of reputation.
The reputation system of SO works quite well balancing and accounting for many factors and parameters. It would require a really strong argument backed up with serious research to change this mechanism.

Thank you all for the lively discussion and for showing me that life is fair after all.

38
  • 82
    News flash: life isn't fair. Commented Oct 23, 2013 at 6:54
  • 25
    The number of people able to understand whether an image processing algorithm is appropriate is always going to be a little lower... but: yes, life isn't fair. I don't propose we can fix that. Commented Oct 23, 2013 at 6:55
  • 4
    Granted, those moments when you put a lot of effort into a post and do end up getting a good amount of upvotes feel good. Commented Oct 23, 2013 at 6:57
  • 2
    @BoltClock'saUnicorn I know life isn't fair. But can we make them a little better?
    – Shai
    Commented Oct 23, 2013 at 6:58
  • 4
    @Telkitty猫咪咪 - this is exactly the points: getting "easy" reps on a trivial answer feels wrong, especially when contrasted with zero reps on an extremely specialized answer...
    – Shai
    Commented Oct 23, 2013 at 7:00
  • 19
    Don't sweat it. Life isn't fair, really! Just bask in the glory of every upvote you do get for the specialized question, because that means that the few people that understand the subject thought your answer was good. And those upvotes count a lot heavier than the popular easy answer votes. Or, at least, that is how I feel about such answers. Commented Oct 23, 2013 at 7:00
  • 7
    @Shai: See? Life isn't fair. (I'll shut up now.) Commented Oct 23, 2013 at 7:23
  • 6
    @BoltClock'saUnicorn: Actually, life does not have a concept for fair or unfair, neither does the universe. It is only us who interpret random events against us (or our understanding of the universe or our moral) as unfair, and random events for us as fair. So, actually, life does not give single fu... Commented Oct 23, 2013 at 7:58
  • 5
    Well you could always post about it on meta and get some upvotes. The bikeshed problem has already been much discussed. Commented Oct 23, 2013 at 9:34
  • 5
    Simple solution: Make your answers dummier! Commented Oct 23, 2013 at 9:35
  • 4
    Life isn't fair. I think we've all been in the same situation. So, at least the unfairness is evenly distributed... which is fair. Commented Oct 23, 2013 at 13:58
  • 2
    @StevenPenny this question is closer in nature to meta.stackexchange.com/questions/31253/… than to your proposed duplicate.
    – Shai
    Commented Oct 23, 2013 at 20:17
  • 2
    Ironically enough, at the time of writing this comment, the former of your answers scores at +8 and the latter +16. Does this mean this question is invalid now? Commented Oct 24, 2013 at 8:32
  • 14
    [feature-request]: Make life fair.
    – apsillers
    Commented Oct 24, 2013 at 13:27
  • 3
    @probablyPekka: [status-bydesign]
    – Shai
    Commented Oct 26, 2013 at 21:44

8 Answers 8

60

This is not only about life being not fair to specialized knowledge questions. There is another difference between the low-hanging-fruit answer and the expert-knowledge post: age of the question.

Your first answer was posted within an hour of the question being asked. The question was still fresh, new, and new questions always draw attention. Your second answer was posted a month after the question was asked.

While answering older questions is really great (there are even badges to be earned), such answers are not seen by as many people in the first few hours of posting as an answer on a new question does.

There are a lot of factors that influence how many upvotes your answers get, including how easy it is to digest for anyone reading your question. How many people see the post is another.

Here comes the clincher though: specialized knowledge answers are rare, compared to the common knowledge posts. Now that your first answer has had its brief time in the sun, it'll be lost in the sea of other such answers. But your second answer will not face such competition. Over time, your specialized answer has a higher chance of garnering upvotes than your low-calorie sugary post. And that is a great thing.

5
  • 2
    So basically what you're saying is: voting/reputation system is working fine - don't mess with it. Message understood.
    – Shai
    Commented Oct 23, 2013 at 9:00
  • 2
    @Shai votes are directly tied to views. Posts that get extra attention through Meta posts, social media, and most commonly time tend to large number of views and consequently also get more votes. My best answers (Which is no where near the quality of your 2nd answer) are also in low traffic questions but they get sustained growth over time. But my easiest answers all got quick upvotes and little activity since. Commented Oct 23, 2013 at 9:06
  • @Shai: Indeed, the voting system works as intended. And don't compare votes between vastly different questions. Compare votes between answers to one question only. Commented Oct 23, 2013 at 9:10
  • @Shai Do you have a proposed change? Nobody's saying it's perfect, but it does seem to mostly work. That's not to say nobody would ever possibly consider making changes of some sort, although the arguments for it would need to be very compelling as voting is something that shouldn't be messed with lightly. Many of us simply don't know of anything that would be better, which is not to say that there is nothing better out there somewhere.
    – Servy
    Commented Oct 23, 2013 at 14:42
  • 1
    @Servy I was kind of hoping to get some new suggestions, but it seems like there are very strong and compelling arguments for leaving things the way they are. So, I'm not the only one who is aware of the flaws in the reputation system, but there is a general agreement that these flaws are "within tolerable boundaries" and there are no good alternatives.
    – Shai
    Commented Oct 23, 2013 at 15:05
49

Rare and specialized answers can only be understood by rare and specialized users. There are obviously fewer of these as compared to trivial answers which are easy and more common.

So we cannot force, ask, or request normal users to vote for a complex answer.

Often these rare and specialized posts have bounties applied which is a good and official way to reward the author very quickly.

Stackoverflow is about contribution, I personally prefer helping and solving someone's issue than getting more reputation. Few upvotes on these specialized answers are more worthy than lots of votes over trivial ones as to make someone understand complex logic is not an easier task. It gives more satisfaction and long term benefits too.

1
  • 2
    It's important to note, though, as Martijn Pieters did that it also takes somewhat rare and specialized users to write these kinds of questions, which means that there are fewer of them and they're easier to find while doing research. This means that, over time, the upvotes should be consolidated a bit better than for common and non-specialized questions and answers which are subject to lots of little variations and duplications. Commented Oct 23, 2013 at 16:04
33

To post, or not to post, that is the question:
Whether 'tis Nobler for the user to endure
The Views and Votes of outrageous reputation,
Or to take Arms against a Backlog of bugs,
And by answering end them: to log out, to AFK
No more; and by being AFK, to say we end
The Heart-ache, and the regexed HTML
That Tony the Pony inherited? 'Tis a badge
of gold to be desired. To log out, to AFK;
To AFK: perhaps to eat Waffles: Aye, there's the rub,
For in that AFK of activity what Unicorns may come,
When we have logged out of this stack overflow,
Must give us pause. There's the reputation
That makes strife of SO answering:
For who would bear the low votes and low views over time,
The random downvote, the pedant's comment,
The disapproval of hard work, the OP's delay,
The senselessness of the Fastest Gun, and the doubt
That knowledge receives in the face of learning,
When he himself might his answers make
With a fastest gun? Who would provide research,
To mock and test under a jittery compiler,
But that the fear of throwing an exception after publishing,
The uncaught exception, from whose cause
No control returns, complicates the answer
And makes us rather be the fastest gun,
Than post answers to questions we may not know.
Thus the dilemma does make Unicorns of us all,
And thus the Zalgo nature of answering
Is pushed out 6 to 8 weeks with the Arial font of consideration,
And experts of high reputation and knowledge,
With this consideration become lost in the why,
And are murdered by meta.

MSO/SO Adaptation of Shakespeare

0
24

Your short answer may get more upvotes, but probably just from the passing crowd thinking "True, that!" upvoting and moving on, doing nothing of much note with what they learned by your post.

Your more detailed answer will actually help someone. If it's a correct post, actually meets the requirement outlined in the question then the poster (and future visitors who have that problem too and find that question/answer) will use that information to achieve something. You're actually helping these people, and that makes a difference.

Sure, some internet points here and there from people who agree with what you say are nice and everything, but actually helping someone and therefore making a difference to their lives (albeit however briefly) is a much better feeling.

12

On Drupal Answers, my 2 of 3 most upvoted answers are things that took significant amount of research and explaining, and one is something that wasn't quite documented and I was able to respond just thanks to the fact someone pointed me to it in the past. So, not really trivial ones. They wasn't upvoted instantly, true, but over the time they gave me fair share of my reputation.

If you want to impress, answer complicated questions and wait. If you want to grind rep, answer questions usable for larger groups. That's all to it ;)

2
  • 14
    So instead of "life isn't fair" - "live long and prosper"?
    – Shai
    Commented Oct 23, 2013 at 7:14
  • @Shai Something like that :D
    – Mołot
    Commented Oct 23, 2013 at 7:14
9

Personally, I find it extremely rewarding if I am able to answer a difficult question, even if it does not get many upvotes, rather than a question which gets the same answer 10+ times.

It shows me that I have very specialized skills (which often requre a deep understanding of the topic) in a certain area, and nobody else was able to answer.

What is better than that?

7

Specialized answers are specialized. The upvote button tooltip says

This question shows research effort; it is useful and clear.

Most people won't be able to accurately judge whether a specialized answer shows research effort. (It might look like it does, but how can they tell it's not just a clever ruse?). Specialized information is, almost by definition, useful to fewer people because it's specialized. Clarity is similar; it's clear if you have some understanding in the field, but impenetrable if you don't.

To some extent this is by design. However, after those initial observations, Martijn Pieters's answer gets more and more weight. Over time, more and more people will see the question, and there won't be as much competition, and there won't be as many trivial duplicates. Non-specialized information is easy to generate and easy to find; so much so that it gets duplicated and diluted left and right. Specialized information, on the other hand, gets trimmed down, consolidated, and condensed.

You'll find lots of answers about silly little boolean operator mistakes, and they'll always be popping up, but for more specialized topics, time eventually starts to produce canonical answers. They're so good that no one needs to ask similar questions anymore, and you'll get all the sweet, sweet, rewards! (Well, if you wrote the canonical answer. But that's what I'm aiming to do too.)

7

When asking highly specialized question I always try to offset that by giving it a few hundred reputation points of bounty. This serves for both potential persons who'd answer the question, getting decent reputation points even without many upvotes, and for me as the question gets a lot more visibility (being specialized it doesn't get lot of upvotes nor views).

The only annoying thing is that I cannot set up a bounty right away after posting the question, even though I'm quite sure that a particular question will not be very popular.

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