33
\$\begingroup\$

It was recently brought up in chat that people are rewarded for trivial answers sometimes, which makes sense because this site is all about short code, but in other aspects, is bad because some answers that took a lot more work get very few upvotes in comparison.

This might discourage people from taking the time to answer longer and harder questions because although reputation is just a number, it's nice to get some reward, tangible or not, for your hard work.

Therefore, to counteract this, what can we do to help promote people answering harder and longer questions?

This is similar to this meta post about answering old challenges that get pushed to the bottom and become inactive, but is slightly different in that this post is about long and non-trivial answers that don't get noticed as much.

\$\endgroup\$
  • \$\begingroup\$ Related. \$\endgroup\$ – Shaggy Jun 28 '17 at 16:50
  • 10
    \$\begingroup\$ To me, this is simply the PP&CG version of the network-wide "early bird gets the votes" problem. Trivial answers are faster to develop and post, and they sometimes get an "early answer" boost. I occasionally post trivial answers, but only when I can't find a shorter solution in my language. What other people vote up is up to them. \$\endgroup\$ – BradC Jun 28 '17 at 16:55
  • \$\begingroup\$ xnor's proposal might help with the trivial answers. (Encouraging answers to hard problems is still a good thing to discuss though.) \$\endgroup\$ – Martin Ender Jun 28 '17 at 17:38
  • 1
    \$\begingroup\$ Honestly, bounties is the real answer. \$\endgroup\$ – Stephen Jun 28 '17 at 17:56
  • \$\begingroup\$ @BradC I agree; that's true, so this is to try to incentivise harder and interesting answers. \$\endgroup\$ – HyperNeutrino Jun 28 '17 at 18:17
  • 18
    \$\begingroup\$ Trivial answers are a symptom: the problem is trivial questions, which were discouraged in the early days but now constitute 90%+ of those posted. \$\endgroup\$ – Peter Taylor Jun 28 '17 at 18:24
  • \$\begingroup\$ @PeterTaylor That raises a fair point too; unfortunately, I don't think we can banish trivial questions or even reduce them significantly, but I think this could be a step forward, if only a small one. \$\endgroup\$ – HyperNeutrino Jun 28 '17 at 18:25
  • 7
    \$\begingroup\$ I wouldn't say that trivial answers get too many upvotes, I would rather say that non-trivial answers get to few upvotes. \$\endgroup\$ – Fatalize Jun 29 '17 at 6:23
  • 2
    \$\begingroup\$ @PeterTaylor And recently there has been an influx of "we don't have a very simple <whatever> challenge yet so here's a trivial one" because they were discouraged inititially \$\endgroup\$ – cat Jun 30 '17 at 16:32
  • \$\begingroup\$ @Fatalize Better. Thanks. \$\endgroup\$ – HyperNeutrino Jun 30 '17 at 16:37
  • \$\begingroup\$ Does anyone know why the display algorithm (which does use randomness, presumably to increase fairness) can't be adjusted even further to counteract the early bird problem? On SO it makes more sense not to do this, since people arrive to find answers and you want the likely sought one to rise quickly to the top (though even here, the first 24 hours could implement a fairer algorithm). But on CG there's no reason why a 3 byte built-in that's garnered 8 votes can't be displayed lower more often, giving newer answers a chance to catch up. \$\endgroup\$ – Jonah Jul 9 '17 at 4:42
  • \$\begingroup\$ @Jonah The display algorithm is fully deterministic. There are three ways to sort answers. You can sort by votes, date of creation, or date of last modification (edits, not comments). There is no randomness. Sort by votes make sense for SO but not PPCG; that's why some answers have huge score gaps. \$\endgroup\$ – HyperNeutrino Jul 11 '17 at 1:41
  • \$\begingroup\$ @HyperNeutrino Weird. My own answer sometimes jumps around (without any score change) so I just assumed. In that case, I amend my comment to say that it should use randomness for the reason I suggested. It doesn't seem like a hard problem to fix... \$\endgroup\$ – Jonah Jul 11 '17 at 1:44
  • \$\begingroup\$ @Jonah Sorting by newest or activity is a good option (unfortunately, newest sorting doesn't exist), but sort by votes promotes drive by votes making a huge score gap. \$\endgroup\$ – HyperNeutrino Jul 11 '17 at 1:46
  • \$\begingroup\$ @HyperNeutrino, I'm saying it should be enforced (or at least, default and opt-out) that here sorting happens by an algorithm which uses randomness to show newer answers always at the top, with a probability high enough that a better answer would in fact be expected to catch up eventually. Relying on educated users and opt-in behavior is effectively deciding not to fix the problem. \$\endgroup\$ – Jonah Jul 11 '17 at 1:50
14
\$\begingroup\$

Run a contest

The chat room is now open for nominations

https://chat.stackexchange.com/rooms/61232/biweekly-contest-for-undervoted-answers

This idea is inspired/copied from Peter Taylor's answer to the linked meta post.

I propose running a contest to reward undervoted answers, as I call them. I envision to run it in a chatroom to avoid cluttering meta.

My current idea is to follow these rules:

  • The contest runs biweekly.
  • Nominate non-trivial answers that don't have a lot of upvotes, and especially those that show that a lot of work was put into them (a short answer isn't necessarily trivial)
  • The winning answer will be decided by the linked answer with the most stars.
  • Attempting to corrupt the process via bribery and/or blackmail will not be tolerated, and may result in moderator action.
  • Prizes will be awarded as bounties; the bounty-giver will award a bounty to the winning answer at the end of the contest. The bounty-giver will rotate between those who sign up to be bounty-givers.
  • The bounty is not necessarily a certain amount and will change based on how much effort was put into it, and scales down inversely to its score.
  • Other users are welcome to make their own selections and award bounties as well.

I have created the chat room here.

\$\endgroup\$
  • 1
    \$\begingroup\$ I'd reword the second point slightly to simply ask people to "nominate non-trivial answers that don't have a lot of upvotes." Maybe also include answers that show that a lot of work went into them. A short, seemingly trivial answer, might not have started out like that, it might have taken a lot of golfing to get to that point. \$\endgroup\$ – Shaggy Jun 28 '17 at 16:53
  • \$\begingroup\$ @Shaggy Good idea; thanks. \$\endgroup\$ – HyperNeutrino Jun 28 '17 at 18:17
  • 1
    \$\begingroup\$ I dislike the idea of you and you only being the final judge. Since we place a lot of emphasis on community, maybe the winner should be based on number of stars \$\endgroup\$ – Beta Decay Jun 28 '17 at 23:27
  • 1
    \$\begingroup\$ Similarly, it's not fair on you to be the main bounty giver. Maybe have a rota so each week, a different person awards the bounty \$\endgroup\$ – Beta Decay Jun 28 '17 at 23:28
  • \$\begingroup\$ @BetaDecay Maybe that would be a good idea. I like your previous suggestion especially. As for the second one, that would work too; we could have a signup of people who want to be bounty-givers. \$\endgroup\$ – HyperNeutrino Jun 28 '17 at 23:32
  • \$\begingroup\$ If we get a moderator on board, these can be marked as events and appear on the yellow side-box. (ROs might also be able to do this - I'm not sure.) \$\endgroup\$ – wizzwizz4 Jul 9 '17 at 18:03
  • \$\begingroup\$ Did this project die? \$\endgroup\$ – Leo Sep 29 '17 at 11:38
  • \$\begingroup\$ @Leo Yes; there wasn't enough support for it. \$\endgroup\$ – HyperNeutrino Sep 29 '17 at 11:59
  • \$\begingroup\$ Oh, sad to hear, I didn't know of it :( \$\endgroup\$ – Leo Sep 29 '17 at 12:01
7
\$\begingroup\$

Upvote them

Literally just upvote answers that take work. Be the solution.

\$\endgroup\$
  • \$\begingroup\$ And comment! Comments help garnerattention too! \$\endgroup\$ – tuskiomi 12 hours ago

You must log in to answer this question.

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