Proposal for an Answerathon

Question

Please click here for discussion of the details of the Answerathon!

I recently came across an intriguing series of competitions on Travel Meta that I thought might make for an interesting proposal for us to do. The idea is that there's a lot of unanswered questions, so they have (had) a monthly competition in the interests of drawing attention to those questions.

The way it worked over there is that in order to remain in the game, users had to answer an unanswered question asked by someone else AND receive an upvote on it. Once the upvote's been made, the user's added to the scoreboard. You're out if a day passes with no upvoted answer, but of course, don't just post an answer for the sake of posting an answer. Last user standing wins.

---

We currently have 3,561 unanswered open questions with a nonnegative score. That's a lot.

Now, if we cross-reference these 3,561 questions with the top tags, we get:

• 918 halacha
• 194 minhag
• 229 sources-mekorot
• 184 shabbat
• 232 tefilla
• 151 history
• 124 parshanut-torah-comment
• 129 blessing

My suggestion is to divide the competition into classes: one for each of the top eight tags (after which we drop into ninety and seventy eligible questions each). Over the competition, users will still be striving to post answers to specific categories (rather than unanswered questions in general). So, instead of (in addition to?) a general leaderboard, we would have a halacha leaderboard, a shabbat leaderboard, etc.

The benefit comes when you consider the amount of research involved in a good answer. By focusing in on one subject, a user would be able to be more effective at answering the questions - and answering them well.

The other change I propose to Travel's rules is that, instead of it being last user standing, it should be that the contest is over when the last user is out, but the winner is the one who answered the most questions. This encourages users not to just post one answer a day (kinda defeats the purpose), but rather to post high-quality answers throughout the competition.

Oh, and for bonus points, it goes without saying that all submissions are eligible for Best Answer Contest and DoubleAA's bounty.

---

There's one obvious problem with this. What's to stop users from spamming answers?

One possible solution is to up the threshold from 1 upvote to 2 in order to be eligible. Another is that if someone abuses the system, they're disqualified from that and all future competitions (assuming that they don't get banned in the process). I don't know of a particularly good answer, or combination of answers, but hey, that's why this is on Meta, isn't it?

---

The purpose of this thread is for two reasons:

1. Is this a good idea? Why or why not?
2. Suggestions for improvements.

Once we have a consensus on going through with this and flesh out the general details, then we can move on to the more specifics.

Please post answers with your ideas, so that we can get community consensus on those as well!

Answer 1

I like this idea.

I suggest first trying this once with one of the tags, seeing how it goes, then iterating with another, etc.

I wouldn't worry too much about people gaming the system or spamming. You haven't proposed a tangible prize, so there's little additional (over and above the imaginary internet points we already offer) incentive to cheat. If people post answers to unanswered questions that at least one other person (net) finds valuable, that's a win, even if they're not all-star answers.

The shorter and simpler the list of rules, the more people may participate, and the less time we'll spend discussing the rules. So save on rulemaking by just assuming everyone's in it in good faith.

There's value in rewarding consistent contributions over time, since that gets people into habits that could possibly be sustained. Also, tracking just one entry per contestant per day makes the recordkeeping simpler and shorter than tracking and counting up lists of entries. I would recommend trying the Travel's model first and seeing how many answers it nets, then considering changes.

It would be fun to track the total yield per contest of the whole community [of contestants]. Post the number of unanswered questions in the target tag at the start, then see how many it's down by at the end. That becomes a score the communtiy can try to beat in the next iteration.