The polynomial averages between entries and votes have a large gap with entries going higher and votes lower. Have we got to the point that the site needs to do something to get the averages closer to each other? If so, what?
Should the site limit the amount of entries?
Should the site require voting to submit?
Should the site reduce the amount of challenges?
Should the site do nothing and see if the trend reverses?
I’m interested in hearing some of your thoughts.
this is not a rant thread. It is a thread to debate what a growing site should or should not do when it sees trends change dramatically.
That is an interesting point, i wonder how not seeing your vote until you vote would help because it would force you to vote if you want to keep track of your score, and i know of several people who don't like voting in challenges they have participated in
It's no real surprise... entries have spiked significantly, with the DPL probably being the biggest driver. Meanwhile, the voting pool, those users that regularly vote on challenges, has not grown substantially. Therefore, unless those users increase the number of images they vote on, the number of votes per image will drop.
ETA: During DPL, total votes per challenge actually increased, about 50% in Member Challenges, and somewhat less in Open Challenges. The number of entries increased by a larger margin, however, and thus it looks like "entries are increasing and voting is decreasing." Based on the observed data, I'd conclude that "entries dramatically increased suddenly, and voting could not keep up."
I'm too tired from writing grad papers to sound technical, even if I could...and hope this doesn't qualify as rant...because it isn't..
I don't vote in many challenges now..because it's very hard to vote on all submissions (which is what I try to do WHEN I vote).
obviously...voting 100% on challenges with 200 entries works better for me than even attempting to vote 100% on challenges with 350+ entries. not limiting entries does give pokey butts like me time to get entry in...but if # of entries is limited folks, such as myself, would either have to get off duff and entry or be left out...goals and deadlines really do help procrastinators, again...like me, correct bad habits :-)
summation? limit entries...so that it is possible to vote the full challenge.
I don't think you have a statisticly significant trend there, frankly. Polynomial fits are famous for allowing the illusion of trends where there are none; this is particularly true at the ends of the data (notice the same problem exists at the beginning of the graph, where it is ignored becuase it makes no sense there.
I think you could argue a slight decrease in votes since the 300 mark; I would be loath to claim any trend in entries without looking more closely at the data, but I certainly don't see a dramatic upswing.
I like this...we once had a challenge that you couldn't enter unless you had given more comments than recived. I don't know how that would work for votes...maybe the same way.
One option is to not make your score visible until you have voted 20% of the challenge.
What about the trolls? ;)
Trolls are people too! :P
And suffer from update addiction like the rest of us.
So they'll just speed through giving everyone the same vote (1-1-1-1-1-1-1-1-1) until they reach the magic number to see their scores. They'll be tossed out at the end, but in the meantime render the score you see utterly useless. Not to mention that this approach would unfairly lock out those who have an ethical objection to voting in challenges they've entered.
A better tactic might be to hide scores until 75% of the entries have received 100 votes. That way, everyone on the site is accountable for the lack of votes, and the voting is more likely to be honest.
Imagine how wild the conspiracy would get if people were forced to vote.
"Why did we make people vote when they didn't want to? My score is 5.1 and I know it would've ribboned before we made the new rule! Of course, my past scores consist of plenty of low numbers but I know very well that my scores would be much better because these new voters don't see my awesomeness. Of course, the old voters didn't see my awesomeness either but we won't talk about that".
Ok, what if challenges were doubled, 2 member, 4 open but still exclusive, and you weren't allowed to vote in the challenge you entered. Entries would still grow over time, but the amount of voting you could do would be cut in half. Now i know it would seem like its less voting, but could the reason less votes are happening because people don't want to vote, or because there's too many entries to simply vote on them all?
And assigning DPL teams to specific challenges each week would still keep the # of submissions per entry about avg, even though they even out each week anyway. And it could do away with the perception of people voting harsher during the DPL season since your not voting on entries you've entered, it has no bearing on your score.
I don't think you have a statisticly significant trend there, frankly. Polynomial fits are famous for allowing the illusion of trends where there are none; this is particularly true at the ends of the data (notice the same problem exists at the beginning of the graph, where it is ignored becuase it makes no sense there.
I think you could argue a slight decrease in votes since the 300 mark; I would be loath to claim any trend in entries without looking more closely at the data, but I certainly don't see a dramatic upswing.
The problem does not exist at the beginning. At the beginning of the graph there are more votes than entries, directly opposite, which would not pose a problem.
