DPChallenge: A Digital Photography Contest You are not logged in. (log in or register
 

DPChallenge Forums >> Web Site Suggestions >> idea for fixing self-serving voting
Pages:  
Showing posts 1 - 25 of 55, (reverse)
AuthorThread
08/04/2003 10:00:47 PM · #1
here's an idea i've been kicking around for awhile. i'd be interested to hear what others think of it, or ways to refine it.

currently, if you're entered in a challenge, and you're doing well, there's nothing to stop you from going around and giving all the pictures that you think are 'threats' to your ranking lower scores.

potentially, everyone could do that, driving all the scores and averages down.

so here's the potential solution:

1) divide the pictures into 2 pools.

2) Here's the kicker: make it so that you can only vote on the pool that you're NOT in. This makes it so that you will vote ONLY on the merits of the pic and not how your vote might affect YOUR OWN ranking.

3) give separate ribbons for each pool. stagger the pools to alternate weeks or parts of weeks so we don't have to have double the current the winners up on the front page at the same time.

thoughts?
08/04/2003 10:07:35 PM · #2
obviously you have given this some good thought. i'm not convinced this is a problem that's worth creating new policies for. regardless of how rampant this voting trend may be, the best photos always rise to the top.
08/04/2003 10:09:46 PM · #3
i agree with achiral.
08/04/2003 10:21:42 PM · #4
Yes, I think that the most people here do not give everyone else a 1 because they think they can win that way.
08/04/2003 10:32:33 PM · #5
With the number of entries in each challenge someone voting a 1 out of spite will have little effect.

e.g. lets start with 200 votes and an average of 6

if the next person votes a 1 then the average will drop to 5.975, not much of a difference

Even the top rated images get a few 1's

Let everyone vote their own way. With the number of entries in each challenge it all works out in the end.

I await the next thread on this very same subject come the next challenge.


08/04/2003 10:32:50 PM · #6
Not to jump onto the pile here, but that would be a very extravagant change to the site, and would have a lot of downsides to go along with it. If this type of voting seemed to be a severe threat to the function or enjoyment of this site, your suggestion would work perfectly it seems. However, at present, it's definately not worth all the drawbacks.
08/04/2003 10:45:30 PM · #7
In spirit, I guess I kinda agree with Mag. I'm just not sure how common it (self-serving) voting is. I don't have a better solution, though, but I don't think splitting it into two different pools would help. If someone is desperate enough to low ball someone, I think they will do it whether or not their picture is "competing" with them. I also think that if someone does that enough, they will eventually get tired of it and move on. Maybe I'm just a naive optimist (though I'm a happy, naive optimist).
08/04/2003 11:00:40 PM · #8
Interesting idea...

Pro: It would cut down on the number of images we have to vote on each week, thus possibly increasing the overall quality of scores. There's no way you can objectively view, ponder, and score 350 photos in one week. I have to put SOMETHING on my timesheet at work.

Con: Statistically, I don't think that vengeful voting is rampant (I, for one, know that the pictures I think are better than mine and could win always end up like 78th). It would create a bunch of hullabaloo that we probably don't need.

Not to usurp the topic, but I think if we are going to change the voting system I think we need a way to vote on the technical merits vs. the subject and its adherence to the topic. Just MHO.

Rob
08/04/2003 11:00:45 PM · #9
I think the "self serving voting" thing could easily be determined by somebody with access to the MySQL score database.

Take a bunch of users who have both entered and not entered photos in recent challenges, and voted on challenges they have entered and not entered. Determine the average vote given when they had an entry in the challenge (and compare it to the score they received for their entry) vs. the average vote given when they did not have an entry in a challenge.

My guess would be that the average would be lower (at least in some cases) for those challenges where the voter had a "competing" picture, validating the "self serving voting" issue. (That is, the thought process of "I'm only getting a 5.2 right now, so I better not vote anything higher than a 5" or the "I'm doing OK but this picture is really good so I better vote it low".)

Personally I'd like to see the scores for participants hidden until the results were finalized. I don't see any tangible benefit to being able to watch the score hour by hour. (Especially since there isn't anything you can do to your submission as the week goes on to try and improve your score.)
08/04/2003 11:03:24 PM · #10
Originally posted by wingy:

If this type of voting seemed to be a severe threat to the function or enjoyment of this site


I think it is considering that this is one of the most frequently posted topics on this site....at least one post per day.
08/04/2003 11:13:14 PM · #11
I believe this is where my idea of the different levels of ablilities would work. Where you couldn't vote in your own level, thus stopping the low voting on images that are a threat to yours.
08/05/2003 12:31:52 AM · #12
ALL THAT I CAN SEE HAPPENING IS PEOPLE PSYCHING THEMSELVES AND EACH OTHER OUT....HUMAN NATURE. IF YOU CHANGE THE SYSTEM THERE WILL BE PEOPLE GIVING LOWER VOTES ANTICIPATING GETTING A LOWER VOTE THEMSELVES...JUST IN CASE. NO SYSTEM WILL BE PERFECT WHEN THE PEOPLE WHO SUBMIT ARE THE ONES WHO VOTE. A DIFFERENT PANEL OF JUDGES MAY BE THE ANSWER BUT THATS NOT THE POINT OF THIS SITE AS I SEE IT. ITS PEERS JUDGING PEERS AND GIVING COMMENTS CONSTRUCTIVE OR OTHERWISE, NOT THE WINNING THATS IMPORTANT.

I ALSO AGREE THAT THEY BETTER PHOTOS TEND TO RISE UP AND DO WELL.....MAYBE NOT WIN BUT THE RATE BETTER THAN CRAP PHOTOS.

Message edited by author 2003-08-05 00:32:47.
08/05/2003 07:54:22 AM · #13
Originally posted by OneSweetSin:

I believe this is where my idea of the different levels of ablilities would work. Where you couldn't vote in your own level, thus stopping the low voting on images that are a threat to yours.


Interesting concept...actually competing with others of your skill level. I equate that to having professional athletes competing with high school kids. Does seem like a positive way to approach things. As our experience grows, we can move up into the "professional" level!
08/05/2003 11:25:49 AM · #14
I don't see it as that extravagant of a change. It could be done totally transparent to the user.

Picture this:

You submit your picture (as usual). It gets randomly assigned to one of the pools (requiring no user intervention).

The challenge starts -- you go to vote. You've been assigned to the opposite pool from where your picture went - so you see a group of pics to vote on, AS USUAL.

You vote on them, but you're able to vote, freely knowing that you won't 'help someone beat you', so you can give each pic the score it deserves.

At the end of the time period, you get to see the results for the pool you voted on, as normal. And you get to see the results for the pool your pic was in, as normal.

Everything "different" happens behind the scene. The only difference is internal logic that prevents people from voting on the same pool that contains their picture.


08/05/2003 11:42:41 AM · #15
If self serving votes is a problem (I'm hoping we are a tad more mature than that) then I have an alternative solution. Make everyones votes a matter of public record. Then if a 1st place person is found to have voted a 1 on the 2nd, 3rd and 4th place shot, peer pressure and public opinion will swat that offender like the gnat they are and the behavior will likely change.

I'd like to believe this problem doesn't exist and that no solution is needed... But if others suspect it happens, it would be easy enough to spot.
08/05/2003 11:46:01 AM · #16
Originally posted by magnetic9999:

The only difference is internal logic that prevents people from voting on the same pool that contains their picture.


There are other differences: You cut the number of votes by a third (or half or whatever). Consequently we get half the comments, or maybe slightly higher since people may be more inclined to comment if they only have half the voting to do.
In the movies challenge, that means we'd have only 30 people maximum in the other pool voting on our images. With low participation numbers like that, one or two grumpy voters make a big difference.

Plus then we'd get the inevitable "I'd have placed higher if I was in the other pool" threads. Not to mention the "Pool A winner was better than the Pool B winner" threads.

Seemingly human nature requires us to bitch at any and all opportunities. I don't think the odd Troll vote by someone who derives his/her self worth from an internet photo contest will statistically affect anything. Especially when there are several hundred entries.

The cream usually does rise to the top, despite the fact that each one of us were cheated on every contest we've ever entered.

:)

I think the reason we should consider making the changes you guys are talking about is when the site is so big, the voting is unmanageable (and yes, I think 350 is starting to push the envelope a bit).

and THAT, is all I have to say about that.

Pedro
08/05/2003 11:52:00 AM · #17
Has anyone (with access to voting data) done a study to see how prevalent this problem actually is?

I think it gets mentioned a lot in forums but more in the sense of people looking desperately for reasons why the photo they were convinced would win only got a 5.5 than in the sense of people saying they have reason to believe much of this is actually happening.

Unless there is evidence in the voting logs to show that the problem is prevalent then I'm not sure I'd feel this would be worthwhile.

I would like to believe that most people here are essentially mature individuals who would not stoop to such behaviour.

When I vote I vote according to my own scale and seldom think about my own entry, if there is one, since I know very well that I'm not able to judge it objectively. So I trust that others are voting as I am, truthfully, and that, if I'm scoring lower than I would have liked, it's because that's the score, on average, that I deserve.


And if you made sense of that, my brain isn't suffering from meltdown as much as I feared... :o)
08/05/2003 11:52:22 AM · #18
Skill level voting won't sort out the self-serving voting "problem" if it exists, but it's the best change this site could make in the near future.
08/05/2003 12:06:54 PM · #19
Just thought i'd have a quick check in the archive ...

The winners of the last ten challenges have recieved, between them, 7 ones and 9 twos. Five of those seven 1's were for Willem's round entry, which I will happily put down to people not getting it's abstract quality - or not liking it, which is fine.

I would suggest the proposal ain't worth the effort, repectfully.

Ed
08/05/2003 12:07:48 PM · #20
2 things:

1) if it's not a problem, why are the average scores on the site so low? Why is it so hard to get a 7, and nearly impossible to get an 8?

it doesn't have to be as blatant as giving others 1's .. it can be as simple as giving everyone else 4's 5's and 6's just to make sure you dont 'give them an edge'. this results in an overall depressed score for the pics on this site.

2) affecting number of comments - personally i think it would result in MORE comments, simply because when you don't have to vote on 75 million entries, you're freed up to give more and better comments to the pics that you are voting on ..

:)

Message edited by author 2003-08-05 12:08:29.
08/05/2003 12:10:53 PM · #21
Originally posted by magnetic9999:

2 things:

1) if it's not a problem, why are the average scores on the site so low? Why is it so hard to get a 7, and nearly impossible to get an 8?

it doesn't have to be as blatant as giving others 1's .. it can be as simple as giving everyone else 4's 5's and 6's just to make sure you dont 'give them an edge'. this results in an overall depressed score for the pics on this site.

2) affecting number of comments - personally i think it would result in MORE comments, simply because when you don't have to vote on 75 million entries, you're freed up to give more and better comments to the pics that you are voting on ..

:)


Oops, forgot to type my reply!

i think you're right mag as afar as that goes: but out off the number of people voting on some challenges, averaging an 8 is really massively high. Even around challenge 1 the winner got onyl 8.5, from twenty votes. The problem is there is no holy grail perfect picture, and people take subject matter into account when voting. Some people dislike studio shots - some people think the same about landscapes - not everyone is going to vote fabulously for any one shot, however good.

That said, I do have an intense dislike for those here whose photography is god-awful and whose voting average is around 3.5-4. Smells very strongly of sour grapes to me.

But it wouldn't change anything - they'd still vote things down.

Ed

Message edited by author 2003-08-05 12:14:58.
08/05/2003 12:11:16 PM · #22
considering the response this suggestion is getting, i don't *ever* and I mean EVER want to hear anyone complaining again about how others are giving them low scores to boost their own pic's standings.

:P

Message edited by author 2003-08-05 12:17:40.
08/05/2003 12:18:09 PM · #23
Originally posted by e301:


That said, I do have an intense dislike for those here whose photography is god-awful and whose voting average is around 3.5-4. Smells very strongly of sour grapes to me.
Ed


Just thought - perhaps they just can't see very well?
08/05/2003 12:18:19 PM · #24
I think that if this were a prevelant problem it would be reflected in the voter profiles. Just compare average votes cast to average votes received. If it shows a large negative difference, one could assume that person was abusing the present voting system.
Whoa! I better go check mine. :<)
08/05/2003 12:54:25 PM · #25
I think there are two things that can be done to get a better result for good pictures that mysteriously get a low result.

1. Do an analysis for incompetant voters. Not sure of the exact method to do this, but hers a try: Each vote is effectivly a prediction on where in the rankings the photo will appear (or possibly a prediction of its final score). Bad predictors will show a large deviation from the final rank/score. I think bad predictors are either going to be attempting to sway the vote, or just be incompetant and need helping. Perhaps a tutorial on art appreciation is needed. This method will punish exessivly high/low voters too.

2. Take the average of a users 10 best scores (or at least disregard their lowest, else experimentation will be punished), and use that number to weight their vote, so that sucessful photographers get a larger vote. Im not sure this idea is that great, but it might be nice to see the voting/ranking opinion of the m99/bod/setz/shiz/jak/drj/etc group seperatley from the great unwashed.
Pages:  
Current Server Time: 02/20/2020 08:38:23 PM

Please log in or register to post to the forums.


Home - Challenges - Community - League - Photos - Cameras - Lenses - Learn - Prints! - Help - Terms of Use - Privacy - Top ^
DPChallenge, and website content and design, Copyright © 2001-2020 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Proudly hosted by Sargasso Networks. Current Server Time: 02/20/2020 08:38:23 PM EST.