| Author | Thread |
|
|
05/05/2010 01:47:52 PM · #76 |
Originally posted by scalvert:
Originally posted by redjulep:
I also like the idea of scrubbing opposing DPL team votes per challenge. |
Be careful what you wish for. DPL participants tend to be among the more "experienced" voters, and discounting the opposing team's votes may have a negative impact as a result. For example, I gave out thirteen votes of 9 or 10 in the Balloon challenge, and three of those went to our opponents (even though the author of one was obvious). None of my competitors voted the Double Exposure entry less than a 6 either, so eliminating those ratings would actually increase the impact of trolls! |
You're missing the point. It's not the head-to-head actual scores which are the problem. The problem is the overall depression of scores by the inexperienced voters who are trying to elevate their own scores by voting down on others. They cut a wide swath, even though their target is very narrow. This phenomenon is palpable. |
|
|
|
05/05/2010 01:51:57 PM · #77 |
Let me stir the pot for a bit...
Do the trolls/hackers have any idea who they are trolling? I mean sure, there are some entries that give away or give many hints of its photographer, but I am talking about the vast majority of the entries.
Given the number of entries, and the random order of the entries in voting, can't we assume that for the most part, the effect of the trolls is pretty evenly distributed over all the entries in the challenge. I mean the absolute scores in a challenge will be lower but the relative postions of entries should remain unchanged. I am missing something here?
Since DPL started, I have noticed a drop in my scores, but I haven't noticed a drop in my percentile finishes. A 5.5 will get you a lot farther these days... |
|
|
|
05/05/2010 01:54:57 PM · #78 |
Originally posted by hahn23:
The problem is the overall depression of scores by the inexperienced voters who are trying to elevate their own scores by voting down on others. They cut a wide swath, even though their target is very narrow. This phenomenon is palpable. |
You're making a huge assumption. Can you prove it? |
|
|
|
05/05/2010 01:56:35 PM · #79 |
Originally posted by Five_Seat:
can't we assume that for the most part, the effect of the trolls is pretty evenly distributed over all the entries in the challenge. I mean the absolute scores in a challenge will be lower but the relative postions of entries should remain unchanged. I am missing something here? |
Right. DPL is ranked by percentile, so voting down a bunch of entries is meaningless. |
|
|
|
05/05/2010 01:56:47 PM · #80 |
So "inexperienced photographers (voters) who try to elevate their own scores by voting down on others" is a big problem here? I think not.
I can't even imagine that there are that many "inexperienced" voters out there who are PURPOSLY trying to bring a high score down....in order that their score will look better.
What is an EXPERIENCED voter? One who has appeared in the top 20 more than 10 times? If so, there are many, many more INEXPERIENCED voters than Experienced. |
|
|
|
05/05/2010 01:58:58 PM · #81 |
Originally posted by kenskid:
What is an EXPERIENCED voter? |
People who have been around this site for a while, and know the general rules and expectations. This would include the overwhelming majority of DPL participants. |
|
|
|
05/05/2010 02:01:34 PM · #82 |
Originally posted by scalvert:
Originally posted by hahn23:
The problem is the overall depression of scores by the inexperienced voters who are trying to elevate their own scores by voting down on others. They cut a wide swath, even though their target is very narrow. This phenomenon is palpable. |
You're making a huge assumption. Can you prove it? |
No, I don't have access to the data. But, there are a lot of early (and late) very low votes going to some pretty good images. I wouldn't be too concerned about this, if more than 20% voting was required and cherry picking from thumbs was prohibited. Look, I'm just suggesting ideas for improvement in a constructive criticism sense. Why not remove the incentive to cast targeted votes?
Message edited by author 2010-05-05 14:02:31. |
|
|
|
05/05/2010 02:04:08 PM · #83 |
Please remember that the original intent of my post wasn't to battle 'unfair' voting (which has already been discussed ad nauseum) but rather suggesting another method to catch people who are using ghost accounts.
Voting is subjective. I.e. what's a 10 to me might be a 3 to you. This must continue to be like it is. Langdon and SC did a very good attempt in catching friend voters etc earlier this year and several members threatened to cancel/not renew their memberships as a result.
But if we could stop or limit ghost accounts which is undermining the entire voting process AND the DPC concept, I think we would make it better for all of the serious users. I know this is being worked on continually. But again, I was just suggesting an alternative method based on consecutive 1's, 2's or 3's which based on statistics and odds would not likely happen unless there are ghost accounts. From there it is possible to find aliases which would then reveal creation dates of these and thereby give us new patterns that might give us a hint that for example quarantine/no voting for new users could be a good idea. |
|
|
|
05/05/2010 02:07:07 PM · #84 |
Originally posted by scalvert:
Right. DPL is ranked by percentile, so voting down a bunch of entries is meaningless. |
This is not nessicarily true either, the overall positioning of your photo is determined by your score. Therefore your percentile finish is directly correlated to your score. With no chance to "hurt" your team by voting lower, leaves the possibility of unfair voting to make your teams percentile finish higher. |
|
|
|
05/05/2010 02:25:40 PM · #85 |
Originally posted by Five_Seat:
Given the number of entries, and the random order of the entries in voting, can't we assume that for the most part, the effect of the trolls is pretty evenly distributed over all the entries in the challenge. I mean the absolute scores in a challenge will be lower but the relative postions of entries should remain unchanged. I am missing something here?
|
That only works if you assume that all the low voters vote on 100% of the entries. If they only vote on 50% and give them all 1s or 2s, the 50% of the entries they didn't vote on will have an advantage.
It should average out, but could also really affect the higher scoring images. I've seen many a challange where the removal of a 1 vote meant the difference between top 5 and top 10. |
|
|
|
05/05/2010 02:32:41 PM · #86 |
Originally posted by scarbrd:
Originally posted by Five_Seat:
Given the number of entries, and the random order of the entries in voting, can't we assume that for the most part, the effect of the trolls is pretty evenly distributed over all the entries in the challenge. I mean the absolute scores in a challenge will be lower but the relative postions of entries should remain unchanged. I am missing something here?
|
That only works if you assume that all the low voters vote on 100% of the entries. If they only vote on 50% and give them all 1s or 2s, the 50% of the entries they didn't vote on will have an advantage.
It should average out, but could also really affect the higher scoring images. I've seen many a challange where the removal of a 1 vote meant the difference between top 5 and top 10. |
Valid point. I guess you have to have a certain number of trolls active to "hit" all the entries.
I don't have the time to go back through past challenges and look at the distribution of 1 to 3 votes over all the entries in each challenge. But that may be somewhere to start... |
|
|
|
05/05/2010 02:33:15 PM · #87 |
I'll be honest - when I first joined the site, I entered a challenge right away and the thought seriously occurred to me to get some friends and family to register just so they could vote on my entry. I knew they would have to vote on 20%, so they probably would have just given lower scores to other entries. It would be the same thing many of us do on other sites that "don't really care" about where the votes come from. It wasn't clear to me initially that this site was any different. Probably the only reason I didn't do it was that my family is small, I have no friends, and I'm too lazy to rally votes. The point is that I am pretty sure that some newcomers think the same thing and don't read the rules or take them seriously and ultimately their friends troll votes can and do taint the overall results. And I think this issue can be alleviated by simply putting up a small obstacle (3 challenge entries before voting). Again, what is the downside of that requirement besides a few less voters?
On the issue of scrubbing DPL votes, Shannon makes a good point and I'm not hung up on that issue anyway. |
|
|
|
05/05/2010 02:37:29 PM · #88 |
Originally posted by Art Roflmao:
I'll be honest - when I first joined the site, I entered a challenge right away and the thought seriously occurred to me to get some friends and family to register just so they could vote on my entry. I knew they would have to vote on 20%, so they probably would have just given lower scores to other entries. It would be the same thing many of us do on other sites that "don't really care" about where the votes come from. It wasn't clear to me initially that this site was any different. Probably the only reason I didn't do it was that my family is small, I have no friends, and I'm too lazy to rally votes. The point is that I am pretty sure that some newcomers think the same thing and don't read the rules or take them seriously and ultimately their friends troll votes can and do taint the overall results. And I think this issue can be alleviated by simply putting up a small obstacle (3 challenge entries before voting). Again, what is the downside of that requirement besides a few less voters?
On the issue of scrubbing DPL votes, Shannon makes a good point and I'm not hung up on that issue anyway. |
Are you sure you didn't rally votes on your first entry there Art?  It seems your first few entries did pretty well. Now if you want to see how a newcomer is suppose to score just have a gander at my first entries... come to think of it a few of my recent ones (especially my current FS) is doing about as well as my first one. N/M I think I just suck lol |
|
|
|
05/05/2010 02:46:47 PM · #89 |
Originally posted by jminso:
It seems your first few entries did pretty well. |
I think Langdon has some code in there to boost newcomers scores for the first challenge to get them hooked. It's like any good drug, there's the initial rush, then the long, painful coming down part, but by then you're addicted. ;-) |
|
|
|
05/05/2010 02:47:18 PM · #90 |
Originally posted by Art Roflmao:
..., I have no friends, |
Not true. You have  and I can prove it :P |
|
|
|
05/05/2010 02:48:15 PM · #91 |
Originally posted by Art Roflmao:
Originally posted by jminso:
It seems your first few entries did pretty well. |
I think Langdon has some code in there to boost newcomers scores for the first challenge to get them hooked. It's like any good drug, there's the initial rush, then the long, painful coming down part, but by then you're addicted. ;-) |
Man I got screwed then. Oh well, I am hooked now regardless and even if I am completely aggravated at my scores right now but oh well. Onward I go. |
|
|
|
05/05/2010 03:32:29 PM · #92 |
Originally posted by TrollMan:
But if we could stop or limit ghost accounts which is undermining the entire voting process AND the DPC concept, I think we would make it better for all of the serious users. |
If you have actual evidence of "ghost accounts" controlled by specific users, then please submit a ticket with the relevant information, rather than calling into doubt the integrity of every other DPC member here in public. |
|
|
|
05/05/2010 03:44:11 PM · #93 |
Originally posted by GeneralE:
Originally posted by TrollMan:
But if we could stop or limit ghost accounts which is undermining the entire voting process AND the DPC concept, I think we would make it better for all of the serious users. |
If you have actual evidence of "ghost accounts" controlled by specific users, then please submit a ticket with the relevant information, rather than calling into doubt the integrity of every other DPC member here in public. |
You're missing the point. Remove the incentive for people to create ghost accounts and or get friends to sign up just to manipulate the vote. |
|
|
|
05/05/2010 03:46:56 PM · #94 |
Originally posted by Art Roflmao:
Originally posted by GeneralE:
Originally posted by TrollMan:
But if we could stop or limit ghost accounts which is undermining the entire voting process AND the DPC concept, I think we would make it better for all of the serious users. |
If you have actual evidence of "ghost accounts" controlled by specific users, then please submit a ticket with the relevant information, rather than calling into doubt the integrity of every other DPC member here in public. |
You're missing the point. Remove the incentive for people to create ghost accounts and or get friends to sign up just to manipulate the vote. |
I reckon we could stop having challenges and just do photo sharing. ...
:P |
|
|
|
05/05/2010 03:50:17 PM · #95 |
Originally posted by karmat:
Originally posted by Art Roflmao:
Originally posted by GeneralE:
Originally posted by TrollMan:
But if we could stop or limit ghost accounts which is undermining the entire voting process AND the DPC concept, I think we would make it better for all of the serious users. |
If you have actual evidence of "ghost accounts" controlled by specific users, then please submit a ticket with the relevant information, rather than calling into doubt the integrity of every other DPC member here in public. |
You're missing the point. Remove the incentive for people to create ghost accounts and or get friends to sign up just to manipulate the vote. |
I reckon we could stop having challenges and just do photo sharing. ...
:P |
I understand you are trying to make this a 'light discussion', but really, how is this kind of comment helpful to the discussion? I think Ken has a valid point and it should be discussed. |
|
|
|
05/05/2010 04:07:04 PM · #96 |
Yes, I'm trying to make it a bit lighter, but at the same time, consider what he said and what I suggested.
As long as it is a competitive site, there will *always* be an incentive to cheat. Even if the prize is "just" a colored .gif. The only way to take that incentive away is to take away the competition.
We currently, and actively monitor for ghost accounts, etc.
For every 10 or so "suspicions" of ghost voting we receive (or find), typically, only 1 or 2 is "substantial" meaning that there *might* be something there. Of those, very few turn out to be true ghost accounts.
What else can we do? |
|
|
|
05/05/2010 04:10:02 PM · #97 |
| Just a thought, but how about knocking out the low 5% and top 5% of scores? 5% was just the first number that came to my mind....... |
|
|
|
05/05/2010 04:13:42 PM · #98 |
I think Ken is right on the three entries requirement. But that is again, just my two cents on that. I cant see what it can hurt.
Message edited by author 2010-05-05 16:15:53. |
|
|
|
05/05/2010 04:15:19 PM · #99 |
Originally posted by JaimeVinas:
I think Art is right on the three entries requirement. But that is again, just my two cents on that. I cant see what it can hurt. |
I agree that is a good idea. I would also note that Member challenges seem to be suffering just the same as Open challenges so even $25 isn't a deterrent to "the voting issue", whatever that might turn out to be.
|
|
|
|
05/05/2010 04:26:16 PM · #100 |
Originally posted by karmat:
Yes, I'm trying to make it a bit lighter, but at the same time, consider what he said and what I suggested.
As long as it is a competitive site, there will *always* be an incentive to cheat. Even if the prize is "just" a colored .gif. The only way to take that incentive away is to take away the competition.
We currently, and actively monitor for ghost accounts, etc.
For every 10 or so "suspicions" of ghost voting we receive (or find), typically, only 1 or 2 is "substantial" meaning that there *might* be something there. Of those, very few turn out to be true ghost accounts.
What else can we do? |
It is about removing 'low hanging fruit'.
The effort required to create an account, submit three entries that go through voting, and then mal-vote is much much higher than allowing someone to creating an account and then immediately vote.
Message edited by author 2010-05-05 16:26:52. |
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 06/17/2025 02:46:51 AM EDT.
and I can prove it :P
