| Author | Thread |
|
|
05/07/2006 07:41:26 PM · #1 |
Originally posted by Leok:
If you don't want low votes - don't enter challenges. I'll bet every photographer here has an image that has scored a 1 - this is how it should be. |
Rackat's "A Splash of Colour", the site's most-chosen favourite;
1 1
2 0
3 0
4 0
5 0
6 6
7 11
8 20
9 24
10 29
Even the best get 'em! |
|
|
|
05/07/2006 07:06:25 PM · #2 |
Thanks, RolandB. (blushing) She's a winner, for sure. I'm very proud of her...and my son! I'd like to take the credit, but I'm pretty sure it was luck. I was a young mom and didn't know what the heck I was doing most of the time...and when I grow up I want to be just like them. :)
Message edited by author 2006-05-07 19:07:24. |
|
|
|
05/07/2006 07:03:22 PM · #3 |
I'm not sure what the goal is here. Imagine the rules were changed to make voting low less likely e.g. by requiring comments on 1,2 and 3's. What will happen?
There will be less votes under 4. Many votes will bias their votes towards votes between 4 and 10. Everyone's scores will increase... presumably everyone will be happier?
Guess what though - the rankings should stay the same (think about it). The same images will ribbon.
We will just get threads on how all those voters giving out 4s are trolls and should be punished... oh wait this is one!
I don't understand how a 4 under the current system can be considered a troll or inappropriate vote. A troll gives a vote of 1. 4 is almost 5 - the score an average photo deserves.
I give lots of 4, 5, 6 and 7s - theses are the scores averagish photos deserve. I give fewer 1,2,3s and 8,9,10s. I know I have given a ribbon winner a 1 - the photo didn't do anything for me and it DNMC in my opinion. Obviously others disagreed. My average vote given is 5.22 - pretty much spot on fair IMHO.
This is the basic idea of voting - freedom of choice. Each voter has their own opinion and will thus vote differently to every one else.
If you don't want low votes - don't enter challenges. I'll bet every photographer here has an image that has scored a 1 - this is how it should be.
|
|
|
|
05/07/2006 04:25:41 PM · #4 |
Originally posted by Marjo:
I have lots of trolls, but no friends or family voting for me. Honestly, nobody...notta! I will admit that I asked my daughter to vote a long time ago when I first joined and she said, "Seriously, Mom, do you want to win that way?". Shut me up rather quickly. ;) |
You certainly raised her RIGHT!!! Adds to my faith in my fellow man :)
|
|
|
|
05/07/2006 04:23:56 PM · #5 |
Originally posted by ClubJuggle:
Originally posted by seenosun:
I have made suggestions in the past about improvements to the system as many others have. Those suggestions have fallen on deaf ears. What I've seen happen is some agree, some disagree and ultimately nothing changes. So why waste my time and energy trying to improve a site that doesn't want to improve. Why waste my time trying to create art for a site that doesn't appreciate art. Do you think a 12th place image deserves nine 1 votes? Here's a suggestion. How about a challenge called Guns, Nipples and Flowers?
The winner is the photo with the most 1's! |
While this is an interesting point, your post fails to mention that you also received more votes of 10 than 7 out of the 11 entries ranked above you. For better or worse, there is a contingent of voters on DPChallenge who subscribe to a "boobies=10" philosophy and vote up any nude shots that are posted. Is it fair that your entry should benefit from this bias?
Art is not created in a vacuum, and whether we like it or not, it is not viewed in a vacuum either. Just as we as artists bring our own experiences and biases to the creation of art, each viewer brings his or her own bias into their interpretation and evaluation of it. That's just human nature, and I frankly don't see any realistic way to get around it, nor do I necessarily think we need to. I believe that our voting system does a reasonably good job of figuring out how our photographs will appeal to a wide audience. It's certainly not the only way to judge art, but it's one that seems to work reasonably well for our purposes. That said, I'm always willing to discuss something better.
~Terry |
If you review the Avg stats of my Photo for low key you will see exactly where the problem is...
Avg (all users): 6.412
Avg (commenters): 7.760
Avg (camera): 6.500
Avg (no camera): 3.714****
I think throwing out the bottom 10% / Top 10% idea is a good one but....i'm sure NOTHING will change. This is the way DPC is.
|
|
|
|
05/07/2006 04:10:20 PM · #6 |
| I'm starting to understand where things are going wrong for me. I'd like to see the voting period extended by another week. That way all the '10' votes I should have got will starting coming in, and help me recover from my 5.5 |
|
|
|
05/07/2006 03:52:42 PM · #7 |
Originally posted by ElGordo:
Then there are those that are simply malicious and vote low on images they perceive to be of higher quality. |
Total conjecture... Show me this troll voter... |
|
|
|
05/07/2006 03:45:39 PM · #8 |
Originally posted by PhantomEWO:
If ya want suggestions:
Throw out the top 10% and bottom 10% of vote scores on each image, then calculate the average on close of voting. That gets rid of most trolls and most friend votes. |
I have lots of trolls, but no friends or family voting for me. Honestly, nobody...notta! I will admit that I asked my daughter to vote a long time ago when I first joined and she said, "Seriously, Mom, do you want to win that way?". Shut me up rather quickly. ;)
|
|
|
|
05/07/2006 03:43:12 PM · #9 |
You know what I just accept what vote I receive. Yes I would like to have a high scoring photograph. I have entered 60+ challenges and I only have broken received a 6.5 or > once, a 6.0 or > four times. Never a 7.0 or >, One TOP 10, and NO ribbons.
But its like the old saying goes; You can satisfy some of the people all the time. You can satisfy all the people some of the time, But you will never satisfy all the people all of the time.
I just have developed thinker skin and enjoy my pictures and try to learn.
-SDW
|
|
|
|
05/07/2006 03:30:09 PM · #10 |
As an example of why I think troll-voting is a fallacy, I know of one image in particular which one a blue ribbon with a score well over 8.0. The photograph had a total of less than 20 votes of 5 or lower (more than half of which were 5's), and only a single vote of 1, no twos and a single 3. Surely, that vote of 1 was a troll!
As it turns out, I happen to know that vote came from one of the administrators. Since I think we all agree that Drew and Langdon are not trying to disrupt their own site, can we just accept that even the most popular photographs on the site might be disliked by someone?
~Terry
|
|
|
|
05/07/2006 03:29:58 PM · #11 |
Originally posted by PhantomEWO:
If ya want suggestions:
Throw out the top 10% and bottom 10% of vote scores on each image, then calculate the average on close of voting. That gets rid of most trolls and most friend votes. ... |
No!... No!... absolutely not!! Eliminate "friend" votes??? No! That would be heresy! LOL!!!!!
|
|
|
|
05/07/2006 03:21:30 PM · #12 |
If ya want suggestions:
Throw out the top 10% and bottom 10% of vote scores on each image, then calculate the average on close of voting. That gets rid of most trolls and most friend votes. The contests would be closer and winners would be a very tight margin. The average photos would be clumped nicely in the middle. Could the SC's run a couple of old challenges under this and see how much difference it makes. If indeed the best photos are pushed out of the competition then it's a bad idea. AND just because the top 5 were there before and not after that may be a more accurate representation of the better photos.
|
|
|
|
05/07/2006 03:20:00 PM · #13 |
Originally posted by seenosun:
I have made suggestions in the past about improvements to the system as many others have. Those suggestions have fallen on deaf ears. What I've seen happen is some agree, some disagree and ultimately nothing changes. So why waste my time and energy trying to improve a site that doesn't want to improve. Why waste my time trying to create art for a site that doesn't appreciate art. Do you think a 12th place image deserves nine 1 votes? Here's a suggestion. How about a challenge called Guns, Nipples and Flowers?
The winner is the photo with the most 1's! |
While this is an interesting point, your post fails to mention that you also received more votes of 10 than 7 out of the 11 entries ranked above you. For better or worse, there is a contingent of voters on DPChallenge who subscribe to a "boobies=10" philosophy and vote up any nude shots that are posted. Is it fair that your entry should benefit from this bias?
Art is not created in a vacuum, and whether we like it or not, it is not viewed in a vacuum either. Just as we as artists bring our own experiences and biases to the creation of art, each viewer brings his or her own bias into their interpretation and evaluation of it. That's just human nature, and I frankly don't see any realistic way to get around it, nor do I necessarily think we need to. I believe that our voting system does a reasonably good job of figuring out how our photographs will appeal to a wide audience. It's certainly not the only way to judge art, but it's one that seems to work reasonably well for our purposes. That said, I'm always willing to discuss something better.
~Terry
|
|
|
|
05/07/2006 03:17:18 PM · #14 |
Originally posted by ElGordo:
Think about what you are saying! Are there as many 10s as there are 4s? For the exceptional image there may be as many 7s as there are fours. Frequently the distribution of scores is not a bell curve but is skewed to one side or the other (that's a not necessarily bad) and when a controversial subject appears the distribution may even be bimodal. |
FWIW, a "bell curve" can exist even when not all the vote values are used. for example:
1 0
2 0
3 0
4 1
5 16
6 44
7 49
8 36
9 29
10 28
That's the vote distribution for this: 
It peaks at 7, falls off in both directions, and has a single outlier vote of 4, nothing else less than 5. It's a highly compressed curve because 4 of the 10 numbers are essentially out of play. It shows consensus that the image is decent across the board, with considerable variation of opinion as to just how excellent it actually is.
Compare that with this one:
1 0
2 1
3 2
4 4
5 25
6 32
7 31
8 53
9 62
10 67
That's JJ's very fine "singled out" entry: 
Here we see a steady climb, no fall-off to the curve at the high end. We can dismiss the 7 outlier votes as "DNMC" knock-downs because the image didn't use shallow DOF, which was mentioned in the challenge description, and we have a very nicely rising curve with a slight hiccup between 6-7 and a general agreement that the image is terrific.
Speaking as a competitor, that's the holy grail: an image that has more 10's than anything else, and a nicely ascending curve to lead up to them. Both are blue ribbon images. The sunset shows ambivalence among voters (it's like winning the presidential election by a handful of votes in the electoral college) while JJ's masterpiece is the equivalent of sweeping 45 states or something :-)
R.
|
|
|
|
05/07/2006 03:08:13 PM · #15 |
Originally posted by bod:
Originally posted by ElGordo:
I did not 'quote' anyone in my statement, I simply made a statement, right or wrong. You obviously missed the point as well. |
Originally posted by ElGordo:
Think about what you are saying! Are there as many 10s as there are 4s? |
Okay, so you asked a question of the person you quoted, but changed the numbers to suit your own ends. Same shady trick. Same difference. |
So if he had said 9 instead of 10 would that change the point he is making? Frankly, I disagree a bit from El Gordo not because of his premise but because of his example of a troll vote being a 4. Troll votes in my mind are 1s, 2s and 3s and there is little to no balancing out those votes. 4s can mean someone is using the 4, 5 and 6 as placeholders where a 6 could be a photo that ends up being a 9 and a 4 may get bumped up to a 5 or 6.
Message edited by author 2006-05-07 15:10:21.
|
|
|
|
05/07/2006 03:05:07 PM · #16 |
Originally posted by ClubJuggle:
Originally posted by cislander:
Why not require a comment on votes of 1 or 2. This way if it's a troll vote, the troll would be identified. If the image truly is poor, then the comments would go to helping that member improve their work which is, I believe, the purpose of this site. No? |
Our past experience with required commenting shows that it rarely achieves its intended goals. ... |
Perhaps comments should be required for all votes of 4 or below.
This will result in higher average DPC scores and fewer general complaints about low scores. Voters rate images to low anyway. It is a win-win situation. :) (Except for that tiny little issue of compulsory comments)
|
|
|
|
05/07/2006 03:04:19 PM · #17 |
| I like your take on this, Dan. Voting the full range 1-10 is the way to make your votes count most effectively. But do you consistantly vote low on all images? Not likely. Your 1 votes are honest and not malicious. I have given 1 votes as well and rarely some 10s. But this is not about honest disagreemant of image quality, rather malicious votes consistantly cast to affect the placement of excellent entries. These malicious voters are not stupid (for the most part) and seem to have a discerning eye for the quality image. Mediocre images, like mine, are largely unaffected. Looking forward to another well deserved 4.8 in a current challenge! There may be no reasonable solutions to the problem, and not all agree that it is a problem! |
|
|
|
05/07/2006 02:52:13 PM · #18 |
Originally posted by cislander:
Why not require a comment on votes of 1 or 2. This way if it's a troll vote, the troll would be identified. If the image truly is poor, then the comments would go to helping that member improve their work which is, I believe, the purpose of this site. No? |
Our past experience with required commenting shows that it rarely achieves its intended goals. Even if we require comments, there's nothing to stop a voter from leaving a comment of, say, "nice photo" or "well done" to accompany their vote of 1. Most likely, such a requirement would do little to deter the very few trolls that exist, and only result in a stream of hostile PM's to the vast majority of honest users who cast their votes in good faith.
~Terry
|
|
|
|
05/07/2006 02:51:29 PM · #19 |
Originally posted by ElGordo:
I believe there is sufficient variety in the challenges that no honest voter would consistantly, over many challenges and votes, produce a pattern of significantly low voting. And I agree, we do not wish to create a strictly conformist venue, but that seems to be what is happening anyway. Out of the box entries usually score abyssmally, though there are notable exceptions. |
No they don't I have proof, thats a common misconception. Outta the box will score quite well, if its nicely composed, perfect exposure, just generally pleasing to the eye. |
|
|
|
05/07/2006 02:49:51 PM · #20 |
To explain "unwarranted" low votes DPCers seem willing to believe most any troll or conspiracy theories and think that the system is broken and needs fixing.
Two possibilities some folks just can't seem to accept are:
1-Some individual voters just might not like a picture
2-Gasp! They might be responsible for giving low scores!
Message edited by author 2006-05-07 14:51:21.
|
|
|
|
05/07/2006 02:49:13 PM · #21 |
Originally posted by Qart:
My guess is that if the voting scale started at 0 or even -10, there would be those for whatever reason would be compelled to click the lowest number available. I've never agreed with those that say that a low score has little overall impact, particularily when an image is scoring rather well. |
I use the full rating scale when I vote on DPC challenges. I usually give a few 1's and a few 10's. If the scale were -10 to +10, I would give a few -10 scores and a few +10 scores in most challenges.
There is a good reason to do this. It makes my vote count more. In the current system, my votes can make a 9-point difference in the total. If you don't give 1's and 2's, your vote can only make a 7-point difference.
If I'm going to take time to vote, I don't see much point in doing something that reduces the impact of my votes. I will continue to gives 1's and 2's to the images that I think are least successful in each challenge.
Every vote does count, but with 250 people voting, individual votes don't count MUCH. This is true whether the individual votes are 1's, 5's, or 10's. I want my votes to count as much as possible.
--DanW
|
|
|
|
05/07/2006 02:40:31 PM · #22 |
Originally posted by mk:
Originally posted by ElGordo:
I think that a system which detects a consistant pattern (more than one challenge and many votes) of significant deviation (how much is enough?) on the low side from the median vote would result in a warning to the deviating voter and continued abuse would result in suspension from DPC. I believe it is unlikely that anyone would maliciously vote high on an entry, but I suppose that is a possibility.
The present system is pretty good but could use some improvemnts IMO. Please share your thoughts! |
I think that any system that essentially requires you to vote like everyone else would be boring. Look at the number of already existing complaints about how it's impossible to do well with an "out of the box" shot. Then imagine getting penalized for having an opinion that was different that other people's. Why can't we just accept the fact that you're going to get some low votes, whether you like it or not? I don't think voting on photography is supposed to be a precise work of machinery. |
I believe there is sufficient variety in the challenges that no honest voter would consistantly, over many challenges and votes, produce a pattern of significantly low voting. And I agree, we do not wish to create a strictly conformist venue, but that seems to be what is happening anyway. Out of the box entries usually score abyssmally, though there are notable exceptions. |
|
|
|
05/07/2006 02:39:18 PM · #23 |
Thats really well said mk.
This pile of junk got two tens. Whats up with that?
There are going to be some real high and low scores, thats just the way it is.
I know another site...and the high is about a 6 and the low a 4. Because they have a system in place for low voters. Well that sucks, you should be able to vote any way you want and let the higher ups take care of the trolls. |
|
|
|
05/07/2006 02:35:51 PM · #24 |
ima get into this one as well.
i like my images that is why i submit them. ifn i thought i was submitting a 3 average image why would i do it?
the beauty of DPC is simply that there are gallons of people that make their opinion known about my image. do i like it when i get hit with gallons of 1's, absolutely not.
i agree with the earlier post that the system is not broken. there is no perfect system. some "troll" (i hate that name calling stuff) will always wanna score my images below what i think it should score.
why is is so difficult to imagine that other people are actually different with different tastes? why is it horrible that some people dont like nudity?
i have a canon, some have nikons, some have pentax, etc. some score high, some score low, some just try to mess up the system. that will always happen.
ifn the system gets changed, the people that whine about "troll" voting will still whine about it. there will always be a "troll" to them.
|
|
|
|
05/07/2006 02:29:16 PM · #25 |
Originally posted by ElGordo:
I think that a system which detects a consistant pattern (more than one challenge and many votes) of significant deviation (how much is enough?) on the low side from the median vote would result in a warning to the deviating voter and continued abuse would result in suspension from DPC. I believe it is unlikely that anyone would maliciously vote high on an entry, but I suppose that is a possibility.
The present system is pretty good but could use some improvemnts IMO. Please share your thoughts! |
I think that any system that essentially requires you to vote like everyone else would be boring. Look at the number of already existing complaints about how it's impossible to do well with an "out of the box" shot. Then imagine getting penalized for having an opinion that was different that other people's. Why can't we just accept the fact that you're going to get some low votes, whether you like it or not? I don't think voting on photography is supposed to be a precise work of machinery. |
|
