| Author | Thread |
|
|
12/03/2008 09:29:29 AM · #26 |
Originally posted by aKiwi:
My photo  finnished 5th with 6.5366.
On the second day of voting it got up to 7.2 As you can imagine I was over the moon. But then from Thursday midday on the score plummeted all the way down to 6.3!
My score curve is pretty flat.
10 x 10
17 x 9.
But
2 x 1
and 7 x 2! Holy SH! So many people thought it was that bad! |
No, so many people thought it was not meeting the challenge. It takes a certain level of sophistication to interpret this as "backlighting", IMO. It's a wonderful image, but I'm surprised it scored as well as it did, frankly.
R. |
|
|
|
12/03/2008 09:29:44 AM · #27 |
Originally posted by smichener:
The problem with the 10 point rating scale we use is that a score of 5 is actually in the lower half of possible scores, while a 6 is in the upper half. This is a "forced choice" set up when there are an even number of rankings. My guess is that when people give a score of 5 they are intending to say the photo is a "neutral" or midpoint photo, but they haven't really thought that a 5 is actually in the lower half of possible scores.
Two solutions to this problem.
1) Make the rating scale a 9 point (odd)scale which then makes a score of 5 a "neutral" score.
2) Give definitions to the numbers. For instance 1=DNMC/worst, 3=needs work, 5=neutral, 7=very good, 9=one of the best. |
Good ideas - the fact that the very best photo in backlighting was only just above a neutral score just seems a bit silly.
If you got a 6.5/10 on a test at school the comment would have been "Must try harder, see me!". I'm talking to you  IreneM - you're only letting yourself down ;) |
|
|
|
12/03/2008 09:35:00 AM · #28 |
Originally posted by Bear_Music:
No, so many people thought it was not meeting the challenge. It takes a certain level of sophistication to interpret this as "backlighting", IMO. It's a wonderful image, but I'm surprised it scored as well as it did, frankly.
R. |
Herein lies the problem IMO, a lack of voter sophistication, if it's not blindingly obvious what you're looking at people automatically give you a low score. |
|
|
|
12/03/2008 09:51:29 AM · #29 |
Originally posted by Bear_Music:
No, so many people thought it was not meeting the challenge. It takes a certain level of sophistication to interpret this as "backlighting", IMO. It's a wonderful image, but I'm surprised it scored as well as it did, frankly.
R. |
I think you are probably right there.
When I submitted I was worried about that.
I nearly submitted another photo that IMHO isn't as good, but showed the back lighting better.
This one here actually.
I guess it goes to show that submitting a winning photo on DPC is not only a photographic task.
1 The photo has to be good.
2 The photo has to look good as a thumbnail for the skim voters.
3 It must be blindinggly obvious that it meets the challenge so it doesn't get DNMC votes.
4 No Nudity for the prudes.
5 No Patriotism
6 No religion
I can see that the fill flash challenge will get some skewed votes as in my opinion good fill flash can't be seen as being there. The less experienced voters wil probably vote for a DNMC. If I choose that topic I'm going to NUKE my subject so It is blindingly obvious that flash was used.
|
|
|
|
12/03/2008 09:53:57 AM · #30 |
Originally posted by IreneM:
Originally posted by JimiRose:
Really I'd love to know just what's happening to the voting at the moment. Not trying to take anything away from the top scorers (in fact quite the reverse), but Backlighting IV was won with a 6.69 and IreneM received 17 votes under 5. |
On the second day of the challenge (last Thursday) my score went from 7.2 to 6.5 in 2 hours and that's where it stayed till last night when I went to bed. I was amazed that it nearly went to 6.7 in the eight hours I slept ;-) |
Glad I wasn't the only one that was suffering that score drop. Mine actually went down whilst I slept. |
|
|
|
12/03/2008 09:54:09 AM · #31 |
Originally posted by JimiRose:
Good ideas - the fact that the very best photo in backlighting was only just above a neutral score just seems a bit silly.
If you got a 6.5/10 on a test at school the comment would have been "Must try harder, see me!". I'm talking to you IreneM - you're only letting yourself down ;) |
The actual scale of achievable final scores gets compressed because of the averaging of all opinions. Nearly all entries will come in between a 3.0 and 8.0. That's only half the actual scale that is being used for voting. Another reason to heavily interpret ones average score and not take it too much to heart. |
|
|
|
12/03/2008 09:55:53 AM · #32 |
Originally posted by smichener:
Two solutions to this problem.
1) Make the rating scale a 9 point (odd)scale which then makes a score of 5 a "neutral" score.
2) Give definitions to the numbers. For instance 1=DNMC/worst, 3=needs work, 5=neutral, 7=very good, 9=one of the best. |
We already have the second one.
1=Bad 10=Good. DPChallenge: It's like Animal Farm on growth hormone. |
|
|
|
12/03/2008 09:59:28 AM · #33 |
| The 10 seems to be this almost unattainable thing...kind of an illusive hope for the photographer. Plus, some people vote on, it seems, 'the prettiness', and then some people seem to actually critique and point out the exact things that may be good/bad about the pic. |
|
|
|
12/03/2008 10:05:32 AM · #34 |
Keep telling you people, there is a database here; SCs know who is voting 1s and 2s to winner photos, why don't we ask how come? Let̢۪s track them down, go to their houses, sit on kitchen table and ask if they have any problems in their lives that need to be fixed. Or just take their computers away for a year or two...
We need answers damned, why 1's and 2's to great photos? This turn into soccer match each game ends 0-0, we need a new system here like football (American football that is) one touchdown, 6 points... we need to improve the numbers here, not drop them. We need to see 7s and 8s and 9s as winners... not 6.
Yeah, here I said it (again) :P |
|
|
|
12/03/2008 10:11:37 AM · #35 |
I will admit it is a lot based on taste.
My 10 yr old just started submitting to challenges, and I have sat with him whilst he votes. I have to really hold my tongue as he votes on what appeals to him and not what I would vote. I figured I wouldn't say anything when he gave a high score to what I would normally give a low score to.
I did comment to him a couple of times when he gave a low score to what I thought was a technically treat photo but tried to keep out of it so as not to affect the voting.
Just remember the Huge age range and cultural differences in all the members here. |
|
|
|
12/03/2008 10:26:41 AM · #36 |
Originally posted by JimiRose:
Originally posted by HawkeyeLonewolf:
But just because something's been done for thousands of years does not make it publicly proper for viewing. |
Quick, someone get the Statue of David a fig leaf. |
Or just give him a fork to poke his eyes out so he doesn't have to see himself, or anyone else naked. |
|
|
|
12/03/2008 10:32:27 AM · #37 |
Originally posted by FocusPoint:
Keep telling you people, there is a database here; SCs know who is voting 1s and 2s to winner photos, why don't we ask how come? Let̢۪s track them down, go to their houses, sit on kitchen table and ask if they have any problems in their lives that need to be fixed. Or just take their computers away for a year or two...
We need answers damned, why 1's and 2's to great photos? This turn into soccer match each game ends 0-0, we need a new system here like football (American football that is) one touchdown, 6 points... we need to improve the numbers here, not drop them. We need to see 7s and 8s and 9s as winners... not 6.
Yeah, here I said it (again) :P |
Who says they're great photos? You? You and a bunch of others? You'd really tell someone that their opinion is "wrong". Would you also go through their wardrobe and dress them? Maybe re-decorate their homes to suit your taste too? Hell, why don't we just declare you King of Opinion and let you decide what everyone else should think? |
|
|
|
12/03/2008 10:39:45 AM · #38 |
Originally posted by Gordon:
There are probably people who reserve 1s for DNMCs and pigeons. |
No! Say it ain't so! Not for pigeons!!! |
|
|
|
12/03/2008 10:53:58 AM · #39 |
Originally posted by Spazmo99:
...Hell, why don't we just declare you King of Opinion and let you decide what everyone else should think? |
I like this idea :D |
|
|
|
12/03/2008 10:57:11 AM · #40 |
My first opinion is, whoever gave 1 or 2 to this photo sucks in photography. Flipping burgers would be recommended job instead. If not a job, please use your camera as a paperweight instead. |
|
|
|
12/03/2008 11:34:43 AM · #41 |
Originally posted by FocusPoint:
Keep telling you people, there is a database here; SCs know who is voting 1s and 2s to winner photos, why don't we ask how come? Let̢۪s track them down, go to their houses, sit on kitchen table and ask if they have any problems in their lives that need to be fixed. Or just take their computers away for a year or two...
We need answers damned, why 1's and 2's to great photos? This turn into soccer match each game ends 0-0, we need a new system here like football (American football that is) one touchdown, 6 points... we need to improve the numbers here, not drop them. We need to see 7s and 8s and 9s as winners... not 6.
Yeah, here I said it (again) :P |
Why is that? What if the contest was entered by all 6s?
Some people vote where they give a 10 to the best photo. I think this is wrong, personally, but it's their choice.
I prefer to "rate" the photo and let the ratings sort themselves out I rarely, if ever, give out 10s or 9s.. most of the photos don't earn them.
JMO.
|
|
|
|
12/03/2008 11:39:49 AM · #42 |
I remember a thread started about a voter who rates pics with a point system...
X creativity,
X technicals,
X 'interestingness'
etc etc
...and people (including me) found it a bit weird...however, I'm beginning to see the logic behind it...although best combined with the stray explanation...
Message edited by author 2008-12-03 11:40:27. |
|
|
|
12/03/2008 01:53:10 PM · #43 |
Originally posted by yospiff:
Originally posted by JimiRose:
Good ideas - the fact that the very best photo in backlighting was only just above a neutral score just seems a bit silly.
If you got a 6.5/10 on a test at school the comment would have been "Must try harder, see me!". I'm talking to you IreneM - you're only letting yourself down ;) |
The actual scale of achievable final scores gets compressed because of the averaging of all opinions. Nearly all entries will come in between a 3.0 and 8.0. That's only half the actual scale that is being used for voting. Another reason to heavily interpret ones average score and not take it too much to heart. |
Yes the final score is an average of all the votes. However, my point was that the individual voter working with a 10 point scale is likely to see 5 as "average" when in fact it is below the center point. Since we can't vote a 5.5 many voters go with a 5. This keeps scores lower than they would be with a 9 point scale.
If this was research, most trained researchers would probably suggest a 9 point scale over a 10 point scale. The flaw is not the voters, it's the tool they are using to vote with. |
|
|
|
12/03/2008 02:46:32 PM · #44 |
| I think it would be quite interesting one week to have a second score for each image that is allocated by a panel from DPC. In that way we could get an interesting comparison between two styles of voting. |
|
|
|
12/03/2008 03:04:54 PM · #45 |
Originally posted by Lutchenko:
I think it would be quite interesting one week to have a second score for each image that is allocated by a panel from DPC. In that way we could get an interesting comparison between two styles of voting. |
We already have subsets of scoring and comparisons of different voting groups. The following example is stats taken from a recent Top-10 challenge entry.
Avg (all users): 6.0877
Avg (commenters): 6.9231
Avg (participants): 5.7561
Avg (non-participants): 6.1604
Average Vote per Users w/ Cameras: 5.756
Average Vote per Users w/out Cameras: 6.160
|
|
|
|
12/03/2008 03:33:43 PM · #46 |
Here's example from the Backlight challenge.
Place: 43 out of 237
Avg (all users): 5.8854
Avg (commenters): 7.2500
Avg (participants): 5.3818
Avg (non-participants): 6.1569
Views since voting: 42
Views during voting: 251
Votes: 157
Comments: 8
|
|
|
|
12/03/2008 03:42:04 PM · #47 |
Originally posted by JimiRose:
Originally posted by Bear_Music:
No, so many people thought it was not meeting the challenge. It takes a certain level of sophistication to interpret this as "backlighting", IMO. It's a wonderful image, but I'm surprised it scored as well as it did, frankly.
R. |
Herein lies the problem IMO, a lack of voter sophistication, if it's not blindingly obvious what you're looking at people automatically give you a low score. |
I don't see how you can say that: the image ended up in 5th place! I don't think any of the four that beat it are unworthy of their positions. The discussion was about OUTLIER votes; I offered an explanation for the low votes. I don't think it's unreasonable that *some* people thought this was not an especially good example of the category "backlighting"; for some people, this shot would not qualify as backlighting at all. I don't see what the fuss is about, the image did very well.
R.
Message edited by author 2008-12-03 15:42:33. |
|
|
|
12/03/2008 04:21:53 PM · #48 |
Here's a good example of why you need to interpret your challenge performance by more than just the final average:
The average vote was 5.25, which at first glance, would tell me "good, but nothing outstanding". However, looking at the voting breakdown, I see that about 40% of the voters thought it was worth a 6 or better. 32% were neutral with a 5, and the average was brought down by a lesser 28% who it failed to appeal to much. (Also notice that 28% did not leave any comments explaining why.)
The average commenters vote was 6.83! All this additional information tells me that this is actually very good photography, but it only had strong appeal for a limited group. |
|
|
|
12/03/2008 04:22:52 PM · #49 |
Originally posted by Bear_Music:
I don't see how you can say that: the image ended up in 5th place! I don't think any of the four that beat it are unworthy of their positions. The discussion was about OUTLIER votes; I offered an explanation for the low votes. I don't think it's unreasonable that *some* people thought this was not an especially good example of the category "backlighting"; for some people, this shot would not qualify as backlighting at all. I don't see what the fuss is about, the image did very well.
R. |
Amen!
I am totally happy with that, and sorry if I came across as not being happy with that.
What ticked me a little was that the score climbed to 7.2 on Thursday. I was very excited, and happy I didn't have a refresh button or I wouldn't have got any work done this week. But after Thursday every time I looked at the image the score was falling. Almost to 6.2. Luckily it came up a bit.
Irene was complaining about the same thing.
Maybe we should allow votes on Wednesday and Sunday only. |
|
|
|
12/03/2008 04:56:11 PM · #50 |
Oooh a double post!
Message edited by author 2008-12-03 16:57:04. |
|
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: 11/10/2025 10:58:14 AM EST.
