| Author | Thread |
|
|
02/01/2006 04:09:38 PM · #51 |
Originally posted by strangeghost:
So we invent trolls as a straw man because we just can't believe that anybody could really disagree with the majority to such an extreme degree. ...there will always be extremists. |
Oh, absolutely. We see a bell curve with hundreds of opinions go up, form a peak, and fall back down. Then there's a few blanks spaces and one more opinion wayyyyy out at the end by itself. I never said that extreme opinions didn't exist– only that they're annoying.
Originally posted by marcellieb:
I feel in a challenge you have an added criteria (above the norm Dof,composition,xposure etc) and that is how well the photo meets the challenge rules. |
Sure, and if you vote by those technical criteria AND how well the photo meets the challenge, you should arrive at a fair score. Why would you need a comparison photo to know whether an image was technically good and met the challenge?
Originally posted by rebs138:
...last semester I had a 65% average in a certain class and recieved an A as my final grade. The test scores were terrible, but someone had to get an A. If a professor failed every student in the class he wouldnt have a job for very long. |
...and if a professor routinely gives A's to students who failed or F's to those who clearly passed, he's not serving the educational system (or his students) very well.
Message edited by author 2006-02-01 16:12:13. |
|
|
|
02/01/2006 04:14:34 PM · #52 |
Originally posted by glad2badad:
Originally posted by marcellieb:
...I feel in a challenge you have an added criteria (above the norm Dof,composition,xposure etc) ...
|
Why is that added? That should be a given for ANY photo to be judged by.
Originally posted by marcellieb:
... I do this by comparing it to other photos in the challenge. |
Why? Let's say the challenge is "Animals". You take an "OK" shot of a horse, but someone else in this challenge has a "Great" shot of a horse. Based on comparison, your image should get a 1, 2, 3??? Your image by itself is decent. Maybe not an 8, 9, 10 - but at least a 5 or 6.
Sorry, not buying into this comparison voting one bit. JMO. ;^) |
Exactly - how can my avg/bad photo of a horse score say a 5 (avg) where a much better photo of a horse scores a 6 for instance. This would then give me a false sense of "goodness" - (lack of diferent word). Score my photo low if it deserves a low score and I will definately do the same - I will leave a comment on why though. I cannot expect my avg photo to get almost the same vote as a good photo.
|
|
|
|
02/01/2006 04:18:34 PM · #53 |
Originally posted by marcellieb:
how can my avg/bad photo of a horse score say a 5 (avg) where a much better photo of a horse scores a 6 for instance. |
Because a 5 might be in the bottom 20% of the challenge, and a 6 in the top 20%. If you gave the lesser horse photo a 1, you'd be putting in on par with a picture dim, blurry of a banana that had nothing to do with the challenge.
Message edited by author 2006-02-01 16:18:48. |
|
|
|
02/01/2006 04:19:22 PM · #54 |
Originally posted by marcellieb:
Originally posted by glad2badad:
Originally posted by marcellieb:
...I feel in a challenge you have an added criteria (above the norm Dof,composition,xposure etc) ...
|
Why is that added? That should be a given for ANY photo to be judged by.
Originally posted by marcellieb:
... I do this by comparing it to other photos in the challenge. |
Why? Let's say the challenge is "Animals". You take an "OK" shot of a horse, but someone else in this challenge has a "Great" shot of a horse. Based on comparison, your image should get a 1, 2, 3??? Your image by itself is decent. Maybe not an 8, 9, 10 - but at least a 5 or 6.
Sorry, not buying into this comparison voting one bit. JMO. ;^) |
Exactly - how can my avg/bad photo of a horse score say a 5 (avg) where a much better photo of a horse scores a 6 for instance. This would then give me a false sense of "goodness" - (lack of diferent word). Score my photo low if it deserves a low score and I will definately do the same - I will leave a comment on why though. I cannot expect my avg photo to get almost the same vote as a good photo. |
Sorry - I didn't give a score example for the "Great" horse image. It might get an 8 or 9. But in the end it doesn't matter WHAT the other image gets. Your shot (in the example) is ok/avg. A score of 5 WOULD be average and justifiable for an ok/avg image.
|
|
|
|
02/01/2006 04:20:44 PM · #55 |
Yikes! This has been fun, but I really need to get on with my 2 hour commute home now. :(
Take care, and don't forget to smile and keep having fun! ;^)
|
|
|
|
02/01/2006 04:22:45 PM · #56 |
Originally posted by scalvert:
Originally posted by strangeghost:
So we invent trolls as a straw man because we just can't believe that anybody could really disagree with the majority to such an extreme degree. ...there will always be extremists. |
Oh, absolutely. We see a bell curve with hundreds of opinions go up, form a peak, and fall back down. Then there's a few blanks spaces and one more opinion wayyyyy out at the end by itself. I never said that extreme opinions didn't exist– only that they're annoying.
Originally posted by marcellieb:
I feel in a challenge you have an added criteria (above the norm Dof,composition,xposure etc) and that is how well the photo meets the challenge rules. |
Sure, and if you vote by those technical criteria AND how well the photo meets the challenge, you should arrive at a fair score. Why would you need a comparison photo to know whether an image was technically good and met the challenge?
Originally posted by rebs138:
...last semester I had a 65% average in a certain class and recieved an A as my final grade. The test scores were terrible, but someone had to get an A. If a professor failed every student in the class he wouldnt have a job for very long. |
...and if a professor routinely gives A's to students who failed or F's to those who clearly passed, he's not serving the educational system (or his students) very well. |
Well - this way of voting helps me. Agree, disagree - its up to you. My scores compare well to the final scores.
|
|
|
|
02/01/2006 04:22:55 PM · #57 |
Originally posted by scalvert:
...and if a professor routinely gives A's to students who failed or F's to those who clearly passed, he's not serving the educational system (or his students) very well. |
You are missing the point... no one failed and no one had an A until the curve was fit to the class! That was the precise definition of who failed and who passed- THE CURVE! It depended on how you did relative to your peers. To me that is what a challenge is- you compete against the opponents IN the competition. |
|
|
|
02/01/2006 04:23:29 PM · #58 |
Originally posted by glad2badad:
Yikes! This has been fun, but I really need to get on with my 2 hour commute home now. :(
Take care, and don't forget to smile and keep having fun! ;^) |
|
|
|
|
02/01/2006 04:29:09 PM · #59 |
I don’t give ones but here is something I though about at the end of voting on the best of 2005.
I usually vote in placeholders 4-5-6 and then when I get through voting I go back and adjust the scores appropriately. Well even though we had a month to vote I did not get the chance to go back and adjust ALL the scores even though I did not vote 100%. So last night I was looking through the entries and see where I gave some pictures a 6 when they deserved a 8 or 9 and some pictures I placed in the 5 placeholder that deserved a 7 or 8. So I did those photographers an injustice. But it was NOT because I wanted to; it was because time ran out on me.
Could some people be using 1-2-3 as placeholders and not getting back to adjust the scores? It’s a thought even though I don’t know why someone would use 1-2-3 as placeholders. Because of my mistake I am not going to use the placeholder system again. I am going to take the time to vote an appropriate score the first time around so no one will have to suffer a lower score than they deserved because I did not get the chance to go back and adjust.
|
|
|
|
02/01/2006 04:29:18 PM · #60 |
Originally posted by rebs138:
Originally posted by scalvert:
...and if a professor routinely gives A's to students who failed or F's to those who clearly passed, he's not serving the educational system (or his students) very well. |
You are missing the point... no one failed and no one had an A until the curve was fit to the class! That was the precise definition of who failed and who passed- THE CURVE! It depended on how you did relative to your peers. To me that is what a challenge is- you compete against the opponents IN the competition. |
However, your "curve" depends on a final adjustment--meaning your "1" must be mitigated by other voters. Imagine if we all voted identically to you--then the result would be an "F" to a competent entry.
Besides, you are ignoring the larger comparison--what about the "grade" of a photo relative to other entries in previous challenges on this site? Or, relative to challenges outside of this site...in the larger world---if you really believe in relative grading you shouldn't doubt for a moment that students work is compared with students in other classes and prior years. And here we are again--why "fail" a photo that has merit--why can't the worst student in this year's class be far superior to the best from last year? (Most of my professors in university had no problem handing out a spate of A's when everyone succeeded in learning the material, by the way.) |
|
|
|
02/01/2006 04:31:26 PM · #61 |
Originally posted by KaDi:
|
Or DPC Radio.. on XM satelite ... that would rock.
|
|
|
|
02/01/2006 04:32:08 PM · #62 |
Originally posted by rebs138:
Originally posted by scalvert:
...and if a professor routinely gives A's to students who failed or F's to those who clearly passed, he's not serving the educational system (or his students) very well. |
You are missing the point... no one failed and no one had an A until the curve was fit to the class! That was the precise definition of who failed and who passed- THE CURVE! It depended on how you did relative to your peers. To me that is what a challenge is- you compete against the opponents IN the competition. |
To expand on this concept of "grading on the curve"; in the hypothetical "class", graded on the curve, it might be that the lowest SCORE on the test might be, say, a 78%, and when the professor grades to the curve this score might very well get a low GRADE, but there are two components; the score and the grade.
The analogy would be that in a hypothetical challenge between the site's masters, all the SCORES might be above 6, and rightly so, but the GRADE would still vary, with the highest 3 scoring averages receiving the ribbons.
Robt.
Message edited by author 2006-02-01 16:33:14.
|
|
|
|
02/01/2006 04:33:49 PM · #63 |
Originally posted by rebs138:
I can't vote that way. In a class I could get over a 90% on every assignment and end up with a final grade of a B or even a C in some cases. |
Understood, my wife got one and one half points off out of one hundred possible on a test in a class for her MBA and got a C on the test. To the student it is just the way it goes. To those of us in the work force,the professor did a bad job designing the test, the curve is not a reasonable standard.
In my work, I judge the performances of myself and others on an absolute scale. If all the work is below standard, it all gets ripped out and done again, the client won't pay for it, so I am not going to reward the guy who did the least bad job. If everyone does good work, we are all happy.
When your only paradime for judging is that of the school then the cruve might seem reasonable , but if you go for a PHD, all work is held to absolute standards, and how the rest of the feild did will not help or hurt any single candidate. I hold that this is the reasonable way to vote, HOWEVER I can't say how you ought to vote because as long as you vote and comment from time to time you are doing the right thing. |
|
|
|
02/01/2006 04:55:39 PM · #64 |
Originally posted by BrennanOB:
Originally posted by rebs138:
I can't vote that way. In a class I could get over a 90% on every assignment and end up with a final grade of a B or even a C in some cases. |
Understood, my wife got one and one half points off out of one hundred possible on a test in a class for her MBA and got a C on the test. To the student it is just the way it goes. To those of us in the work force,the professor did a bad job designing the test, the curve is not a reasonable standard.
In my work, I judge the performances of myself and others on an absolute scale. If all the work is below standard, it all gets ripped out and done again, the client won't pay for it, so I am not going to reward the guy who did the least bad job. If everyone does good work, we are all happy.
When your only paradime for judging is that of the school then the cruve might seem reasonable , but if you go for a PHD, all work is held to absolute standards, and how the rest of the feild did will not help or hurt any single candidate. I hold that this is the reasonable way to vote, HOWEVER I can't say how you ought to vote because as long as you vote and comment from time to time you are doing the right thing. |
In any event, "voting" = "scoring the test" and "average vote received" = "grading the test"; they are different things. It's not up to us, IMO, to grade the pictures; our job is to score them.
R.
|
|
|
|
02/01/2006 05:01:13 PM · #65 |
| You just gotta live with "ones", nothing you can do about it. It's all a part of DPC life. It is not so much a "troll" issue, but more an issue of someone who sees your photo as better than theirs and votes you a "one" so their photo has better chances, although that will never work. Or is that the definition of a troll? |
|
|
|
02/01/2006 05:23:00 PM · #66 |
With all due respect ... I dont think that the 1s come from "trolls"
stupid people...people whose circa 1994 .99 dot packard bell monitor are so out of whack that a razor sharp image is called OOF in a coment or only have a range of 1.5 stops on that same monitor having the frigging nerve to say that an image needs more contrast. No these are not trolls but self-important twits that dont understand what the he** dof is in an image (its fuzzy)... These are the ones dishing the ones IMHO... Sometimes I am simply amazed stunned and shocked at some of the scoring that comes from some of our sitemates... Is there anything that we can do about it? ... Nope. Except, accept them for what they are. Whatever that is
Thank you... I'm better now :)
Message edited by author 2006-02-01 17:25:36. |
|
|
|
02/01/2006 05:50:02 PM · #67 |
I've wondered about the effects of some people who give ones and two more often than others. I don't think it really has an effect on the general outcome, or the rank rather, 33rd, 34th, 35th best, etc. If each voter is fairly consistant in their voting criteria then we all get equally proportionate treatment. If a high scoring person votes on every entery and a low scoring person votes on every entry then each entry will get, from these two voters, a "good" score and a "lower" score. But when averaged together the better shot will still score higher than the worse one. In the end the rank should be about the same right?
Although without some voting "standards" you really can't compare two pictures from two different challenges by using the score as they probably voted on by different voters with different criteria.
As far as your "dpc career" average goes... Well if that really matters to you then... I don't know. I think you need a new hobby or something. |
|
|
|
02/01/2006 06:07:14 PM · #68 |
Originally posted by briantammy:
I've wondered about the effects of some people who give ones and two more often than others. I don't think it really has an effect on the general outcome, or the rank rather, 33rd, 34th, 35th best, etc. If each voter is fairly consistant in their voting criteria then we all get equally proportionate treatment. If a high scoring person votes on every entery and a low scoring person votes on every entry then each entry will get, from these two voters, a "good" score and a "lower" score. But when averaged together the better shot will still score higher than the worse one. In the end the rank should be about the same right?
Although without some voting "standards" you really can't compare two pictures from two different challenges by using the score as they probably voted on by different voters with different criteria.
As far as your "dpc career" average goes... Well if that really matters to you then... I don't know. I think you need a new hobby or something. |
Consistant voters have never been the issue from what I gather and if one voter is tougher than another, so be it as that is the right of individuality. Often placements are seperated by very narrow margins and after all, this is a challenge and placements mean alot to alot of people particularily when potentially ranking highly. Deliberate lowball voting on an image that clearly is above average is a fact and beyond dispute in my opinion. The motives behind these actions are anyones guess... sour grapes comes to mind as does the misguided belief that in a small way one is helping their own cause. You may say that in the end the effect is insignificant. I don't buy it... the effect is very significant to the person who receives a couple of 1's ten minutes before the challenge close. |
|
|
|
02/01/2006 06:08:52 PM · #69 |
Ah, this explains it! I was wondering how a picture that ranked 182nd place got already 232 views the day the results came out. Thanks pitsaman for including my picture (the underwear kids).
As for the 1's - different strokes for different folks, it's all for fun (at least for me). I got some REALLY nice comments from a number of people and 12 favs so far, so I'm happy - the picture brought a smile to many faces - that's enough for me. Some people just like landscapes, some like flowers, some like macros and some like people shots. I think people vote more for what they like and less on the technical merits of the picture. I personally do not hand out 1's, but I can't begrudge those that do - it IS one of the voting options. :-) |
|
|
|
02/01/2006 08:51:30 PM · #70 |
Curious, what has been the lowest score of a winning photo in challenge history?
Not to pick on anybody but the Surprise Challenge didn't have any entry that I thought were 10s yet the top ten received over a 100 perfect votes. Guess how many 1s the top 20 got? The answer: 1. You people are making a big deal about absolutely nothing. The only problem in the vote (if any) are the high votes not the low ones. Hell, I got nine 10s for my Best of 2005 entry and it certainly didn't deserve it in my opinion nor did it deserve the four votes of 1 but those votes cancelled out and then some. If anything I think people are too nice in their voting.
Message edited by author 2006-02-01 20:55:57. |
|
|
|
02/01/2006 09:07:49 PM · #71 |
edit...nevermind, I'll play nice. ;^)
Message edited by author 2006-02-01 21:12:23.
|
|
|
|
02/01/2006 09:26:59 PM · #72 |
And the psychologist in me has to also point out that our challenges highlight the behavior of a group, and can't tell us anything about the behavior of an individual voter. In virtually any group characteristic or behavior, you will have a range that often falls on a bell shaped curve when plotted. That means you will always have a large middle group, and two smaller "tail" groups. The typical pic in a challenge is always going to get some very low votes and very high votes. It's not trolls (no such creature has even been proven to exist to my satisfaction). It's the expected and normal behavior of a group. Really interesting or controversial pictures will often have very unusual vote distributions (luckily DPC plots it for us after a challenge is over), where there are multiple peaks. However, the group behavior, the bell curve, will virtually always be there. Some pics will get a lot of low votes, most pics will get some low votes, and even excellent pictures will get theirs too.
-------------------------------
Sorry, but I cannot go with you totally on this one. If voting was 100% subjective in nature, then I'd agree we should expect nothing more than normal distribution curves, reflecting different tastes etc.
But...in voting on pics, some weight must surely given to fairly objective issues (focus, exposure etc). To me, a voter giving a '1' indicates that the voter thought it was an all-round crap pic, on both objective and subjective grounds. If the pic was technically good, then I do not see how a '1' could possibly be justified.
Q.
texttext |
|
|
|
02/02/2006 04:37:09 AM · #73 |
Originally posted by scalvert:
Originally posted by David.C:
Voting against an imaginary ideal is comparing each entry against something that does not and can never exist. |
C'mon David... if somebody walked up to you with a single photo, surely you could tell them it was a good or bad picture without requesting a portfolio of comparison images to judge it against? Each of us has an idea of what a good, bad or average photo looks like, and we don't carry around a field guide for reference.
Voting by comparison with the competition is fundamentally flawed... if a hundred people entered the challenge with the same image taken with the same camera and settings under the same conditions, you might as well throw up your hands because you could never distinguish the good from the bad. |
A single photo taken by itself does not a competition make. But sitting that aside for the moment; when I look at an image I can give my opinion of it. But going any further than that and getting into what might or might not improve the image and I am no longer looking at the image -- I would then be looking at what that image could be, and perhaps IMO should be. But I wouldn't be looking at the image in front of me anymore. That's fine for general viewing, and is IMO what is asked for when critiquing, but within a contest I feel it is a disservice to the contestants to not base the vote on what they entered. So I don't base my vote on what I think the image could or should be -- I base it on what it is.
Perhaps it will help me be understood if I state how I vote. Like I've said, I don't base my vote on what the image is not, I don't imagine what it could be if it were done as well as it could possibly have been done and then compare. I don't start all images at 0 and add points until I run out of reasons to add points. I don't start all images at 10 and subtract until I find no more flaws. What I do is start in the middle, 5 or 6 because there is no 5.5, and then look at the image. I don't look at anything else, or imagine anything else -- I just look at the image. Does it move me? Does it speak to me? Does it draw me into it? Does the focus work? (please note this is not 'is it in focus?', it is 'Does the focus work?) Does the saturation work? (see focus) Does the composition work? Does it convey the challenge topic? All sorts of things like this, but they all have in common that they move the image away from the completely neutral, 'nothing to notice' middle of the road score I start them from. This is the first pass.
When I've finished going thru the images, I will strart looking over the thumbnail page. This pass can be considered the scoring equivalent of using Levels on an image. I move the top impacting images (both positive and negative to their respective poles. Keep in mind, I rarely have to move anything more than 1 point for the highest and lowest to reach their respective extremes. This is just a matter of using the entire voting scale, to give the distribution the greatest dynamic range possible. This seems to be what you disagree with, but there you have it. I start at the extremes (high first, because it's more fun) and decide based again on impact, if this one impacts me more or less than the others I gave the same score to (and how it compares to the scores above and below it). This is not a slide-rule comparison, but one of subjective impact -- the net result is those images that are able to impact me more than once will be moved more than those that were only able to impact me initially. This is my second pass and is a direct comparison of one challenge entry with another.
The third and final pass is for any additional commenting I may feel like doing. I comment as I score in both the first and second passes, but this pass is focused on commenting. Again, I start at the extremes and work my way in, commenting on anything I have something to say about. And again, I start with the upper extreme -- simply because they are more fun to comment. But keep in mind the images that impacted me heavily will likely already have a comment on them.
So there you have it, the first pass is focused on impact, the second on the repeatability of that impact and the third on comments. I will and have given a 1 and 10 in the first pass, which makes the second pass easier. But I find it interesting to note that I may not even know why I like (or dislike) an image to give it a good score. Jacko's entry in the first "Silhouettes" challenge is a good example as reading my comment will show. I gave it a 9 and commented on the first pass (I don't remember if I added it to my favorites at that time or after the challenge, but added it just the same). All without the slightest idea why I liked the image. I have a better idea of why I like it now, but not fully -- but it's impact hasn't changed.
A look at my average vote cast shows what I could have said before looking -- I give move 5's than 6's during the first pass, and they make up the bulk of the scores I give out. Looking at past challenges I see my average vote cast is fairly well in agreement with the average of the average score in the challenges. That is, some challenges have an average score lower, some higher, but my average vote cast compares well. It is lower than the 5.5 mean of the voting scale, but that is to be expected with the much larger number of beginners compared to those who know a thing or two about taking pics. I don't expect that to change any time soon -- and I'm certainly not going to start giving out higher votes just to see my average vote cast go up.
So, to sum up, I compare challenge entries against other challenge entries and I use the entire voting scale to do it in. The way I vote doesn't take long, and has me looking at and thinking about the images when they are in front of me far more than I would if I was just assigning a number based on some predetermined technical scale. I also find the images tend to clump together, so it is easy to say this clump are the 7's, this clump are the 3's and so on. There are a few that are hard to place; they require more time and attention to place well -- and the above general approach ensures they get it. It's fair, It's fast, it's easy and it make sure those images that deserve more attention get it.
BTW, I would have no problem at all rating 100 identical images -- they would all most likely get a 5 or a 6.
David
|
|
