|
| Author | Thread |
|
|
07/11/2004 03:29:59 PM · #51 |
This is a very good discussion with a lot of valid points made. One that rang true for me regarded the perceived benifits of a karma system. Really the only thing we can point to is the fact that it should discourage those who really troll vote, that is, consistently vote way outside the norm, and do so for the bulk of a challenge.
I would submit, though, that the current situation does not warrant such a change. We really have very few voters who practice voting "strategies" that would be affected by the karma system, and therefore the karma system would provide little or no benefit in this area.
On the other hand, the karma system would disrupt the statistics and make the voting database and challenge calculations much more complex.
The idea of not seeing the thumbnail view until some milestone, e.g. 20% or full voting, has been passed has been discussed before and seems to be widely supported. Unless it is completely eliminated until after a voter finishes voting, there are ways (as has been posted previously) to circumvent the system. If you can't see the thumbnail page until you're finished voting (and you are not able to vote further) there is really little value in the thumbnail page at all.
I do feel that eliminating the thumbnail page would have a downside. Many voters, possibly a majority, go back and bump up/down after their initial vote. I do this, and I find that I change scores on a number of photos each challenge. It's a gret "sanity check." I think the user community would not support completely eliminating this feature.
That said, I do think the "selective voting" is a significant problem. The forums are loaded with discussions of scores, including vote totals; very often the vote totals for images will vary widely at nearly identical times. If they were voted on random fashion, everyone should receive nearly the same number of votes. I think that a change should be made to discourage selective voting, but I'm conflicted about what is the best path.
Disclaimer: The opinions expressed above are my own personal opinions.
|
|
|
|
07/12/2004 10:24:46 AM · #52 |
Hi!
I read this thread and it's quite interesting. There is no doubt that the voting system could be improved. I don't like the two separate pools voting. That would give 2 winners for each challenge? If not, all the results would have to be merged, therefore this is not resolving the problem that it's supposed to solve.
I like the Karma system, but as people pointed out, this tend to punish people whos tastes tend to differ from the majority. Lets say someone is really likes gore or something, he might not like smooth pictures of people or pink smelly flowers, but will prefer a dark content, black rose, blood and all. Should he be punished because of his tastes? I don't think so. And another good point is that people like to compare their pictures from one challenge to another to improve themselves. With Karma voting scheme, they will not be able to compare past with future pictures.
I would propose a few things instead:
1- The addition of a meta-data when submitting a picture, a new box where the submitter tells how he interprets the picture to meet the challenge. This would be visible to all voters.
2- I would keep the bad voting scheme detection.
3- The addition of a checkbox in the voting system. This box would be called "Doesn't meet the challenge". From there I see two possible ways to deal with that checkbox:
- You can either vote or check the box, they are exclusive from one another. If a picture gets 25% boxes checked at the end of the voting period, it's disqualified.
- You can always vote (or not if you don't want to), even if you checked the box, but you are asked to vote as if the picture was in a proper challenge. If a picture gets 20% boxes checked, it's score is lowered by the same amount as the percentage of checked boxes. For example, I get 7 on a picture, but there are 30% of the voters who have checked the box, my final score would then be 4.9 (7 - 30%).
4- I would balance the voting. Let's say the average vote is 6.6 in a whole challenge. My personnal voting is 4.8, the system could automatically raise my average to 6.6 or so. With statistics this is quite easy to do. That would make sure that everyone has a "same voting value", as there are people who tend to give high scores to please, there are others who score low because they judge more critically of for whatever reason. A 7 should be a 7 for everyone (that's why judges mostly have voting templates to guide their judgments in sports for example) and I think that balancing the votes would correct the situation.
5- I would help people by giving the chance of voting with a template. That could look like this:
1- Does this picture attracts you artistically (Do you like the subject, the whole image?) 1 2 3 4 5 6 7 8 9 10
2- Do you think the picture is well done technically (good focus, good composition)? 1 2 3 4 5 6 7 8 9 10
3- Does this picture has a WOW factor (Are you immediately impressed by anything in the picture?) Yes Maybe No
From your anwsers, the score you should give is : 7
That's it. What do you think? ;)
Message edited by author 2004-07-12 10:32:51. |
|
|
|
07/12/2004 11:22:37 AM · #53 |
Like many, the update button and I have become very well acquainted. When I enter a challenge I intensely watch the score. Not that I place much value in the score as a measure of the worth of the image or even of its quality of execution. No, the score doesn't reflect that, instead it gives a measure of international mass appeal; and I find that it gives little else but that. For the quality of the shot I watch for comments as they are the primary means of determining the quality of the image. The new additions of views, favorites and wish list to the stats add additional elements to guage the image. The number of views I like the most since if it is significantly higher than the number of votes I feel it shows that I am making the views think and come back for another look.
I'm getting to a point -- honest.
While I enjoy getting all of the comments, the comments I value most (good or bad) are from the experienced photographers; those who have shown they know what they are talking about. But commenting is hard, and as I found out when I commented on every entry of a recent challenge, commenting on a lot of images in each challenge degrades the value of the comments. While I would love to get more comments, with this in mind I don't realisticly see it happening.
I told you I would get to the point.
The karma system has been misinterpreted as a means to 'punish' those that do not conform, which is far from fact. While it is true some images invoke strong reactions in a few individuals that are far from the normal reaction, but those images are not present in enough quantity to significantly affect an individuals karma. A good image, in general, is accepted as a good image regaurdless of content. The addition of a weight to each vote gives the score new character; instead of simply measuring international mass appeal, it becomes a measure of the actual quality of the image -- with experience, ability and knowledge of the voters having a greater impact.
While a karma system changes the character of the vote, it does not eliminate the old character. The vote will still measuring mass appeal but will no longer have that as the major factor.
As far as stats are concerned, what stats will it disrupt? Placing a histogram of both the original and weighted distributions is sufficient to once again not replace, but merely add to what is already available.
David
|
|
|
|
07/12/2004 12:05:11 PM · #54 |
The karma system does not only punish the one who has different tastes, but also the one who had a different way of calculating his appreciation.
I know there are people who looks at a photo giving it a zero at start, and points adds-up with qualities. It's a nice photo, +3, meets the challenge +2, good technique +2, etc.. some others vote starting at 10 and taking points out as they see defaults.. bad focus -4, doesn't meet the challenge -3, etc...
There are people who give mostly under 5 scores because that's the way they vote, should their vote count less?
I think we should balance the voting of everyone instead of giving credit to those who have voted in the same way the mass did.
And a good photographer doesn't make a good judge. And a bad photographer can be a very good judge.
Message edited by author 2004-07-12 12:06:29. |
|
|
|
07/12/2004 12:09:30 PM · #55 |
And thinking again, balancing the votes would even solve part of the troll problem.
Let's say I vote 1 on all pictures. The system would balance my voting to meet the overall average, let's say 6.5. That way, every picture I voted on would have a 6.5 vote from me after balancing.
|
|
|
|
07/12/2004 12:16:29 PM · #56 |
Originally posted by procyon:
And thinking again, balancing the votes would even solve part of the troll problem.
Let's say I vote 1 on all pictures. The system would balance my voting to meet the overall average, let's say 6.5. That way, every picture I voted on would have a 6.5 vote from me after balancing. |
What the...
Perhaps you haven't realized that by doing this, after the balancing, every photo would have the exact same score at the end of the challenge. That would certainly be balanced, though.
David
Of course, then we would all get blue ribbons. :)
/edit: clarity
Message edited by author 2004-07-12 12:17:35.
|
|
|
|
07/12/2004 12:41:27 PM · #57 |
Originally posted by procyon:
The karma system does not only punish the one who has different tastes, but also the one who had a different way of calculating his appreciation.
I know there are people who looks at a photo giving it a zero at start, and points adds-up with qualities. It's a nice photo, +3, meets the challenge +2, good technique +2, etc.. some others vote starting at 10 and taking points out as they see defaults.. bad focus -4, doesn't meet the challenge -3, etc...
There are people who give mostly under 5 scores because that's the way they vote, should their vote count less?
I think we should balance the voting of everyone instead of giving credit to those who have voted in the same way the mass did. |
The challenge winners usually have a vote around the 7.0 mark. With that as the average vote for that photo, a person would have to vote a 3 or less to be hit with bad karma for that photo. But that is just one photo, karma is a trend of all the photos and as the vast majority are around the 5.0 to 6.0 range, a vote of 1 or 10 would be needed to be adversely hit on the average. The chances of the various voting styles falling outside the +/-3 point leeway is pretty slim. And if you look at statistics for very long you realize that over time 'pretty slim' equates to no effect.
Originally posted by procyon:
And a good photographer doesn't make a good judge. And a bad photographer can be a very good judge. |
Perhaps, perhaps not, but the karma system does not take the skill of the photographer into consideration, only his ability to critique over a large number of entries over time.
David
/edit: formatting
Message edited by author 2004-07-12 12:42:00.
|
|
|
|
07/12/2004 01:19:49 PM · #58 |
Originally posted by Britannica:
Originally posted by procyon:
And thinking again, balancing the votes would even solve part of the troll problem.
Let's say I vote 1 on all pictures. The system would balance my voting to meet the overall average, let's say 6.5. That way, every picture I voted on would have a 6.5 vote from me after balancing. |
What the...
Perhaps you haven't realized that by doing this, after the balancing, every photo would have the exact same score at the end of the challenge. That would certainly be balanced, though.
David
Of course, then we would all get blue ribbons. :)
/edit: clarity |
I think you didn't understand the system I'm proposing.
In my example, all my votes would have been raised to 6.5 because all my votes were equal values, all 1's in my example.
Now let's say I vote on 5 photos, I vote 4, 6, 10, 10 and 5. The average in the challenge is 6, mine is 7. A simple system could reajust my vote for it to balance with the average. My votes could count for example: 2, 5, 9, 9, 5 to average at 6 because the system would have analyzed that I votes to high in general. Then another voter has the same taste but a different way of scoring. He puts 1,3,6,7 and 3, averaging at 4. The system could easily raise his votes to 2, 5, 9, 9, 5 because it would have detected his votes to be too low in general.
That would not give every picture the same vote, it would give each vote the right weight in the process.
|
|
|
|
07/12/2004 01:27:49 PM · #59 |
| Anybody notice that the last-place finish in the Color Purple challenge received a 9 from the lone non-camera voter? Ooooohhh.... bad karma! |
|
|
|
07/12/2004 01:36:47 PM · #60 |
|
|
|
07/12/2004 02:24:04 PM · #61 |
Originally posted by Britannica:
The addition of a weight to each vote gives the score new character; instead of simply measuring international mass appeal, it becomes a measure of the actual quality of the image -- with experience, ability and knowledge of the voters having a greater impact.
While a karma system changes the character of the vote, it does not eliminate the old character. The vote will still measuring mass appeal but will no longer have that as the major factor. |
I thouught mass appeal was what we were suppose to be using to determine the winners in our challenges. Why is the opinion of someone who has been voting in the last 100 challenges more valuable than that of a newcomer? How does a karma system measure quality? If you want to be rated on the quality of your photography you would seek out a panel of experts, not the masses of the nether world we call the internet. Is the goal now to change the character of the vote? I thought it was to refine the system to make it harder to abuse? I am trying hard to see what advantage there is to the karma system but, for the life of me, can't find any substantial benefit or improvement. I think the only noticeable change if it were adopted would be to concentrate the votes around a norm that gradually moves upward, and set up a new hierarchy of the big karma members with their suck-ups & wannabees. |
|
|
|
07/12/2004 02:28:40 PM · #62 |
Originally posted by Britannica:
.... the karma system does not take the skill of the photographer into consideration, only his ability to critique over a large number of entries over time. |
I beg to differ with that. I can't find it now but there was a proposal that, in part, would give greater weight to the votes of past ribbon winners. Isn't that the same as taking the skill of the photographer into consideration? |
|
|
|
07/12/2004 02:34:39 PM · #63 |
Originally posted by procyon:
I think you didn't understand the system I'm proposing.
In my example, all my votes would have been raised to 6.5 because all my votes were equal values, all 1's in my example.
Now let's say I vote on 5 photos, I vote 4, 6, 10, 10 and 5. The average in the challenge is 6, mine is 7. A simple system could reajust my vote for it to balance with the average. My votes could count for example: 2, 5, 9, 9, 5 to average at 6 because the system would have analyzed that I votes to high in general. Then another voter has the same taste but a different way of scoring. He puts 1,3,6,7 and 3, averaging at 4. The system could easily raise his votes to 2, 5, 9, 9, 5 because it would have detected his votes to be too low in general.
That would not give every picture the same vote, it would give each vote the right weight in the process. |
Your right, I didn't understand it correctly. But now that I do, I don't see it as doing anything other than taking away the individuals right to vote the score they feel the photo deserves based on their expeirience and ability.
David
/edit: clipped excessing quoting
Message edited by author 2004-07-12 14:36:02.
|
|
|
|
07/12/2004 02:53:02 PM · #64 |
Originally posted by coolhar:
Originally posted by Britannica:
.... the karma system does not take the skill of the photographer into consideration, only his ability to critique over a large number of entries over time. |
I beg to differ with that. I can't find it now but there was a proposal that, in part, would give greater weight to the votes of past ribbon winners. Isn't that the same as taking the skill of the photographer into consideration? |
I remember the post as well:
Originally posted by jcordina:
I would like to suggest a different system which I call an 'effort system'. Here the users that invest effort in the website get rewarded accordingly. Its also very simple to implement. Each user is given a weighted rating and the lower the rating, the less effect he has on the final votes. A user starts off with a low value. Now the more he votes in terms of photos and challenges, the higher his rating gets. In addition, the more he comments whereby his comment is marked as useful by a user, the higher the rating still. Also his rating becomes much larger if the user submits a photos and it gets voted in the top ten. Thus the better photographers become better voters (sort of like judge status). Also I suggest that when voting a user is presented only with a photo by photo vote request (not the thumbnail page) until he or she has reached the required percentage of voting. This would avoid users from just choosing the pictures to vote for (and thus randomising further those pictures one should vote for and creating a better statistical distribution).
In summary I think these would be adequate weights where the rating is the effect one has on the total voting
Rating = 0.1*Votes cast plus 0.3*Comments made and marked as useful
plus 0.6*Pictures in top ten |
and,
Originally posted by jcordina:
Originally posted by melismatica:
This method sounds like Brownie points to me. |
A very valuable voter and good photographer should also be able to produce photos of high quality and get in the top ten of challenges, something a good little worker with no experience might never able to achieve. According to the weights, the good photographer would always get more weighting on final votes. You might say that a good judge is not necessarily a good photographer but we should not distinguish sayers from doers here. Also the comment has to be marked as helpful for it to count towards the final rating. |
But it was not referring to the karma system, but the system jcordina suggested earlier in the thread.
Such additions as extra karma for ribbons won, forum posts, comments given, etc. could be added to the system, but I don't think they would not be of much benefit.
David
/edit: more complete quoting
Message edited by author 2004-07-12 14:56:15.
|
|
|
|
Current Server Time: 03/12/2025 02:29:17 PM  |
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: 03/12/2025 02:29:17 PM EDT.
|
