Yatchen wrote:But when you abstain from the vote it means you a) don't care enought o vote or b) you don't love the image. So it basically is a neutral vote not to vote. I'm personally not about to waste my time voting neutral when my non-voting is neutral enough for me.
Well we're getting into degrees here. A vote of 5 out of 10, or 3 out of 5, whatever is the middle point, is different from not voting at all. If you love an image, one you'd normally rate a 10, or if you just like an image, rating it just 6-7, those two would have equal weight of +1 in an up/down system. It's not a matter of not caring, it's a matter of not viewing the image in an extreme light.
If you look at it in terms of straight averages, you have an image that already has two votes: a 1 and a 10. If you can insert a 5 into there, it insulates the rating a little from more extreme high or low votes, whereas a non-vote in a straight up/down system doesn't have the middle ground protection.
Yatchen wrote:What if you could vote something up from a 1-3 (just an example) range but vote it down by one?
I'm not talking about just a random image; I'm just talking about abusing the rating system to have your favorite images at the top of the "top images" section, which I know happens. And there aren't enough people who even rate images on a regular basis to have an accurate judge anyway.
But even that can be 'abused'. There's no such thing as a perfect system - only a system that hasn't been exploited yet.
And not enough people voting is part of the problem, but that's why this discussion was created in the first place. We fairly routinely get about 20 pages of images a day, and at 15 images a page that's an average of 300 new images that a person could vote on in a given day. Even if you take down the reposts and disableds, I'm sure that it'd still be in the 250+ range on a normal day. I can only go by my own feelings on the matter, but I think a significant part of the low voter turnout is the fact that voting on an image is a very cumbersome process.
Yatchen wrote:When I talk about an image going from 9 to 2, I'm assuming that the image deserves the 9 in the first place.
Which, unfortunately, is an assumption that -cannot- be made. Who's to say that you, who gives an image a 9, is right while someone who votes that same image a 2 is wrong? As I think Merun is alluding to, maybe they legitimately think that the images in the top ratings aren't as good as their favorites. Are they wrong? Is it bad of them to vote down an image they don't like as much? I honestly think no, because art cannot be judged objectively.
Maybe that person would vote the image down without any other factors (e.g. the existence of an image they like better), maybe not, but we can't assume that it's a problem because people are subjective creatures and are prone to introduce their own biases into their thinking. Is it abuse? Or is it just using the system as intended?
I don't think a single person voting an image down is a bad thing. Now if this person were to enlist alt accounts and spam votes an image, that's a different story. But one person-one rating, I think is fair, even if it does cause dramatic swings. However, this becomes much less of an issue if you make the system easier to use and make it so that people want - or at least are willing - to vote for a lot of images even if they don't absolutely love or absolutely hate it. Doing that will give you a more accurate representation of the opinions of the people on Shuu and protect against these alleged 'abuses'.
Merun wrote:Normally, the more rating there is, the more accurate the rating is. The problem is those with extreme rating which is in fact, not the same scale as most user so it's in the end, not accurate.
Well, what is the scale for 'most' users? Do most people for 1,5,10 for images? Do most people only vote for images they really like, giving them ratings of 7-10 (myself included)? There is no set standard, especially when you consider the 10-point scale range, which I think is too wide and an issue, especially since it's too easy to compare it to the standard school grading scale in the states with 7 being pretty average, and anything much lower than that is bad, which I'm pretty sure isn't the intent. I think that would be at least be partially alleviated by narrowing the scale.