Or at least remove the ability to dislike, like Youtube did. The problem is that people don't use it wisely, and seems to form gangs to bash down on a comment. Case in point: I see comments of people complaining about the "no edits!" rule getting likes, but whenever I counter argument (even when presenting a perfectly valid point), they go straight to disliking it.

Put this in your custom CSS.

Nope, that won't solve the problem, since the problem in itself is on the way people use it, and the auto-cutoff (if any) will still trigger for any who didn't change it.

You're asking for site functionality to be removed/disabled on the basis of hurt feelings. This is what we in the business call "unreasonable".
You can also adjust the threshold at which comments are hidden by going into your settings. Just set it to some impossibly low number, use the CSS, and forget that comment ratings even exist.

It's not unreasonable. Unreasonable is that people in the right can get their message hidden from randoms because people in the wrong downvoted it simply because they didn't like it. Unreasonable is that people nag for MONTHS over a rule change that impacts only the minorities.

Wanting to remove a function from a site that actually is used pretty well simply because it has been used against your claims is not a valid reason for its removal.

There's a reason Youtube removed the ability to dislike comments...

It was easily abused.

tsukiyomaruzero said:
There's a reason Youtube removed the ability to dislike comments...

It was easily abused.

Except. They didn't though. I just checked ands it still there clear as day. And saying that is like saying Ccomments altogether should be removed as well because those can and sometime are abused.

united_gamers said:
Except. They didn't though. I just checked ands it still there clear as day. And saying that is like saying Ccomments altogether should be removed as well because those can and sometime are abused.

Well, yeah, the dislike comment button in Youtube still exists, but does nothing.

tsukiyomaruzero said:
Well, yeah, the dislike comment button in Youtube still exists, but does nothing.

Double checked and you are right. In which case though I think its asinine to have a useless function. It should much like this does and total like and dislike number. That being said I still do not believe such a function should be removed Just because 1 out of 5 people misuse it should not detract from the other 4 that use it responsibly. Again my comment about comments still stands.

united_gamers said:
Double checked and you are right. In which case though I think its asinine to have a useless function. It should much like this does and total like and dislike number. That being said I still do not believe such a function should be removed Just because 1 out of 5 people misuse it should not detract from the other 4 that use it responsibly. Again my comment about comments still stands.

1/5? Make it 5/6 and I might believe.

Anyhow, it's on comments, not on the image post itself.

Actually, this reminds me: people also misuse it on images too! They can just like, you know, blacklist guro and snuff, but they instead opt to downvote it and comment "eww, gross!" and stuff.

tsukiyomaruzero said:
Unreasonable is that people in the right can get their message hidden from randoms because people in the wrong downvoted it

sounds like what you actually want is the default score threshold to be reduced, so comments/posts in the negative aren't hidden

bipface said:
sounds like what you actually want is the default score threshold to be reduced, so comments/posts in the negative aren't hidden

... Know what, that could be a good way too.

tsukiyomaruzero said:
Actually, this reminds me: people also misuse it on images too! They can just like, you know, blacklist guro and snuff, but they instead opt to downvote it and comment "eww, gross!" and stuff.

Working as intended.
It's not supposed to be score within its own isolated genre, or even any real marker of artistic quality (let's face it, a high rating doesn't mean a post is objectively good from an artistic standpoint) it's just a general "I like this"/"I don't like this" button.

I see the function used just fine 98% of the time.
If you're seeing downvotes everywhere you comment, perhaps the problem lies closer to home.

magnuseffect said:
Working as intended.
It's not supposed to be score within its own isolated genre, or even any real marker of artistic quality (let's face it, a high rating doesn't mean a post is objectively good from an artistic standpoint) it's just a general "I like this"/"I don't like this" button.

That really calls to question what it's supposed to be used for. If you don't like a particular subject, say cub, shouldn't you really be dealing with that with your blacklist rather than downvoting all the cub art you see? It's not a general like-to-dislike ratio because if you don't like particular subjects or themes, you're encouraged to blacklist it and avoid it altogether instead of downvoting. The up/downvotes don't mean anything as far as site management goes, either (a high score won't stop an image from being deleted or a commenter from getting a neutral/negative record, nor will a low score make an image get deleted, or stop a commenter from being in the right). As it is, I don't understand the intended difference between an upvote and favorite (they both indicate you like it), and the overall score isn't a general like-to-dislike ratio because most people who dislike it either blacklist or ignore it, and it's not supposed to be a score within its own genre because people will downvote based on the subject matter... so what's the point of the votes?

When it comes to comments specifically, you also have the ambiguity of whether someone is (dis)liking what the comment specifically said, the way it was said, the topic in general, or something else. One of my comments got a negative score when I just gave my reasoning for tagging and not tagging certain things; I simply said I couldn't tell if the grovyle was male, and there was nothing to say the grovyle is Grovyle the Thief (no distinct markings or accessories, no mention of it in the source). It got downvoted, and I have no idea why. Did I miss something and was wrong? Did people just not like that I didn't originally tag it male/male or that specific character? Should/could I have by the site guidelines? I have no idea, and the downvotes don't tell me anything, they just give an impression of negativity. I did the same thing with other posts, giving my reasoning for tagging or not tagging certain things, and got a positive score for it, so what is it telling me?

You also have the issue of dog-piling, where people sometimes up/downvote because it already has high number of up/downvotes and they want to be part of the crowd. Or the opposite, some people only vote to counteract the current score but wouldn't vote if it was already going the way they think it should. Some people vote as a "(no) thank you". Some upvotes act as an uncredited "same here" reply. What is the score supposed to mean, if anything?

watsit said:
That really calls to question what it's supposed to be used for. If you don't like a particular subject, say cub, shouldn't you really be dealing with that with your blacklist rather than downvoting all the cub art you see? It's not a general like-to-dislike ratio because if you don't like particular subjects or themes, you're encouraged to blacklist it and avoid it altogether instead of downvoting. The up/downvotes don't mean anything as far as site management goes, either (a high score won't stop an image from being deleted or a commenter from getting a neutral/negative record, nor will a low score make an image get deleted, or stop a commenter from being in the right). As it is, I don't understand the intended difference between an upvote and favorite (they both indicate you like it), and the overall score isn't a general like-to-dislike ratio because most people who dislike it either blacklist or ignore it, and it's not supposed to be a score within its own genre because people will downvote based on the subject matter... so what's the point of the votes?

e621:rules

Refusal to Use Blacklist
Suggested Suspension Length: 3 to 7 days
This category includes:

Creating posts, threads, or comments that complain about artwork that can be blacklisted
Wrongfully claiming lack of knowledge about the blacklist system to avoid punishment from the previous rule

The "Refusal to Use Blacklist" rule says nothing about voting, only about making text posts. To my knowledge you can vote however you want, on whatever you want, provided you're only doing it with a single account.
As for score vs favcount, the latter is pure unadulterated positives, while the former is downvote-weighted. They're just not the same metric, and if anything it's more useful to have both together. order:score_asc is not the same as order:favcount_asc after all. Well, even if that weren't broken, I mean.
But really, what's it to you if an image gets downvoted? What's your stake in it? What compels people to contribute to comment sections with little more than screeching 'It's porn, why are you downvoting it!?' ad nauseum just because a little number at the side of the page is red?

Don't get me wrong, I love comment sections like that, but I think everyone's far too attached to sets of numbers that ultimately, as you're saying, don't mean a whole lot in the grand scheme of things unless you're looking for general popularity-based feedback. Here's one of mine if we're sharing

The youtube dislike button still does something, but it's harder to tell since there is no visible change. Hitting dislike on a comment doesn't bring the score down, it actually cancels the next upvote. So if a person dislikes a comment and then someone upvotes it, the score doesn't change.

thirtyeight said:
The youtube dislike button still does something, but it's harder to tell since there is no visible change. Hitting dislike on a comment doesn't bring the score down, it actually cancels the next upvote. So if a person dislikes a comment and then someone upvotes it, the score doesn't change.

Really?

Always been under the impression it was broken
during the Google+ age and hasn't been fixed since;
Peeps sorta just rolled with it. ╹ ╹)

Wow, OP took his whining about people's dislike of the 2kinds edit rule all the way to the forums. Geez

magnuseffect said:
The "Refusal to Use Blacklist" rule says nothing about voting, only about making text posts. To my knowledge you can vote however you want, on whatever you want, provided you're only doing it with a single account.

Sure, but my point is that people use and are encouraged to use the blacklist to filter out stuff they don't like. If they don't see the content they dislike because of that, they're not likely to be voting on those posts despite disliking them. Some people may not use the blacklist and will still downvote content they don't like, and there's no rule against that as long as they're not otherwise being disruptive, but the votes are still going to be skewed because not everyone who would downvote does. And similarly, not everyone who likes a post is going to upvote it, they may favorite it instead (or they may favorite and upvote, who knows). Plus it's lacking context; a score of 40 can be 40:0, 70:30, 200:160, etc... the same score meaning very different things. Given such a small, uncontrolled sampling, the vote tally is going to be pretty useless at telling you anything good or bad.

magnuseffect said:
As for score vs favcount, the latter is pure unadulterated positives, while the former is downvote-weighted. They're just not the same metric, and if anything it's more useful to have both together.

Weighted according to what, though? For some people downvoting means 'I may like the subject matter, but this isn't well done', and for others it means 'this may be well done, but I don't like the subject matter'. And for others it may mean 'I don't like this style', and yet others its 'I don't like the artist'. When different people have unrelated reasons for giving their -1, the result doesn't tell you anything. It doesn't help either that up/downvotes don't get carried over from deleted inferior posts.

magnuseffect said:
But really, what's it to you if an image gets downvoted? What's your stake in it?

I just find a post's score is at best redundant or meaningless information, or worse, gives a false impression of quality and worthiness, making people overlook what they otherwise might not (and if the site could be made more efficient by posts just having a fav count (or fav and upvote-only), that would be a benefit). Consider a hypothetical post that has 70 favs and an 100 score, and another that has 70 favs and an 80 score. The gives the impression that the former is better than the latter, but it really says nothing worthwhile. Did the former get a dogpile of extra upvotes? Is the latter a repost that lost a bunch of upvotes? Did the latter just manage to catch more people who have a random problem with the image?

And when it comes to comments, it's important to note the site hides all comments with a score less than -10 by default. Doesn't matter why it's that low, or how many upvotes it has either. If there's 11 more downvotes than upvotes for any reason, the comment is invisible to most people. And as my and your comments show, there are people that will downvote innocuous comments without an apparent reason (I'm sure the downvotters have their reasons, we're just left to guess what they are), and it hardly seems worth forcing them closer to being hidden.

Post score is still a quick lazy way to get the good stuff first. It may be flawed, but still works rather well, and way better then non order at all.

And I would argue if 10 people bothered to downvote a comment, what was said was probably not worth saying. Moat people don't vote at all unless something stood out as especially stupid/gross/etc

camkitty said:
Post score is still a quick lazy way to get the good stuff first. It may be flawed, but still works rather well, and way better then non order at all.

Going by favcount, or favs and upvotes-only, would be functionally similar, especially if it's just a quick lazy measurement. Posts that have more likes than other posts still act as a rough guide for what's more worth your time. On the other hand, I find if I'm looking at particularly selected content, going by the amount of likes offers a better guide if it's not offset by people who don't like the selected content. To give an example... some people don't like TF stuff, but I do, so the downvotes by people who don't like TF are worse than useless to me as it hides the upvotes of people who do like it. People who normally like TFs not liking a particular post says much more than TF fans liking a post and a similar number of non-fans disliking it, but the current system gives little distinction between the two results.

camkitty said:
And I would argue if 10 people bothered to downvote a comment, what was said was probably not worth saying. Moat people don't vote at all unless something stood out as especially stupid/gross/etc

Maybe, but it would depend on why the comment was downvoted, which you can't know without looking at the comment. But I'm also not much of a fan of users telling other users what's worth looking at before you ever know it even exists, and gives in too much to a mob mentality. It's one thing if it's against the site rules, but that's not how it works. I mean, there's plenty of creepy comments that get upvoted and are still plainly visible, while some comments get downvoted and hidden for striking a nerve but don't say anything particularly bad or controversial.

watsit said:
Weighted according to what, though? For some people downvoting means 'I may like the subject matter, but this isn't well done', and for others it means 'this may be well done, but I don't like the subject matter'. And for others it may mean 'I don't like this style', and yet others its 'I don't like the artist'. When different people have unrelated reasons for giving their -1, the result doesn't tell you anything. It doesn't help either that up/downvotes don't get carried over from deleted inferior posts.

Weighted according to what, though? For some people upvoting means 'I may not like the subject matter, but this is well done', and for others it means 'this may not be well done, but I like the subject matter'. And for others it may mean 'I like this style', and yet others its 'I like the artist'. When different people have unrelated reasons for giving their +1, the result doesn't tell you anything. It doesn't help either that up/downvotes don't get carried over from deleted inferior posts.

To give an example... some people don't like TF stuff, but I do, so the downvotes by people who don't like TF are worse than useless to me as it hides the upvotes of people who do like it. People who normally like TFs not liking a particular post says much more than TF fans liking a post and a similar number of non-fans disliking it, but the current system gives little distinction between the two results.

Do you operate on the assumption that people are actually mob-downvoting select posts, and not that any downvotes are spread throughout the genre as a whole? I also want to remind you that after a post hits about a day of age, the majority of people who find it are going to be specifically searching for that artist, character/s, or subject matter.

I just find a post's score is at best redundant or meaningless information, or worse, gives a false impression of quality and worthiness, making people overlook what they otherwise might not

Tell me how favourites and upvotes on their own don't do any of this.
Anyone who's judging posts and comments entirely on their score is already a lost cause anyway.

magnuseffect said:
Weighted according to what, though?

False equivocation. A likes-only system (favcount and upvote-only) isn't a weighted scale. The point is to reduce the amount of noise in the resulting value, and to avoid invalid comparisons. Like I mentioned, if you're looking for a specific subject, having downvotes by people who don't like that subject is a useless measure, and further, having a 'this is well done' vote counteracted by a 'I don't like this subject' vote is less than useless since it scores the same as a no-vote, despite meaning different things. And the fact that you can't know if a score may be low because people who like the subject didn't like a particular piece, or because more people than normal downvoted it based on subject matter, adds another level of uncertainty to interpreting the score.

magnuseffect said:
Do you operate on the assumption that people are actually mob-downvoting select posts, and not that any downvotes are spread throughout the genre as a whole?

In general, no. But that's exactly the problem, it's not a general occurrence. Some posts may just catch the eye of more people who don't like the subject for no apparent reason, and some posts don't, creating inconsistency (noise) in the value.

To be clear, I'm not saying my suggestion is a perfect system. There's no such thing. What I'm saying is that I believe it would be slightly better and overall more useful.

magnuseffect said:
Tell me how favourites and upvotes on their own don't do any of this.
Anyone who's judging posts and comments entirely on their score is already a lost cause anyway.

It's not that simple. I don't have time to look at every picture of every subject I like. A post's score is one aspect that determines whether I actually click on a post or not, and with being a factor that determines whether I go to a post or not, cutting out any useless or less-than-useless noise from it can only be beneficial.

A comment's score is another matter entirely. I do think that's completely useless since you generally read a comment before seeing its score. At best, it can help cut down on the banal "same here" or "no u" responses, but it doesn't need to auto-hide comments that hit the magic -10 threshold, especially since there's no control for why a comment gets the score it does. As I said earlier, I don't think giving users control over another user's comment being visible is a good thing.

watsit said:
Some posts may just catch the eye of more people who don't like the subject for no apparent reason, and some posts don't

This is still sounding an awful lot like it's just you assuming that posts rated lower than you'd like have been unfairly targeted "for no apparent reason", and not that your specific personal preferences may not be perfectly in-line with general opinion.
You can't really judge most post ratings alongside one another unless they all feature the same artist or popular-characters anyway, as those tend to overwhelmingly be the biggest factors in vote accumulation. Aside from the animated tag, anyway.

magnuseffect said:
This is still sounding an awful lot like it's just you assuming that posts rated lower than you'd like have been unfairly targeted "for no apparent reason", and not that your specific personal preferences may not be perfectly in-line with general opinion.

No, I'm quite aware that my particular preferences aren't going to match anyone else's, so there's going to be inherent variability in the score vs what I like (hence why I don't use just the score for anything). This is only about making sense of what the score is supposed to mean and how it should be interpreted.

You claim it's a general like-to-dislike tally, but it can't be since people are encouraged to blacklist what they dislike so as to not see it (and consequently not vote on it). The whole point of the tagging system is to make it easier for people to find what they like and avoid what they dislike, which would inevitably skew such a tally. Plus given the varying reasons people would have to upvote and downvote something, combining the up and downvotes is essentially comparing apples to oranges.

If it's not that, the other option would be how well received a post is for what it is. But that currently has the problem that it gets "random" amounts of downvotes due to what it is. In other words, if scoring how good this particular apple is, you're getting some points removed because it's an apple. The number of points removed from any particular apple for being an apple is unknown because it's unknown how many detractors ended up seeing a given apple.

My suggestions aim to alleviate issues with the latter, making the score more useful to search results. If I know I'm likely to like something (because I searched for it), there's no purpose in getting a deducted score from people who don't like it. This doesn't completely fix it, it will still have flaws, but I think it would be more useful than it currently is.

watsit said:

If it's not that, the other option would be how well received a post is for what it is. But that currently has the problem that it gets "random" amounts of downvotes due to what it is. In other words, if scoring how good this particular apple is, you're getting some points removed because it's an apple. The number of points removed from any particular apple for being an apple is unknown because it's unknown how many detractors ended up seeing a given apple.

Let's say we've got someone who does not dislike tag x enough to blacklist it, and may enjoy some posts with that tag, but their opinion on individual posts varies based on specific aspects that aren't always present and the more avid fans aren't bothered by but isn't notable enough to have an extra tag, and maybe it's not something they can often distinguish just from seeing the thumbnail. Maybe they're also seeing it due to a specific artist or character tag they're searching. Is this person not entitled to leave an anonymous, textless, negative opinion on a post they weren't able to be sure if they'd enjoy or not? What if the reason they decided to check the post in the first place was due to a high score? Would you consider them a "random" downvoter because their reason for downvoting may not be apparent to a more dedicated fan of tag x?

magnuseffect said:
Let's say we've got someone who does not dislike tag x enough to blacklist it, and may enjoy some posts with that tag, but their opinion on individual posts varies based on specific aspects that aren't always present and the more avid fans aren't bothered by but isn't notable enough to have an extra tag, and maybe it's not something they can often distinguish just from seeing the thumbnail. Maybe they're also seeing it due to a specific artist or character tag they're searching. Is this person not entitled to leave an anonymous, textless, negative opinion on a post they weren't able to be sure if they'd enjoy or not?

That's what a 'no vote' is for. If you could potentially like something, but don't, you don't upvote it, and as a consequence the score will be lower than it could be since it doesn't have your vote. If enough people don't give their vote who otherwise may, other people who come across the pic will see that not a lot of people liked it compared to similar posts.

However, by downvoting you're effectively taking away someone else's upvote. A low score could mean that not a lot of people liked it, like before. Or it could mean a lot a of people did like it, but there were also a lot of people who didn't like it and took away the upvotes (even if the reason they downvoted has nothing to do with the reasons other people upvoted). It's just adding more uncertainty to the score.

magnuseffect said:
What if the reason they decided to check the post in the first place was due to a high score? Would you consider them a "random" downvoter because their reason for downvoting may not be apparent to a more dedicated fan of tag x?

I would consider that counter-voting. Where the person wouldn't have voted if the score was nominal, but only took away an upvote because they felt it had too many. It is random in the sense that you can't know how many people are doing that or what their criteria for "too many" is, which is adding to the uncertainty.

watsit said:
That's what a 'no vote' is for. If you could potentially like something, but don't, you don't upvote it, and as a consequence the score will be lower than it could be since it doesn't have your vote.

No, a no-vote is for indifference. A downvote is for active dislike.
Another reminder that to vote in the first place one has to have both found the post and decided it's worth their time to open.

magnuseffect said:
No, a no-vote is for indifference. A downvote is for active dislike.

I know, but considering how every downvote counteracts an upvote, the way it currently works fails to account for the difference in reasoning between upvotes and downvotes. Again, the apples to oranges comparison. If enough people don't like something who otherwise would, it would be reflected in having a lower than normal amount of likes, downvotes aren't needed for that. Given the way the site works, having a tagging system designed to help people find what they like and avoid what they dislike, I see no legitimate reason to take away someone else's +1 on a post you dislike instead of simply not giving your +1. It's really starting to sound like you need to have some way to express negativity about posts you find that you dislike, and can't let people that do like it be.

watsit said:
I know, but considering how every downvote counteracts an upvote, the way it currently works fails to account for the difference in reasoning between upvotes and downvotes. Again, the apples to oranges comparison. If enough people don't like something who otherwise would, it would be reflected in having a lower than normal amount of likes, downvotes aren't needed for that.

It also fails to account for the difference in reasoning between one upvote or another.

It's really starting to sound like you need to have some way to express negativity about posts you find that you dislike, and can't let people that do like it be.

You've got me. I revel in the deep psychological torment I inflict upon every user whose upvote I have unjustly stolen.

Do you have any specific examples of posts being downvoted to the point of a score far lower than expected for a post of that subject matter, post date and time, artist/character/species content, and level of objectively-defined quality? When you find such a post, if the score's not in the negatives how do you know it's due to downvoting and not just a coincidental lack of upvotes?

magnuseffect said:
It also fails to account for the difference in reasoning between one upvote or another.

That may be so, but an upvote is still an upvote. Someone thought there was some good reason to see it. It's not then paired with a downvote that may have been given for a reason that has no bearing on the reason it was upvoted. And either way, that issue would exist whether there are downvotes or not. The downvotes don't help with it, they make it worse.

magnuseffect said:
When you find such a post, if the score's not in the negatives how do you know it's due to downvoting and not just a coincidental lack of upvotes?

Exactly. You can't know if it's because people in general didn't upvote it, or if there's a number of people who did upvote it with a similar number of people who downvoted it (5:0 has the same score as 30:25 and 100:95). That's my point, and it's an important distinction that's lost with the current system. What's the purpose of the score then if you can't be sure how much a given post's been lowered by downvotes, and whether those downvotes have any legitimate bearing on the upvotes they remove?

watsit said:
That may be so, but an upvote is still an upvote. Someone thought there was some good reason to see it. It's not then paired with a downvote that may have been given for a reason that has no bearing on the reason it was upvoted. And either way, that issue would exist whether there are downvotes or not. The downvotes don't help with it, they make it worse.

We're getting away from the fact that vote reason is ultimately irrelevant. The voting function does not care why a vote is given, there's no regulation on voting beyond one-per-person, and the occasional case where someone intentionally goes out of their way to blanket-downvote a user/artist/character/whatever.
I get that that the lack of meaning is what you have a problem with, but I really don't think removing the downvote button is a solution.

Another thought. Would you rather people hit a downvote button and go away, or drop a negative comment every time they see a post that they don't like, but can't blacklist?

(5:0 has the same score as 30:25 and 100:95). That's my point, and it's an important distinction that's lost with the current system.

And in your system these posts would have respective scores of 5, 30, and 100, despite the latter being even more disliked proportionally than the middle post. Hell, it would still have a higher score than a 30:0 post. Does that seem right to you? Should a post with 100 upvotes and 500 downvotes be given a higher score than a post with 80 upvotes and no downvotes at all?
I think it would make more sense to either leave things as they are and show the vote ratio, or if there were no downvotes at all, to instead display it as a likes-to-views ratio.

You're not convincing me that upvotes as the only information is anything but a nightmare scenario.
How about if you could pick-and-choose which setting you're viewing or searching by? Though I imagine any of this would probably be hell to implement, I don't know much about the danbooru(?) code.

magnuseffect said:
We're getting away from the fact that vote reason is ultimately irrelevant. The voting function does not care why a vote is given, there's no regulation on voting beyond one-per-person

Which is precisely why comparing upvotes to downvotes is meaningless. There's no control for why an upvote was given compared to a given downvote, but they cancel out all the same.

magnuseffect said:
I get that that the lack of meaning is what you have a problem with, but I really don't think removing the downvote button is a solution.

I admit I'm not very good when it comes to optimal ways for displaying statistics to make them most useful. However, I do strongly feel the current way that's simply score = upvote_count - downvote_count is one of the least useful ways to express it. If I know I like a particular subject, the number of downvotes is going to have far less weight than the number of upvotes, but the score gives them equal weight for any and all reasons. Especially since the site is trying to help you avoid the things you don't like and would otherwise downvote, so another way to consider it is that downvoting is more indicative of the system failing than a reflection of the image itself (the uploader badly tagged something, the more pertinent tags were missed, the ideal tags don't exist and maybe should, or people aren't using their blacklist right).

magnuseffect said:
Another thought. Would you rather people hit a downvote button and go away, or drop a negative comment every time they see a post that they don't like, but can't blacklist?

The latter is against the rules if it's being a disturbance or not constructive:

https://e621.net/wiki_pages/1638#spamming said:
Spamming or Trolling
Suggested Suspension Length: 3 days, or Permanent
This category includes:

• Excessively communicating the same phrase, similar phrases, or pure gibberish
• Creating comments, forum posts, or threads for the sole purpose of causing unrest
• Causing disturbances in forum threads or comments, such as picking fights, making off topic posts that ruin the thread, and/or insult other members
• Making non-constructive or derailing forum posts or comments.
• ...

If you don't like something, you can ignore it without telling everyone.

magnuseffect said:
And in your system these posts would have respective scores of 5, 30, and 100, despite the latter being even more disliked proportionally than the middle post. Hell, it would still have a higher score than a 30:0 post. Does that seem right to you? Should a post with 100 upvotes and 500 downvotes be given a higher score than a post with 80 upvotes and no downvotes at all?

Compared to the alternative of the former having a -400 score despite having 25% more upvotes than the latter that has an 80 score, yes, I would prefer the post with 100 upvotes show it has more likes than the one with 80.

magnuseffect said:
I think it would make more sense to either leave things as they are and show the vote ratio, or if there were no downvotes at all, to instead display it as a likes-to-views ratio.

Showing a likes and views count (as separate absolute values, not a percentage or difference tally) would likely work better than the current system too, however given the way the site works (anonymous access, multiple accounts allowed for one person), even a semi-accurate view counter could be problematic I think.

magnuseffect said:
How about if you could pick-and-choose which setting you're viewing or searching by?

Well, it would still leave issues with dog-piling and counter-voting if most people are seeing both (I'm talking both ways here, I don't want the upvote count inflated because of the number of downvotes either), but sure, it would be a positive step to have a user option to see the upvote count separately.

watsit said:
If I know I like a particular subject, the number of downvotes is going to have far less weight than the number of upvotes, but the score gives them equal weight for any and all reasons.

Though you're arguing that even the best reason for downvoting is inferior to the worst reason for upvoting.

Especially since the site is trying to help you avoid the things you don't like and would otherwise downvote, so another way to consider it is that downvoting is more indicative of the system failing than a reflection of the image itself (the uploader badly tagged something, the more pertinent tags were missed, the ideal tags don't exist and maybe should, or people aren't using their blacklist right).

I notice you're continuing to only talk about voting in the contexts of people bashing on a niche subject matter. There's still no blacklist options for things like bad anatomy or perspective, or a poorly-written comic story. Some artists just get a ton of views from long-standing popularity, and pull in a lot of upvotes from that. I for one don't want to see popularity-based upvotes.

The latter is against the rules if it's being a disturbance or not constructive:
If you don't like something, you can ignore it without telling everyone.

That wasn't a suggestion, it's what you're going to end up with if you nuke downvoting.
I'd like to pull an ex-admin quote from an old thread.

Chaser said:
People can downvotes what they want as long as they don't shit up the comments about how they dislike it.

Regardless, a lot of negative feedback is still allowed. I don't remember seeing mass-bans for the comment section on the last page of Adam Wan's Mystic Tantra. Not to mention both comments and voting were a large factor in JasonAfex's departure from the site. If anything, the people whose comments exist only to complain about the downvotes are the disruptive ones here. I think you're overestimating what's bannable when it's not involving the Refusal to Use Blacklist rule.

watsit said:
Compared to the alternative of the former having a -400 score despite having 25% more upvotes than the latter that has an 80 score, yes, I would prefer the post with 100 upvotes show it has more likes than the one with 80.

In that scenario would you immediately assume every downvote was from one of these hate-downvoters, and none were from fans of the subject matter? Your narrative is mostly that downvotes from people who aren't into the subject matter should be invalid, but you still want to shut down votes from people who are into the subject matter and simply find the post to be terrible. A high-view low-score post can still have more upvotes on it than a low-view unanimously-upvoted post, but the latter can compete due to downvotes. Removing downvotes is a shortsighted solution to a single niche issue, and is still ignoring that within site guidelines there's absolutely nothing wrong with downvoting a post you found yourself on but could have blacklisted.

magnuseffect said:
Though you're arguing that even the best reason for downvoting is inferior to the worst reason for upvoting.

I'm not saying that, I'm saying that downvotes in general have less weight than upvotes when I already know I tend to like what I'm searching for. That doesn't mean there's never any reason to downvote, or that there isn't a reason I would agree with, but I don't think it's worth the added uncertainty in the score by treating each downvote as equal to each upvote.

magnuseffect said:
Regardless, a lot of negative feedback is still allowed.

Never said otherwise. The rule I linked to (anchors aren't working...) even states:

unless the entire thread is joining in and enjoying an offensive or harassing comment, someone who posts insulting comments should get flagged. We will have to use common sense when enforcing this; people can have opinions and state them, but that does not allow them to make someone else feel bad or unwanted. If no one complains or gets offended, AND it’s not an outwardly offensive remark, then it will be okay.

The rule isn't about keeping comments positive. If you don't like something, you're free to say it as long as you're not being disruptive, or hating on people who do like it, or spamming. If the comment section is jumping in on and having a good time with the negativity, have at it. But all the same, if the comment section is being positive, trying to start crap because you don't like it is likely to get a moderator's attention. Same if you post the same banal "this is crap" type message to every post you don't like.

magnuseffect said:
In that scenario would you immediately assume every downvote was from one of these hate-downvoters, and none were from fans of the subject matter?

That's the problem, I can't assume anything about why a downvote was cast. Some reasons I care about more than others, but the system as it currently is treats every downvote as equal to every upvote, which is a bad way to tally a score.

watsit said:
I'm not saying that, I'm saying that downvotes in general have less weight than upvotes when I already know I tend to like what I'm searching for. That doesn't mean there's never any reason to downvote, or that there isn't a reason I would agree with, but I don't think it's worth the added uncertainty in the score by treating each downvote as equal to each upvote.

Yet you'd like to introduce more uncertainty by hiding downvotes. Not having downvotes factor in gives you less information.
If YOU personally already know you're most likely going to like what you're searching for, why do you need score at all? I'm well aware this argument is going nowhere and by this point nobody else is going to bother reading, but I've just got to know.

[edit addition]

watsit said:
It's not that simple. I don't have time to look at every picture of every subject I like. A post's score is one aspect that determines whether I actually click on a post or not, and with being a factor that determines whether I go to a post or not, cutting out any useless or less-than-useless noise from it can only be beneficial.

Going back to this
Assuming this means you would open the same number of posts, what are you gaining from nullifying downvotes? You're not gaining more posts, you're just shifting the bias so that you're viewing images a lower proportion of active voters enjoyed. Given that a big factor in this argument is that most viewers would be people who searched for the content in the first place, how is that helpful to you?

Or do you instead use a search cutoff threshold and you're complaining posts you like don't meet that threshold? Wouldn't you just have to raise your threshold again when the score counts inflate, once more losing out on less-disliked posts that you might have liked?

magnuseffect said:
Yet you'd like to introduce more uncertainty by hiding downvotes. Not having downvotes factor in gives you less information.

You don't introduce uncertainty by removing uncertainty. Removing noisy information is a valid way to improve the signal to noise ratio, so while it does remove some information, the information that remains is clearer.

magnuseffect said:
If YOU personally already know you're most likely going to like what you're searching for, why do you need score at all? I'm well aware this argument is going nowhere and by this point nobody else is going to bother reading, but I've just got to know.

Because as I've said, I don't have time to look at every picture I may possibly like. Not all pictures are equal, and the score is one factor that helps me weed through posts to find pictures more worth my time. But as it is, the score is a relatively weak factor because of its problems, which is why I'd like to see it get better so it can be more helpful.

magnuseffect said:
Assuming this means you would open the same number of posts, what are you gaining from nullifying downvotes? You're not gaining more posts, you're just shifting the bias so that you're viewing images a lower proportion of active voters enjoyed. Given that a big factor in this argument is that most viewers would be people who searched for the content in the first place, how is that helpful to you?

I'm gaining more certainty in what the score represents. As it is now with every downvote being equal to every upvote and the score being the simple difference between up and downvotes, I have no idea why a score may be the way it is. Every issue that exists within an upvote-only system will still be there with downvotes, but with the added complications. Don't know why someone upvoted? Having downvotes doesn't clarify that at all, but you also now don't know why someone downvoted either, or whether a downvote has any relevance to an upvote. Like a piece that has low upvotes and dislike a piece that has high upvotes? Still happens if you have downvotes, but now also you can like a piece that has high downvotes and dislike a piece that has low downvotes.

By removing the uncertainty associated with downvotes, the idea is the score will be more stable in the aggregate, so when a given post's score is lower or higher than normal, it means something about how people who like or don't mind the subject took to a particular piece, which is obviously helpful if I'm deciding to look at a piece or not. If fans of a subject all gave their +1, I don't care as much about people giving a -1*, but if fans were reluctant to give their +1, I don't have as much need of seeing people's -1. No this won't be perfect, yes there will still be a fair amount of variability and differences of opinion, but even in the best case, having downvotes with a score that treats them as equal to upvotes does not help. With the scoring system as it is, it has to go farther outside of the average a kind of post usually gets to mean something.

• Not to say there can't be legitimate grievances with a piece, or even some I may agree with, but a simple -1 conveys nothing useful about what those problems may be.

watsit said:
You don't introduce uncertainty by removing uncertainty.

Not how it works. You're talking about the uncertainty within the reasons for downvoting, ignoring the certainty in that someone disliked it enough to downvote in the first place.

I'm gaining more certainty in what the score represents.

The intended use of the voting system is this many more people upvoted than downvoted. You're chasing phantoms when you get caught up in whether a "low post score" is due to downvotes or not, and even if it is due to downvotes and not just a lack of upvotes, it's more likely that any significant volume of downvotes comes from the people who were looking for the content in the first place, unless the thumbnail is just that provocative.

Every issue that exists within an upvote-only system will still be there with downvotes

Tried to point out in my last post all you're arguing for is for worse-rated posts to float up. Everything you're saying is indicating that you personally just don't like where some posts are at score-wise, and you're pinning it on downvotes specifically from people who don't like the tagged content, simultaneously admitting that most of those won't see it because they're not looking for it. I think you'll find under your proposed system score is even less useful than it is now, though it wouldn't surprise me if you wouldn't even admit it then.

If fans of a subject all gave their +1, I don't care as much about people giving a -1*, but if fans were reluctant to give their +1, I don't have as much need of seeing people's -1.

I'm stepping into dangerous territory poking at wording semantics, but I notice the +1 votes in this scenario are from fans, but the -1 are from people. Given we're talking about a tag-subject and not individual images, why this implication that fans don't downvote? Are TF fans a hivemind who only think the way you do, and wouldn't dare touch that down-arrow on a post they think is bad? Are they not a real fan if they ever downvote a post?

the idea is the score will be more stable in the aggregate, so when a given post's score is lower or higher than normal, it means something about how people who like or don't mind the subject took to a particular piece

Something that also does this: downvotes.
Removing all the downvotes from people who like or don't mind the subject matter is taking away a whole lot of information.
I think you'll find people who like similar things have far more varied opinions than you give them credit for.

I'm running out of ways to say the ideology is getting in the way of the logic.

magnuseffect said:
Not how it works. You're talking about the uncertainty within the reasons for downvoting, ignoring the certainty in that someone disliked it enough to downvote in the first place.

I'm not ignoring it, but I do think it's less important than having a more meaningful score. The downvotes don't exist in a vacuum, they're directly correlated to the upvotes, and relating them that way creates uncertainty with the result since there's no assurance the upvote reasons have anything to do with the downvote reasons.

magnuseffect said:
The intended use of the voting system is this many more people upvoted than downvoted.

Which is a meaningless statistic, since the given upvotes may have no bearing on the given downvotes. There's no way to know why someone upvoted or downvoted, so seeing how many more people upvoted than downvoted tells you next to nothing useful.

magnuseffect said:
You're chasing phantoms when you get caught up in whether a "low post score" is due to downvotes or not

I don't think I am because, as I said, it's an important distinction. Think of it this way. Say a game is being made, and it's targeting a niche audience. It has a gameplay loop that most people don't particularly care for. You, however, are a fan of that kind of gameplay and would probably like it. You also know other fans of that kind of game are likely to like it, while the general player-base may like it, not care, or dislike it. The game comes out, you look for information about whether or not to buy it, and you see it has a metascore of 5 (tallying +1s that mean 'buy it', -1s that mean 'don't buy it', and 0s that mean 'eh, whatever'). Do you not think it would be important to distinguish between +10:-5 (not that many people cared for it, even the fans who otherwise would) and +100:-95 (quite a few people recommend it, likely the fans who expectedly would, and quite a few say to avoid it, likely the more general player-base who expectedly wouldn't like it)? In the former case, it's telling you that even if you normally like this kind of thing, nobody really liked this particular game that much so it may be something to skip, while in the latter case, it still found its audience and is probably worth checking out.

magnuseffect said:
Tried to point out in my last post all you're arguing for is for worse-rated posts to float up. Everything you're saying is indicating that you personally just don't like where some posts are at score-wise, and you're pinning it on downvotes specifically from people who don't like the tagged content, simultaneously admitting that most of those won't see it because they're not looking for it.

Either you're really not understanding what I'm saying, or you're being purposely disingenuous. I could just as well say that all you're arguing for is for niche subjects to be sunk in the rankings. Everything you're saying is indicating that you personally just don't like that some low-scored posts may be more popular than they currently appear, and need downvotes to keep their scores down where they belong.

Now, if we're done with the strawmen, it would be nice if you countered my arguments. Notably, you have not explained how I'm wrong that treating every upvote as equal to every downvote is an apples-to-oranges comparison (just because there are legitimate reasons to downvote doesn't mean it's all equal to upvotes in the aggregate). Or how I'm wrong that the variability with the reasons for downvoting is adding more noise/uncertainty to the resulting score (just because it adds some information doesn't mean it's not also bringing in more noise). Or how I'm wrong that the issues with upvoting -- not knowing why an upvote was cast, that you can still not like a piece that has many upvotes, etc -- don't get fixed by having downvotes, let alone make it worse. Or how I'm wrong that the distinction between 5:0 and 95:90 is important.

magnuseffect said:
I'm stepping into dangerous territory poking at wording semantics, but I notice the +1 votes in this scenario are from fans, but the -1 are from people. Given we're talking about a tag-subject and not individual images, why this implication that fans don't downvote?

Sloppy wording on my part. In general, fans of a particular subject are more likely to upvote pictures that subject before other people, and more likely to downvote after other people, but you're right, that's not always true and it varies on an individual level.

magnuseffect said:
Something that also does this: downvotes.
Removing all the downvotes from people who like or don't mind the subject matter is taking away a whole lot of information.

Which would be all fine and dandy if we were just talking about downvotes. But we're not. Those downvotes aren't their own separate thing that you only deal with if you want, they're being directly compared to upvotes and affecting the score without any checks that it's a valid comparison.

Personally, I'd like to see two separate counters for upvotes and downvotes. Seeing them separately would give a better idea of how many people liked or disliked a picture, rather than looking like divisive pictures are just getting ignored. Just my two cents.

iago1 said:
Personally, I'd like to see two separate counters for upvotes and downvotes. Seeing them separately would give a better idea of how many people liked or disliked a picture, rather than looking like divisive pictures are just getting ignored. Just my two cents.

I agree but I think the exact values should be somewhat obscured to discourage voting wars.
Say you partition the space of 'number of total votes', and create a scale (which could be expressed as color or font size):

a. Totalvotes <20
b. Totalvotes <100
c. Totalvotes <500
d. Totalvotes <2500
e. Totalvotes >=2500

This could indicate 'how comprehensive' a given vote ratio was.
Then you could normalize the actual vote ratio to 10:x / x:10, and apply the style determined by the above table.
For example, Watsit's example '95:90' could be expressed as 10:9, rendered in style C, since Totalvotes = 185 for this case.

However, when ordering searches by upvotes / downvotes, I think people would still figure out a way to guess the raw values with some accuracy.

watsit said:
The downvotes don't exist in a vacuum, they're directly correlated to the upvotes, and relating them that way creates uncertainty with the result since there's no assurance the upvote reasons have anything to do with the downvote reasons.

meaningless statistic

Irrelevant. Upvote reasons don't matter, downvote reasons don't matter. It's about the raw numbers, and keeping the system as simple as it can go.

Game score stuff

But of a weird analogy given that the game industry tends to use either averages (eg. Metacritic) or ratio+count (eg. Steam)
Yes, it would be useful to distinguish between those, and I've already been pointing out ratio+count as a better alternative.
But getting back to your example of having only the raw tally difference available, I've been agreeing that a score close to zero needs more context. But even if you just add the vote count you get additional context for whether it's low because nobody's seen it, or low because a bunch of people didn't like it. If anything I'd see lower value on the one a lot of people actively disliked than the one people either didn't notice or were indifferent to.

+10:-5 (not that many people cared for it, even the fans who otherwise would) and +100:-95 (quite a few people recommend it, likely the fans who expectedly would, and quite a few say to avoid it, likely the more general player-base who expectedly wouldn't like it)

If votes are visible, 10-5 tells me few people paid attention to the game in the first place but two-thirds of voters found it enjoyable, while 100-95 tells me a larger number of people played the game, and it disappointed nearly half of the voters.
Now, we're jumping from one form of media to another, but there's a general positive trend when talking about optional +/- voting across most online entertainment media, to the point that normally anything below around 80% positive is a big red flag.

Either you're really not understanding what I'm saying, or you're being purposely disingenuous.

I understand what you're saying, I'm just not sure you realise the implications.

I could just as well say that all you're arguing for is for niche subjects to be sunk in the rankings. Everything you're saying is indicating that you personally just don't like that some low-scored posts may be more popular than they currently appear, and need downvotes to keep their scores down where they belong.

Please quote anywhere I've said that users who want to upvote a post should not be able to do so, or that low scores should be allocated any way other than downvotes from people who didn't like that post, or a simple lack of post interest. I'm not advocating for a system with only downvotes.
But you appear to be advocating for the removal of (or, to be reasonable, option to hide) all downvotes based on not wanting to see (down*)votes specifically from people who do not intrinsically enjoy the subject matter. Then when I bring up downvotes from users who do enjoy the subject matter you tell me those are also irrelevant because they're counted alongside the ones you object to, despite being presumably a higher proportion than other downvotes. You've also been talking about a downvote "taking away someone else's" upvote, and insinuated that I might downvote as a form of harassment.
^{*I imagine you wouldn't mind if someone who didn't like the subject matter came in and upvoted.}

it would be nice if you countered my arguments.

You say after sidestepping a lot of mine yourself.

Notably, you have not explained how I'm wrong that treating every upvote as equal to every downvote is an apples-to-oranges comparison (just because there are legitimate reasons to downvote doesn't mean it's all equal to upvotes in the aggregate). Or how I'm wrong that the variability with the reasons for downvoting is adding more noise/uncertainty to the resulting score (just because it adds some information doesn't mean it's not also bringing in more noise).

Every time you've dropped apples to oranges I've responded, I'll go again.
Yes, one downvote reason is not inherently equal to another downvote reason, or an upvote reason. But the voting system doesn't pretend there's supposed to be any regulation on that. /help/voting doesn't even suggest what voting is for, that's all in the mind of whomever looks at the little number with the arrows next to it. It just happens to be that, as is, the majority of people roughly agree on what voting does here. Noise is irrelevant, whatever number is next to a post is the number that site users have collectively determined through voting should be there, and whatever meaning you derive from that specific number is your own business. Post scores aren't inherently relatable to one another. A perfect duplicate post might have a completely different score depending on as little as the date each was posted. There are no oranges here, only apples.

Or how I'm wrong that the issues with upvoting -- not knowing why an upvote was cast, that you can still not like a piece that has many upvotes, etc -- don't get fixed by having downvotes, let alone make it worse.

I'm gonna have to tie this in with the next one, aren't I?

Or how I'm wrong that the distinction between 5:0 and 95:90 is important.

I have continued to agree that a distinction between 5:0 and 95:90 is extremely important, but apparently for different reasons than you feel.
I've already explained how %positive + count is a better option, but I feel like you're dragging this argument back down to talking about a near-zero vote, where it's far less clear whether it's just due to a low total view/vote count. How about we go for 100:0 and 200:100. 100 is a more substantial number of votes, and these still have the same score. But 100% of people who felt strongly enough to vote on the former enjoyed it, (presumably. Not letting you run off with my above argument on that one,) while only 67% (I'm rounding for simplicity's sake) enjoyed the latter but 33% express dislike. I'm pretty sure you're immovable on this, but I'm gonna take it to an absurd extreme anyway. let's add a post with 10000:9900. Same score, almost evenly negative. If you just remove the downvotes entirely you're not necessarily clearing noise/uncertainty from the score, you're just removing almost half the opinions.

Pretty sure both our positions on this are set in stone, I just like to see where things I disagree with end up.

magnuseffect said:
Irrelevant. Upvote reasons don't matter, downvote reasons don't matter. It's about the raw numbers, and keeping the system as simple as it can go.

And the current system doesn't get you that since you don't see the raw numbers, just the resulting tally of +1s and -1s with no context.

magnuseffect said:
Now, we're jumping from one form of media to another, but there's a general positive trend when talking about optional +/- voting across most online entertainment media, to the point that normally anything below around 80% positive is a big red flag.

I would disagree that anything below 80% positive is a big red flag. It depends on the kind of score a type of game usually gets, and in the case of percentages, the number of people that actually voted/scored. More niche subjects are, by their nature, going to get less overall views and less positivity than things with more general appeal. So if you already know you like a given subject, it should to be measured against what it normally gets. If the same like-minded people in general like a particular post of a given subject as much as the last, having more people seeing it and lowering the score because it's not their thing shouldn't be much of an influence.

magnuseffect said:
Please quote anywhere I've said that users who want to upvote a post should not be able to do so, or that low scores should be allocated any way other than downvotes from people who didn't like that post, or a simple lack of post interest. I'm not advocating for a system with only downvotes.

I never said you wanted a downvote-only system. I was using hyperbole to suggest that you just want subjects you don't like to score low by including downvotes, in the same way you suggested I just want my preferred subjects to score high by excluding downvotes. It's called a strawman argument, taking a false interpretation of the other person's argument (that we want to see scores higher or lower) and attacking that (it's a personal crusade, not based on logical reasoning).

magnuseffect said:
But you appear to be advocating for the removal of (or, to be reasonable, option to hide) all downvotes based on not wanting to see (down*)votes specifically from people who do not intrinsically enjoy the subject matter.

Less "not wanting to see" and more "has no direct correlation to and is less useful than upvotes". Again, I've said there are legitimate reasons to downvote. But as you yourself even say:

"the voting system doesn't pretend there's supposed to be any regulation on [voting]"

So if there's no regulation on voting, what basis is there for comparing (by means of a tally or percentage) upvotes and downvotes? What good does it do to see downvotes without context on the same footing as upvotes given no regulation on them? That's exactly what an apples-to-oranges comparison is.

magnuseffect said:
Then when I bring up downvotes from users who do enjoy the subject matter you tell me those are also irrelevant

Not irrelevant, but there's no way to separate them out. There's no way to know which downvotes are over relevant criticisms. And because all downvotes are lumped together, whether they be relevant, semi-relevant, or irrelevant, it's only reasonable to give them less overall weight in light of upvotes. If people have criticisms about a piece, there is a comment section, so it's not like they'd be silenced on the matter.

magnuseffect said:
Noise is irrelevant, whatever number is next to a post is the number that site users have collectively determined through voting should be there

Except "site users" don't all see the various posts, a post's score isn't a site-wide tally of all users' opinions. Again, the tagging system and blacklist is designed to keep people away from subjects they don't like while helping people find things they do like. It's not perfect at that, but that's its goal, and as such, scoring is inherently skewed on that bias. "Noise" (specifically, reducing it) is very relevant if you want the score to mean anything, and it's obviously intended to mean something, therefore reducing noise in the score should be a goal.

magnuseffect said:
How about we go for 100:0 and 200:100. 100 is a more substantial number of votes, and these still have the same score. But 100% of people who felt strongly enough to vote on the former enjoyed it, (presumably. Not letting you run off with my above argument on that one,) while only 67% (I'm rounding for simplicity's sake) enjoyed the latter but 33% express dislike. I'm pretty sure you're immovable on this, but I'm gonna take it to an absurd extreme anyway. let's add a post with 10000:9900. Same score, almost evenly negative. If you just remove the downvotes entirely you're not necessarily clearing noise/uncertainty from the score, you're just removing almost half the opinions.

If some kind of post normally gets around 100 upvotes, and one suddenly has 10000, that's a clear indication of something very abnormal going on with that post. Seeing the 9900 downvotes in addition to that doesn't really help clarify what the deal is. But even seeing a jump to 200 upvotes would mean it likely stepped outside of the usual realm for that kind of post, but was still well received for/despite the subject matter being looking for. The fact that 33% of voters disliked it doesn't change the fact that more than the usual number of people liked it. If any of the additional 100 downvotes have any relevance to what people enjoy about the subject, there's no way to know from that number.

Not gonna spend as much time on this one

watsit said:
And the current system doesn't get you that since you don't see the raw numbers, just the resulting tally of +1s and -1s with no context.

I'm still not arguing the current system is optimal

If the same like-minded people in general like a particular post of a given subject as much as the last, having more people seeing it and lowering the score because it's not their thing shouldn't be much of an influence.

This one's clearly wrapping back to e621 cases, and I keep asking why you think more people are seeing specific posts and skewing the vote, and I'm getting nothing back. I can think of a few reasons but nothing really widespread enough for the level of reaction I'm seeing.

It's called a strawman argument

I know what a strawman is, and I'm sorry you thought I was using one. A funny thing with individual interpretation of logic is it's not grounded in objective fact, and continuous arguments are often just a clash between different individuals' logical interpretation. I tend to get carried away and say things in ways I shouldn't, but what I truly believe is happening in this thread is that our individual logics are incompatible to the point that this is my only real outlet to say that I believe your logic is wrong. There's no way to resolve this without bringing in neutral third-parties to decide which of us is more/less wrong on which points.

So if there's no regulation on voting, what basis is there for comparing (by means of a tally or percentage) upvotes and downvotes? What good does it do to see downvotes without context on the same footing as upvotes given no regulation on them? That's exactly what an apples-to-oranges comparison is.

Community voting activity. We're collectively giving meaning to what essentially is just an abstract, and you're arguing against the de facto usage given by the general userbase.
Don't get me wrong, you're entitled to your opinion that vote reason should have some kind of score influence, but good luck getting a clear majority to agree that voting isn't working the way it should.

You're saying that because vote reasons are different, the votes might as well be different classes of object.
I'm saying that because the votes are already clearly defined as comparable to one another within the system, vote reason has no impact on what they are.
Again you're entitled to an opinion that the votes shouldn't be as clearly defined as they are, but that doesn't override that your apples and oranges argument does not apply to current voting implementation.
There's no progression for either of us here until someone's opinion changes.

Not irrelevant, but there's no way to separate them out. There's no way to know which downvotes are over relevant criticisms. And because all downvotes are lumped together, whether they be relevant, semi-relevant, or irrelevant, it's only reasonable to give them less overall weight in light of upvotes. If people have criticisms about a piece, there is a comment section, so it's not like they'd be silenced on the matter.

As established earlier, the comment section allows for a smaller range of opinion than voting. You could even take that in the other direction and point out that a lot of what gets people banned via the creepy comments rule are reasons for upvoting.

It's not perfect at that, but that's its goal, and as such, scoring is inherently skewed on that bias.

You appear to be implying that the blacklist and tagging system is failing to screen out people who aren't looking for the subject matter, but is there concrete evidence that this is making a significant impact on score variance within a specific commonly-blacklisted niche tag? I need more data than "because I think it is."

Except "site users" don't all see the various posts, a post's score isn't a site-wide tally of all users' opinions.

But it is a tally of the opinions of people who decided to vote on the post, and that's all the system wants.
Score is already tied more to viewcounts than whether one post is actually comparable to another. That's why I keep offering a ratio system, as removing downvotes while otherwise keeping voting as-is does nothing but inflate scores by removing any negative input, but a ratio instead shows you what's behind the viewcount bias.

That one's where my "strawman" came from.

If some kind of post normally gets around 100 upvotes, and one suddenly has 10000, that's a clear indication of something very abnormal going on with that post. Seeing the 9900 downvotes in addition to that doesn't really help clarify what the deal is. But even seeing a jump to 200 upvotes would mean it likely stepped outside of the usual realm for that kind of post, but was still well received for/despite the subject matter being looking for. The fact that 33% of voters disliked it doesn't change the fact that more than the usual number of people liked it. If any of the additional 100 downvotes have any relevance to what people enjoy about the subject, there's no way to know from that number.

But all you're really seeing is that more people saw the post. Maybe the 100-vote unanimously-positive post is just missing a little something that could have made people click, or the specific tag combinations land it in less searches, or it went up right before someone's mass-upload and got flooded out of the first page of searches. In general online behaviour, people are typically a lot quicker to upvote than downvote something. Maybe 80% positive isn't severe enough to concern you, but what do you think about something when it hits 30% positive?

I want to be clear I'm not putting this here as an ad hominem, but is it just something you've never understood because you tend not to have a negative opinion on anything? I'm genuinely curious whether the reason our logics cannot meet is that you've never seen a collectively-voted negative opinion that you've agreed with?

lafcadio said:
You're asking for site functionality to be removed/disabled on the basis of hurt feelings. This is what we in the business call "unreasonable".
You can also adjust the threshold at which comments are hidden by going into your settings. Just set it to some impossibly low number, use the CSS, and forget that comment ratings even exist.

Honestly if the site didn't care about hurt feelings, a lot of the rules wouldn't exist.

greatwhitetrillium said:
Honestly if the site didn't care about hurt feelings, a lot of the rules wouldn't exist.

Sometimes, I read a message, and immediately know that its author recently received a strike for leaving a creepy comment.
This was one of those times. I was not wrong.

bitwolfy said:
Sometimes, I read a message, and immediately know that its author recently received a strike for leaving a creepy comment.
This was one of those times. I was not wrong.

To be fair, that's a fairly instant assumption any time somebody necros a threat from a year ago. Makes me wish for an auto-lock.

greatwhitetrillium said:
Honestly if the site didn't care about hurt feelings, a lot of the rules wouldn't exist.

Removal of site functionality is not comparable to punishing individual users.