Posted under General
Sacriven said:
Not mine. This is the 4th time the guy flags with that reason. It seems he completely hates chanta's art style. I mean, "Leisure Suit Larry faces" as flag reason? We are beauty experts now?
It looks like a vendetta against the artist or art style. At this point, the mods are probably considering a warning if one hasn't already been given - it's definitely vandalism IMO.
Just a heads up, but I've changed help:flag notice to be a little more in line with howto:upload and howto:flag, given that the "extremely large breasts/penises" clause is really subjective and shouldn't be something for users to decide solely whether to flag on e.g. post #2605869. If we left that in, I feel as if it would provide incentive to flag everything tagged with gigantic breasts and/or huge penis, and that's not necessarily the correct thing to do.
Apologies if I acted a little too early on this, but I felt like the sooner this would be enacted, the better. Feel free to make changes as you think suitable.
Someone check out who flagged post #2605993 and warn them please.
This isn't the correct use of flagging. We have other 'seizure' inducing images, but all this ugoira is is a fun way to try your luck.
Mikaeri said:
given that the "extremely large breasts/penises" clause is really subjective
No, there's a metric for it based on head size. The wiki pages for those tags clearly state the threshold and that picture goes way over it.
That flag is too vague and could be more descriptive, but it's still valid.
Hoobajoob said:
No, there's a metric for it based on head size. The wiki pages for those tags clearly state the threshold and that picture goes way over it.
That flag is too vague and could be more descriptive, but it's still valid.
Then it should be judged by that metric.
That flag isn't vague, it's plain wrong. The flag itself might be valid if for that reason, but proof in the pudding shows that it isn't. Given an approver will probably treat it as such though, such a reason might as well be implied.
I was contacted by someone else about that post, so that's why I mention it.
In regards to post #2605869 and post #2605993, I've sent warnings to both flaggers over invalid flagging.
The user in post #2605869 had access to a valid reason to potentially flag the post, but used a flag that did not describe the actual potential problem with the image and instead described a subjective dislike of the content, thus is in no uncertain terms an invalid flagging reason.
The user in post #2605993 flagged it for a reason that is currently not valid in flagging the post, I've recommended to the user that they bring the issue to the forum to be discussed on whether or not policy on such posts need to be created.
@Flopsy
Since I consider it as unfair when I don't answer you, I respond now.
So there are actually two things which is how I see flags and how I see mutual(!) quality control.
So, I think flags are the complete opposite than approvals. From the way how they are looked at and how they should be created. Approvals are subjective, you don't need to reason why you approved a post, post can be approved because of style and that in a row and it usually takes a long time that a Janitor receives punishment because of their approvals and they are positive quality control
Flags instead are a ground for discussing (see this topic), they need to be objective and you can't flag too many posts in a row, because unsuccesful flagging will be fast punished as vandalism (see the Leisure Larry Suit flags: Permanent flags against, here an artists style, are fast discovered and punished. Then flags need to be descriptive and there is a list full of requirements (see the flag notice).
So, that's that. I don't know how relevant this message is to you, but it should say that flagging can be pretty difficult and is not easy and needs to be developed by the flagging users over time. My flags are usually pretty long besides the image sample flagging.
As for quality control per se: Flagging is useful to control users or approvers. If there weren't such a system we prpbably don't even need the Contributor rank. Contributors are users that are trusted enough for their uploads. Flagging on the other hand is prohibiting that an user is constantly uploading crap and therefore abusing their position. So one could say that flagging is a negative control and if you want to give the flagging system a body, flags probably need to be done on a daily base. If one post is flagged then it should signal to the uploader that the post probably has some issues while it is flagged active. If the flag is succesful then it is signaling to the uploader: "Stop, don't upload such images".
If the flags and approvals are executed with responsibility, then this is a working system, since the goal is, like I mentioned, that certain images with certain issues shouldn't be uploaded.
As you can guess it'll take a long time that a Contributor/Janitor receives punishment for their uploads if they continue. But then on the other side there were flags that told them "Stop", but if the uploader doesn't learn, then it is their fault (and should be given a neutral or negative feedback). That is still no personal attack or insult or whatever, but it ensures that there isn't too much crap uploaded to Danbooru.
If they still continue then it is probably time for demotion...
post #2613854
I know I've flagged this post, but this image was flagged twice. And the other reason, the one with the western artist, looks a bit off to me.
Provence said:
Since I consider it as unfair when I don't answer you, I respond now.
So there are actually two things which is how I see flags and how I see mutual(!) quality control.
So, I think flags are the complete opposite than approvals. From the way how they are looked at and how they should be created. Approvals are subjective, you don't need to reason why you approved a post, post can be approved because of style and that in a row and it usually takes a long time that a Janitor receives punishment because of their approvals and they are positive quality control
Flags instead are a ground for discussing (see this topic), they need to be objective and you can't flag too many posts in a row, because unsuccesful flagging will be fast punished as vandalism (see the Leisure Larry Suit flags: Permanent flags against, here an artists style, are fast discovered and punished. Then flags need to be descriptive and there is a list full of requirements (see the flag notice).So, that's that. I don't know how relevant this message is to you, but it should say that flagging can be pretty difficult and is not easy and needs to be developed by the flagging users over time. My flags are usually pretty long besides the image sample flagging.
I know that flagging isn't a complete free-for-all. That would've been worse, but the way it works now still doesn't do the site any good.
As for quality control per se: Flagging is useful to control users or approvers. If there weren't such a system we prpbably don't even need the Contributor rank. Contributors are users that are trusted enough for their uploads. Flagging on the other hand is prohibiting that an user is constantly uploading crap and therefore abusing their position. So one could say that flagging is a negative control and if you want to give the flagging system a body, flags probably need to be done on a daily base. If one post is flagged then it should signal to the uploader that the post probably has some issues while it is flagged active. If the flag is succesful then it is signaling to the uploader: "Stop, don't upload such images".
If the flags and approvals are executed with responsibility, then this is a working system, since the goal is, like I mentioned, that certain images with certain issues shouldn't be uploaded.
As you can guess it'll take a long time that a Contributor/Janitor receives punishment for their uploads if they continue. But then on the other side there were flags that told them "Stop", but if the uploader doesn't learn, then it is their fault (and should be given a neutral or negative feedback). That is still no personal attack or insult or whatever, but it ensures that there isn't too much crap uploaded to Danbooru.
If they still continue then it is probably time for demotion...
At least you're being honest about your intentions. That lets us get to the heart of our disagreement. The way I see it, the kind of flagging you're describing is destructive and pointless. It's true that the effort that went into pictures, the skill of the artists and the relevance of the content can be ranked semi-objectively (e.g. "post #2447088 is more skillfully drawn than post #2615728, which is more skillfully drawn than post #535171."). On the other hand, the question of whether a given picture is good enough and relevant enough for Danbooru is NOT objectively answerable.
In fact, our disagreement can be viewed as being about which mechanism Danbooru should use to determine what is "good enough". You, and some other people in this thread, favor the idea that uploaders and approvers should limit their uploads/approvals to posts that they are fairly sure that several other approvers would re-approve, since such posts are flag-resistant and those who flag them, by your reasoning, in the wrong. Conversely, if approvers often find themselves on the losing side of the tug-of-war with the flaggers, then they should take this to mean that their approval criteria are in need of adjustment.
That's how I understand what you're saying, and I think you're wrong. I don't think approvals need to or should operate by consensus, not even partial consensus. Posts that some, or most, users/approvers consider "crap", perhaps because the damn leg is too short or whatever, are appreciated by other users, and those who don't like them can just ignore them or filter them out. What is needed is a minimal quality barrier (i.e. an initial thumbs-up from a contributor or approver) to ensure that there is no massive influx of low-quality and off-topic posts that render filtering ineffective. There is no such influx right now, and there hasn't been one since I came here.
Long story short: Flaggers should stop flagging, because what they are doing is to exclude posts (i.e. posts that were deleted after a flag, deleted because no-one dared to approve them or never uploaded because people thought it was no use) that add value for many users and aren't substantially reducing anyone's ability to use the site. That is destructive and pointless, that is what I've been saying since I first posted in this thread and that is why I don't think we're getting anywhere here. In the end, it's up to Albert and the mod staff to run the site as they see fit.
That what you describe is the situation that deleted posts are not able to get re-approved.
Your reasoning is also opening a big hole, because if you say that certain posts are liked by some users, then that automatically means that we don't even need the Contributor and Janitors in the first place. One can argue about the point of those, but if you turn down flags for this very reason then that also effects approvals. Because non-approved are also deleted but there might users who liked that certain post.
Your last sentence is a bit...disappointing. I mean if you say it like that, then this whole discussions makes no sense :P.
I guess you are also a bit misleaded. When I say that flags should be objectively, then I mean that you should put your opinion as neutral as possible. It is like writing a mini review for a post. That is pretty well comparable with a review for a video game: The content is both time written objectively, but you still have an opinion. Is it in the end objectively? Off course not, but when I say that the eyes are lopsided then I give the Janitors more reason to not approve the post as when I write that I don't like the eyes.
And what is the goal of flagging? I guess I can quote albert here who told me once in a longer DM chain that the flagging system is very open on purpose. They've also told me if a post doesn't find approval, then it probably isn't worthy of Danbooru's standards. Now one can argue what "Danbooru's standard" is meaning, but considering that Janitors are demoted and that this is still albert's site, it is his standard in the end. If they see that something is going haywire or is not fulfilling the requirements, then they are demoted. That goes for Approvers, Flags, Contributors and Approvals.
That in turn means that albert offered us a toll to get rid of certain posts that users don't like. They should write as objectively as possible why a post isn't worthy of Danbooru's standards.
Now to the question: Why would someone flag in the first place? A pretty difficult question in my eyes and I've told you that it is meant as a tool for quality control by Albert (I think). And that is how use it.
Some people are just ignoring bad posts as they are not interested in a post. But then I also demand respect for those users who think that a post is not good enough (anymore) for Danbooru. In the end the discussion in those comments won't solve anything and maybe some users are receiving a ban, because of violation agains howto:comment. What I mean with respect is that one can argue against a flag, but it should be in a non-rude tone and only treat this image. Whataboutism ("There are other posts that are worse but not flagged") for example is a really bad argument, since the only thing it reaches is that more posts could be flagged and it doesn't address the quality of the flagged post.
This isn't a reply to anyone in particular, but objectively, a lot of the posts that do get flagged are flawed. There are different degrees of leniency that different people have, for many reasons, and we have to make a way to respect those two with a better compromise.
Recently after a set of undeletions I did, someone went about flagging 10 or so of them. I wasn't worried about that as a bug causing them to immediately enter deletion had me more concerned. The thing I wonder about is, if one decides that an image does not represent the archive on an individual basis, and then someone else approves the image afterward, it's losing us energy, and we could probably prevent this recurring cycle, but how should we?
I don't blame anyone for thinking perhaps an image that was deleted 5 years ago should stay deleted, if it isn't something particularly special. What I'm getting out of this is that perhaps the idea that an image is flagged in more than a few instances without a prior notice of the weak points of an image is creating more fights, when we could encourage the use of the comment system or a forum topic to head off these fights before they start. That's where my next few suggestions come in: Checkboxes for adding objective tags anonymously by approvers who don't feel an image is good enough and requires a second opinion, so that the images can be in a blacklist which separates those images out without needing flagging, and giving dissenting approvers a chance for images they do feel might belong to at least be given conditional acceptance.
As it is, we have a thread for deletion appeals, perhaps we should encourage approvers who feel that an image is flag worthy to more often use the detailed message part of the approver box and moderation queue more often to make some sort of declaration that an image is in their mind flag worthy. This might discourage at least some approvals and perhaps lead to fewer flags.
I also brought up a comment in a separate forum topic #13673 responding to Mikaeri, and I suggested that we use the tags perspective_warp, anatomical_distortion, and proportional_abberation to denote an objective tag that describes both accidental and intentional use of odd elements of physical design, including cartoon anatomy for obvious deviations from the norm. From reading the current thread, I'd take it a step further, and possibly allow approvers to mark for these things when selecting poor quality or no interest as a check box, adding the tags in a hidden manner, protecting those who use these tags the way we protect flaggers. Having more descriptive tags in general for flaws in images gets people thinking more specifically of whether it should be approved or not.
From there, we can start encouraging the use of blacklisting to extend to these tags, and save flagging for more egregious examples of poor work. If an image gets marked with a tag like anatomical_distortion, then it enters a blacklist region that those who do not want to see can avoid, while those who do not want the image deleted or who want to avoid having approval arguments can get the choice without bothering the first group. Then when an approver chooses it, they have it tagged already with these extra descriptors of image flaws so that a persons concerns were brought up before hand. If we already know an image is flawed, and it is approved, we can say that at the very least, someone had a word in edgewise.
For that reason, I think the poor quality mark could count as a proto flag, and in any subsequent flagging, it will be the first reason, and secondary flaggings will either have to back it, or bring up an original problem with the image which wasn't noted by the first.
I also wonder if perhaps we should use a thread to separately allow approvers to ask if an image will be given these cautionary tags or will be flagged if they go ahead with an undeletion. If deleted images are marked with more notice on why they were not approved, it means we reduce the problem of appeals on images that might not deserve them, and would give those who have images that do deserve consideration a chance to show their appeals have more substance.
Cutting down on both the demand for flags and appeals with objective tags to make the dividing line between approval and deletion will perhaps make this possible, and perhaps leaves us room for accommodating measures that satisfy both sides. I do see one possible flaw to this whole approach, which could be that some approvers might add objective quality tags which obviously don't belong, in which case I'd leave it to the moderator/admin levels to be able to remove these tags if contested. That said, I don't think this could be as easily abused since the point is to encourage the separating of the tagged out images for those who don't want to see art with them, and may perhaps reduce the issue of flag vandals.
Another side idea I'd suggest, is perhaps a moderator level tool (if it doesn't already exist) to allow mods/admins to invalidate single flags, reverting to the prior approver(or to the contributor) if the flag is found to be exceptionally frivolous, but this last idea is just an afterthought and perhaps just not worth it.
Updated
As it is, we have a thread for deletion appeals, perhaps we should encourage approvers who feel that an image is flag worthy to more often use the detailed message part of the approver box and moderation queue more often to make some sort of declaration that an image is in their mind flag worthy. This might discourage at least some approvals and perhaps lead to fewer flags.
Well, then first we have to encourage Janitors to sctually use the tools inside the mod queue. I only know from two other Janitors that they are constantly using the "No interest" etc. buttons with Ephyon and Nitrogen09. There might be some other persons, but if so then it is inconsitently ranging from three Janitors to seven. Randoom numbers, but it should illustrate that this is unlikely to happen even if it's a nice idea. We have to start by the basics, though.
And those objective tags are already covered by the bad anatomy (or other tags). What we could need is a more descriptive definition of this tag. But creating new tags will only create a jungle that is pretty obscure and it would take too much time to have an actual discussion under Janitors. In other words: if it get's implemented, then there probably won't be any discussions at all and it will most likely be ignored.
I suggest that it should stay that way.
If someone is really doing some mass flagging against something, then the user should contact a moderator or better admin that is observing the flags. I did that with your undeletions for example. Those flags should have been monitored. That said: If an image gets undeleted, than it probably should be put to higher scrutiny. Especially if it's a long-term deleted post, since they are pretty easy to flag with "objective" reasons (and no, that doesn't mean that my undeletions are better/worse than yours for example.). That the images were deleted so quickly is a problem, though. It should be addressed now in issue #2865.
Updated
If deleted images are marked with more notice on why they were not approved, it means we reduce the problem of appeals on images that might not deserve them, and would give those who have images that do deserve consideration a chance to show their appeals have more substance.
I agree strongly with this.
From there, we can start encouraging the use of blacklisting to extend to these tags, and save flagging for more egregious examples of poor work. If an image gets marked with a tag like anatomical_distortion, then it enters a blacklist region that those who do not want to see can avoid, while those who do not want the image deleted or who want to avoid having approval arguments can get the choice without bothering the first group. Then when an approver chooses it, they have it tagged already with these extra descriptors of image flaws so that a persons concerns were brought up before hand. If we already know an image is flawed, and it is approved, we can say that at the very least, someone had a word in edgewise.
I don't agree so much with this.
I, personally, don't flag posts because I have a seething hatred for anything containing anatomical mistakes. I will sometimes go into the bad anatomy and bad proportions tags intentionally and comb them for posts that are particularly bad; I think they function as good aggregates for potentially problematic posts. I wouldn't go in and flag just anything, as there are plenty of images in those tags that are very nice to look at and just happen to have a couple errors. I use my own judgement, just as I would for an untagged post, which may or may not be the same judgement as the person who added the tag.
Putting bad_* tags on an automatic blacklist would actually do much more to categorically devalue those pictures than just letting them be as they are now. I prefer a system where the better posts remain visible, and the worse posts do not, and the current flagging/deletion system provides that just fine.
How are you planning to do this? I also agree with it, but like I said, most Janitors don't even use the "X mods didn't like this post enough to approve it".
I guess we'd run into a problem here. The amount of approvers per se would be ok, though.
But in that regard, I see some Janitors that don't even approve that much in the first place. So I guess the biggest problem here is the inactivity of some Janitors.
I can't verify that it is possible to achieve this plan given the weak number of generally active approvers. I just figured at least the mechanism is in line with the progressive use of blacklisting that has grown over the past few years, and we could extend it here. The scale of the idea is also such that testing would take a long time; perhaps features could be brought online over time, starting with the checkmarks for adding tags part by itself.
If several of the most active non-approvers in the forum volunteered to use the mod queue and approve and mark images as not worth approving, such a plan could work. Depends on who wants it and who would actually come online regularly, I don't have enough continuous bits of time through the day to do it like I had before. Many contributors are on here every day, while approvers appear to be a much rarer bunch to expect consistent activity with. Perhaps a couple promotions should go to those in the top contributors camp.
And I see your point feline lump, my view would be that the auto-blacklist would separate out the images into a separate tier, without having to worry about them getting thrown into the deleted box. Those who opt into seeing that lower tier would get the benefit of keeping those images active, and there'd be fewer battles over that, since while it might devalue the images as you said, it also keeps them from falling below the floor, since they're already in many cases considered deletion worthy by some, but not all, so a blacklist sounds like a potential way to reach a compromise. If that compromise comes at the wrong expense can be argued, but I think it may be useful, at least to those of us who don't agree with the current set of anatomy flaggings, while keeping them out of sight of users who really don't want to see them. I'd really like to hear what some of the anti-flag camp thinks about this.
Oh, and off-topic, I talked to Type-Kun about the bugged deletions, so its all good.
Updated
@buehbueh
I'm in total agreement of the second paragrph.
But in order to take discussion about active Janitors (first step) further, I think we should start a proper topic and not use this topic for this stuff. Flagging is only tangentially related to this...
About the third paragraph...Not so much. The only thing that this would create is shifting the blame a bit.
I also don't think that there is really a problem if the flag is written "objectively". As I said, and I repeat this, Janitors should look at the reason first and then at the image, so they know what is fishy there (or not). There is actually no room for arguments over a post's flag, but it leaves room for open discussions in the comment section. And I prefer that much more than a system where a set of images gets such a tag and is then thrown untransparantly into this category. It is like bad anatomy, but only with more work but the same outcome.
Hmm, no opinion from an admin or mod regarding the proposed flagging system by buehbueh and/or the lack of active approvers (actually, the latter is more @albert 's part)? I mean it is nice to have a well and good written comment, but it doesn't help if everything is staying still.
post #2617277
"It's awful" is clearly an invalid flag reason.
And this too:
post #2616965
"LEWD Kanna"
By the way I have a question:
If someone is approving such a post because of flag vandalism, isn't there a possibility that if the post gets flagged with a correct and valid reason that this Janitor can still give their opinion to this post?
Like
user 1's flag is invalid ("Lewd character")
Janitor 1 approves this post because of flag vandalism
-> User 2 is flagging the post with a valid reason: Shouldn't Janitor 1 still be able to approve this post?
Updated
