Is there a way to add another alert reason for posting untrue information?

[beaglelover]

It would be great if we could alert on a post because the information posted within it is false. Thanks.

[hlthe2b]

1. Usually gets covered under the 'RW memes' or Conspiracy theories types alerts...

I get what you are getting at, though, but I'm not sure how the jury could make that truth determination in many cases.

[CloudWatcher]

19. Wishful thinking, not right wing memes nor conspiracy theories

The #1 post for most of yesterday was a false thread about how "Trump couldn't land at Bozeman because the grifter owes $$$ from 2018". Totally false with 282 recs so far. It was true that he owes a lot of money, but false that it had anything to do with not landing at Bozeman.

Lots of folks complained in the thread that it was bogus, but so many just loved the idea that it got rec'd up to #1.

We have lots of people rec'ing threads that make them feel good, facts be damned. This is not a good look for DU.

https://democraticunderground.com/?com=view_post&forum=1002&pid=19314733

[Think. Again.]

2. That's a good point, a "mis/dis-information" alert definition.

But how to keep it from getting mired in subjectivity?

Edit to add:

Perhaps that same alert could be applied to debunked deepfakes?

[Maru Kitteh]

4. Many alert categories are fairly subjective so it would be no different. nt

[Maru Kitteh]

3. Agreed, especially since one prolific newcomer seems intent on spreading disinformation

which makes us look FOOLISH. They've been urged by many people in the community to stop. The fact that they insist on continuing anyway leads me to believe it to be a malicious act.

[stopdiggin]

5. can certainly understand your thrust

but there would probably need to be some sort of proviso involving 'knowingly false'.
to avoid penalizing those that are themselves being misled ...

and that leaves a 'jury' in a pretty tough spot ...

[consider_this]

15. See my suggestion, reply#13

[ms liberty]

6. I agree. Please. It's getting out of hand and it's not a good look for us.

[Skittles]

45. I think maybe that was the point

as alluded to in the Admin response further down, it's not particularly serious if someone posted something that fooled them, but if you do it over and over and fail to react to DUers responses regarding authenticity, that is a form of TROLLING

[birdographer]

7. That's tricky

One time I saw something on some other site and I wanted to share it here for whatever reason (I forget what it was). I truly believed it was valid, it was from a trusted source. BUT, later turned out to be false. I posted in good faith that it was true. I went back and deleted the post, but by that time someone could have alerted on me. It was unintended on my part.

I think that if we comment that it is false and suggest that the post be deleted as being false, and the person deletes it, that should suffice. If there are comments about it being false and it is left, maybe instead of an alert there could be some kind of request that it be removed by admins? I worry that alerts would flood the jury process.

[consider_this]

16. See my suggestion, reply#13

[Silent Type]

8. The most important thing to do is post the facts in a response. That gets noticed quickly if enough people say,

“That ain’t right because. . . . . .”

[TwilightZone]

9. Usually, that would work, but in one specific case...

the poster has basically stated that they don't care if the posts are true and that they aren't going to fact-check.

[Silent Type]

10. Well, yeah, that would tick me off too.

[CloudWatcher]

20. We seem to have lots of folks willing to rec anything that makes them feel good.

I've seen too many posts that were completely debunked in the comments, yet kept getting rec'd because it was a story we wished was true. People would (I assume) see the subject, go "cool" and rec it, without even glancing at the replies.

A terrible look for DU. Makes us look like a cheap echo chamber of lies.

[live love laugh]

11. It'd be difficult because who's providing the "truth"? The jury is not be able to do research.

[orange jar]

12. I agree with others who say it's probably not logistically possible, but I get your sentiment.

I've been thinking about how DU could have a reporting system similar to YouTube's — you select the closest rule violation (like, maybe spam or kooky content in this case) and then there'd be an option to include additional context/information in a text box.

[consider_this]

13. This idea might work...

If it could be implemented.
If a post could be alerted for false info, an area is provided to state why or link to proof, BUT
Once alerted, immediately some very noticeable text is posted right at the top of post such as 'this post is under evaluation for possible disinformation'.
The poster is notified and given the option to self delete and has a set amount of time to do so. If they do not, it goes to jury, and jury will have information provided by the alert to judge.
This way, if poster put up the info without knowing it was fake, they have the chance to correct without penalty.

[progree]

17. "If a post could be alerted for false info, an area is provided to state why or link to proof"

Yes, that would require the alerter's reason to be re-implemented. The alerter's reason was done away with several years ago, and consequently I don't alert on many posts because I know most jurors wouldn't see the problem without an explanation or more context (sometimes while on jury duty I look up a thread to look for what the big problem is if I can't see it myself, I remember in one case for example, the alerted post was using a far right source in another country and no way would I have known that).

The alerter's reason was brought back for Latest Breaking News violations, but not for any other group/forum posts.

Late Edit, 9/2 606 PM CT The last sentence above is not quite correct, I believe now after experimenting that the only Latest Breaking News violation that has an alerter's reason box is:

"Don't violate a forum or group's Statement of Purpose"

I've experimented with 4 other kinds of violations of LBN posts, like "Don’t peddle right-wing talking points" (etc.), "No graphic content", and "No kooky, extremist" etc., and none had alerter's reason boxes.

(While experimenting, I was careful to "close" and not to actually "Report this post&quot

[dpibel]

24. Entirely agree

Blocking the ability to explain your alert puts an unreasonable burden on jurors to figure out what the problem is.

There are times when I don't alert because the problem is not obvious, and you can't rely on jurors to do any research.

Then again, maybe this is just a matter of old people longing for the good ol' days when we did things right...

[mopinko]

14. there r a couple rules that dont trigger a ffr. like copyright or wrong forum.

if someone posts in good faith, there’s no penalty. but admin can see who the repeat offenders r, and take appropriate action.

it is a good idea. i wd also like a- i think this person is a troll.

[demmiblue]

18. It would be a nice, and needed (imo), feature.

I saw three today. One of them is currently the #1 recommended thread.

[PedroXimenez]

21. at least some indication that it matters, at all

so much talk of misinformation in social media, and DU seems to have no policy at all about it. Lots of policies, i didn't see any about misinformation.

People are saying a rule would be hard to enforce, how about just a policy that says it matters? Put aside enforcing it, how about something saying DU is against misinformation?

Posting this after seeing a thread which was misinformation, which a few posters pointed out but they were all ignored.

[Hekate]

22. Still needs to be addressed, for all the reasons stated. I am quite distressed at the number of times ...

… this year that I have swallowed mis/disinformation. At best I go back and remove my Rec. At worst, I have to go back to a person IRL and retract something I told them was true.

We all have a stake in maintaining DU’s reputation for purveying factual information. Help us do that, please.

[highplainsdem]

25. It's a difficult problem to deal with in a forum like this one where people are in basic agreement on so

much, because it's such a rare problem.

The usual reaction of DUers to being told an OP is incorrect is to delete it or correct it. I'll delete if there's been little response, and I try to remember to post an explanation and apology as a reply first, and I've often emailed people who engaged with the thread, even just reccing it, to explain and apologize.

If there's been more engagement it's sometimes best to correct the OP and edit the thread title as well to add in caps that it's an update or correction, but if you do that, it doesn't lock the thread, and sometimes it would be best to lock it.

EarlG, would it be possible to add an explanation feature to the self-delete feature, so the person self-deleting can explain why they did so? That would help more than simply adding the explanation in what will be the final reply in a self-deleted thread.

I've been on, and sometimes moderated, boards where all posts were reviewed by a mod before they were released for others to see, or where at least all newcomers' posts needed moderator approval for a certain amount of time. But that requires a lot of moderators and isn't a viable option here. (I have memories of logging in on one board on family holidays and finding pages of new posts to review and release, when I was supposed to be fixing breakfast for a dozen people and getting pies for dinner in the oven.) I was wondering yesterday if reviewing posts in advance of release still might be a possible option if a single user was repeatedly causing problems posting misinformation...but that still creates a burden for the people who'd have to review those posts, especially if that person posts a lot.

No one here should be posting misinformation repeatedly. No one should be ignoring requests to correct threads after mistakes are pointed out to them. They shouldn't require babysitting, if they're adults. If they're intentionally malicious, they shouldn't be on the board...and MIRT does a very good job of weeding out most trolls.

I honestly didn't have the impression the now-former DUer posting so much misinformation recently was a troll. I think they.were posting what they thought would be impactful and help Democrats and hurt RWers, and they said as much early on, IIRC defending posts like the one using AI art to suggest Taylor Swift had endorsed the Democratic nominee, saying that was fine if it upset Republicans reading DU. But that isn't the purpose of DU, which is for Democrats to share info and post info for other Democrats. That newcomer misunderstood what this community is about, and I was hoping through yesterday that they would FINALLY understand DU and stop posting the kind of stuff creating problems.

Now that DUer is gone, and I hope it will be a long time before another DUer repeatedly posts misinformation but doesn't want to correct it.

But we still need some sort of alert for.misinformation.

It can't be the regular jury system, which isn't perfect but at least allows for quick responses to an alert. I stopped being available for juries a while back because I don't use my phone for DU and.I might not look at my tablet or one of my laptops for a long time even while logged in. I figured if I don't see a jury request pretty quickly, it's better not to be one of the DUers contacted.

A quick response time, though, isn't always compatible with fact-checking.

It might be necessary to have fact-checking alerts go first to forum hosts (since they'd be the.ones most familiar with any.similar mistakes), but if none of them can take the time to fact-check, then to a list of volunteer fact-checkers, ideally DUers with a certain number of years and/or posts here. They won't need a separate private forum for discussions, unless that would be more convenient.than group emails (which can be enabled on all board software I'm familiar with). Just one news story from a reliable source should be enough to request a correction, with the software both sending a message to the person who posted the.misinformation and temporarily hiding the thread until a correction is made. It should be hidden in that interim so it dorsn't keep getting recs and replies.

And those temporary hides shouldn't count toward FFRs. Unless, perhaps, the person posting the misinformation refuses to correct it.

It still isn't a perfect solution.

I don't know of any perfect solutions.

Ideally, boards should have large enough teams of moderators with most of the moderators having a lot of experience, and with those mods responding to alerts as well as reading as many new posts as possible. Most forums don't have that. Being a moderator or admin of a board can be exhausting. Last year the senior moderator of one of the largest subreddits asked me to join the moderators, after we'd corresponded and they'd made changes I'd suggested. It was flattering but there was no way I could commit to so much time there, even though they use automated moderation software as well to alert them to problem language suggesting bigotry, etc.

If there are enough experienced moderators, they can hide posts and write to the person who made the problem post to fix or delete it, and that's the very quickest way to deal with problems. Alerts can be sent directly to mods instead of to a jury.

But we don't have that here.

Btw, I saw some posts here (#17 was one) suggesting the alerts jury members get no longer show the reason a post was alerted on (except in LBN?). That surprised me, and if that's correct and jury members aren't now given the reason for an alert, I hope that will be changed.

[Hekate]

27. TY, hpd, for a long & thoughtful response. It's going to get more complicated with AI...

When I talked w hubby about our dear-departed RubyShoo before she was shown the door, his eyebrows raised when I mentioned the sheer number of posts in a matter of weeks. In his professional opinion this was an AI-generative volume of output.

[highplainsdem]

29. I don't know. There have been bots posting on Twitter, even from verified accounts, exposed at times

when someone they've replied to tells them to ignore previous instructions and write a poem, or something like that. But I didn't have the impression Rubyshoo was AI. It doesn't take much time to copy something from Twitter and add a thread title and an emoji or two, and she's said in one reply in a discussion thread she'd started about content creators - and where someone asked her if she was one - that she was retired (so much for my early guess that she was very young) with too much time on her hands, and she was distributing content.

Which she was...but unfortunately it was often misinformation.

I don't know if we will get bots here like the ones on Twitter, and I wouldn't advise people here to post a lot of "ignore previous instructions and..." messages to newbies.

What I'm most worried about, re AI, is a tsunami of misinformation and deepfakes starting just as early voting starts. There's already evidence of Russian deepfakes used to influence elections elsewhere in recent months. I suspect they're stockpiling it for maximum impact and minimum time to fact-check and debunk.

[EarlG]

31. Quick response to the question:

"Would it be possible to add an explanation feature to the self-delete feature?"

This is not necessary because in any case where someone has mistakenly posted incorrect information, they can simply edit their post instead of self-deleting it. Self-deletion completely wipes all record of a post from the forum -- even the Admins can't retrieve it. But if you're just correcting misinformation, there's nothing wrong with simply editing your post, and then you can add any additional context you like.

With regard to a system of expert moderators being faster and more efficient than the Jury system, I can assure you that that's not the case. We ran a traditional moderating system with trained moderators for a decade on DU, so I have a great deal of experience with that system. You're right that it is extremely time consuming and laborious. The people who volunteered were great, and were game for the work, but even so, over time, institutional bias creeps in, "rule creep" occurs, and because discussion forum communities tend to naturally divide themselves into groups, accusations of moderator bias and attempts to "work the refs" become frequent. You can imagine, for example, how much pressure moderators would be under during Democratic presidential primaries.

With a traditional moderator system, you need a private space for moderator communications, which can become scandalous if (or rather when) those discussions become public, because people don't like to be talked about behind their backs. Ultimately the community ends up spending an awful lot of time talking about how it moderates itself, rather than getting on with discussions that the forum is supposed to be about.

The Jury system was specifically designed to resolve the problems that we encountered over ten years of working with a moderator system. It is exponentially faster -- for example, a spambot post can be posted, alerted on, and removed within minutes, even in the middle of the night -- and it produces outcomes that are just as good, if not better, than a traditional moderator system. There are significantly fewer alerts and removed posts under the current Jury system than there ever were under the old moderator system, yet the community as a whole is relatively civil.

[highplainsdem]

33. OK. I can see that the jury system can sometimes be faster than moderators quickly responding to

an alert, especially if any discussion of what to do is required instead of moderators acting quickly when they see a clear violation. But often quick editing of a problem post is a much better response than simply removing the post - especially if a fairly low number of removed posts can.result in permanent suspension.

You do need experienced moderators, though.

accusations of moderator bias

There's jury bias, too. There wouldn't be a jury blocklist at all if there wasn't. And the more heated any debate on DU, the more likely some people are to abuse the jury system to go after others. I've never forgotten a thread I saw here where there was open discussion by some DUers who'd been here for years about trying to get rid of another DUer who'd been here for years by alerting on all their posts for any possible reason in the hope of rolling the dice enough and getting enough hides to get them FFR'd.

Nor have I forgotten how often I had posts alerted on in the 2019-2020 primaries forum. I'd get email about many of the alerts, along with guesses about who was alerting.

It's a crowdsourced substitute for moderation, but it has plenty of flaws.

There are significantly fewer alerts and removed posts under the current Jury system

I suspect one reason there are fewer alerts is that unless the motivation is animosity toward the DUer being alerted on, most DUers DON'T want to alert on a problem post or vote to remove it if that makes it at all likely that DUer will be suspended. There have been a lot of times I've refrained from alerting on a post for that reason, and I even emailed you about one when it was a newcomer I didn't want to see get a hide even though they'd ignored advice on the board (and you told me to alert on them).

I wish the number of hidden posts resulting in an FFR resulted in only a temporary suspension, with that DUer then offered a second chance (but that second chance allowing for more than one hide before a second, permanent suspension).

Moderators who are in contact with people posting messages that are problematic can get a much better sense of which ones are trolls or unable to adapt to a forum's rules than random DUers on a jury are likely to have.

But again, it takes experienced moderators.

[EarlG]

35. Well, no offense, but ironically there's a lot of misinformation in your post that I need to correct

The Jury system is not "sometimes" faster than expert moderators. It is significantly faster in every single situation. In the last 24 hours, the longest time between an alert being sent and a decision being rendered by a Jury was 16 minutes. Blatant rule violations, such as commercial spam posts by spambots, are typically dealt with in under five minutes, even in the middle of the night. There is no moderator-powered system anywhere on the Internet that can beat that response time.

I've never forgotten a thread I saw here where there was open discussion by some DUers who'd been here for years about trying to get rid of another DUer who'd been here for years by alerting on all their posts for any possible reason in the hope of rolling the dice enough and getting enough hides to get them FFR'd.

And... what happened? Do you have any evidence that those DUers were actually able to pull off this scheme? Unlikely, because we have all kinds of systems in place to prevent it from happening.

1) You can't just mass alert on all of someone's posts, because as soon as you get an 0-7 LEAVE result, you are blocked from alerting again for 24 hours.

2) When you hit alert, it is sent to seven random DUers. If any of those DUers appear on the alerted member's Jury Blocklist, they are replaced. If any of those DUers posted in the thread that the alerted post is in, they are replaced. If any of those Jury members do not pass the "dice roll" which occurs during the selection process -- which is affected by their longevity on DU, how clean their own record is, etc. -- then they are replaced.

3) The Jury process is anonymous, so you don't know the identity of the person whose post you are adjudicating. Yes, if you really care, you can figure out how to get around this. But you are just one of seven people serving on this Jury. You have no idea who the other Jurors are, and you have no way to find out. You only have 30 minutes to make a decision, or you are replaced. Even if you did have a way to find out who the other Jurors were, it would have to be a massive coincidence for you to be on the same Jury with enough of your collaborators to make a difference. And bear in mind that once you have been called, you cannot be called again for at least 18 hours (under normal circumstances).

In other words, in order to rig the system, you would need to somehow figure out a way to get all of your friends onto the same Jury, multiple times.

AND EVEN THEN...

If you have a post removed which does not break the rules, you can appeal it, and I will put it back. So even if someone somehow manages to get a non-rule-breaking post removed nefariously -- which they can't, let's not forget -- I'll just restore it and remove the strike from the person's record.

Also, I can see the Jury Log which shows who is sending alerts, and who is getting alerted on. If a person, or group of people, ganged up on someone and started alerting all of their posts, first, they'd likely get stopped by the 0-7 lockout, and second, I'd see it in the Jury Log and it would be blatantly obvious what was going on.

Nor have I forgotten how often I had posts alerted on in the 2019-2020 primaries forum. I'd get email about many of the alerts, along with guesses about who was alerting.

Neither you nor anyone else knows how many times you were alerted on, because we do not make that information public. Anybody can say whatever they like in an email! Especially during primary season, when folks just love to stir shit behind the scenes. As for "guesses about who was alerted" -- that is just pure speculation and rumormongering. Most of the time when people make these accusations, they are completely wrong. But they don't know that, because nobody but Admin has any way to know who alerted on anybody else's posts, unless the alerter makes it public themselves.

I suspect one reason there are fewer alerts is that unless the motivation is animosity toward the DUer being alerted on, most DUers DON'T want to alert on a problem post or vote to remove it if that makes it at all likely that DUer will be suspended.

But this is the exact opposite of what you said above. First you suggested that there are roving gangs of DUers openly going around mass alerting on people in order to get them suspended. Now you're suggesting that DUers are shy to alert on other DUers because they *don't* want them to get suspended.

The reason there are fewer alerts than there were under the previous moderator system is simply because it's not possible to "work the refs" under the Jury system, so nobody tries. In the last 24 hours, DUers have sent a total of 20 alerts. Can you believe that? Under the previous moderator system, we would have seen ten times that many alerts in the same time period. Seriously, we would have seen 200 alerts. That's because under a moderator system, people know that their alert is going to be directly adjudicated by a small group of specific people -- maybe some of whom are even friends with the alerter! -- so they do what they can to influence those moderators. It is simply not possible to influence a Jury in the same way.

In the 13 years that the Jury system has been in operation, I've been aware of exactly one serious attempt to undermine it. In that situation, a DU member was using two accounts, which both had multiple thousands of posts. Both accounts also had Star Memberships under different names, in order to increase their Jury chance. The person would alert on a post, and then would refresh their other account in a separate browser, hoping to get called to that Jury.

Note that even going to this insane amount of trouble, they only had the chance of filling ONE Jury slot out of seven. And for what it's worth, as soon as they started doing this, they were immediately caught and banned. What a waste of time.

I wish the number of hidden posts resulting in an FFR resulted in only a temporary suspension, with that DUer then offered a second chance (but that second chance allowing for more than one hide before a second, permanent suspension).

I don't necessarily disagree with this and may revisit the way it works in future. However, I will say this: It is really, really, really easy to NOT get suspended on DU. Look at all the many hundreds of people posting all around you, some of whom have been here for a couple of decades, many of whom have never had a single post removed. I really don't feel super strongly about offering repeat rule-breakers multiple chances. Once someone starts receiving post removal warnings, my feeling is that it is up to them to moderate their OWN behavior. If they cannot do that, then they will inevitably end up suspended.

[highplainsdem]

36. No, I didn't contradict myself.

You wrote

But this is the exact opposite of what you said above. First you suggested that there are roving gangs of DUers openly going around mass alerting on people in order to get them suspended. Now you're suggesting that DUers are shy to alert on other DUers because they *don't* want them to get suspended.

And I made it clear those were different types of DUers. I wrote

I suspect one reason there are fewer alerts is that unless the motivation is animosity toward the DUer being alerted on, most DUers DON'T want to alert on a problem post or vote to remove it if that makes it at all likely that DUer will be suspended

Emphasis added for this reply.

You wrote

The Jury process is anonymous, so you don't know the identity of the person whose post you are adjudicating.

When I was on juries, if I didn't recognize the post from having seen it earlier, it was simple enough to find in a search to look at it in context, because I do think context is important. And again, this is something where an experienced moderator can make a better decision, probably, than a jury. Forum interactions often have a history, and it can make more sense to talk to the people involved than to see if randomly chosen DUers without context can help the situation if they're alerted to one post and that somehow happens to be key. Sometimes what gets removed isn't the worst that was said in an exchange. Sometimes simply deleting an entire subthread where a debate got too heated makes much more sense than anything else - deleting it, not penalizing people for it.

You also wrote

Neither you nor anyone else knows how many times you were alerted on, because we do not make that information public. Anybody can say whatever they like in an email! Especially during primary season, when folks just love to stir shit behind the scenes.

The people I got these messages from were among the most highly respected here, and anything but shit-stirrers. It is true that they couldn't know for sure who was doing the alerting. I had no reason to disbelieve them about the alert, though.

You like the jury system, and you've explained why. I'm sure there are some times it might be more helpful than having experienced moderators.

But I also feel that communication between mods/admins and members posting stuff causing a problem can go a long way to keep problems from developing. Moderators talking to Rubyshoo early on about why her posts were causing problems might have headed off so much that went wrong. The jury system by comparison seems rather scattershot.

[EarlG]

37. You're correct

I mischaracterized what you wrote, and should have made a better effort to read it properly. My apologies.

When it comes to "so much going wrong" with Rubyshoo, I dunno about that. They were causing a disruption, as indicated by that thread you started which resulted in lots of people calling them out by name, complaining about their behavior, and calling for me to make changes to the site rules in order to deal with them. I guess I could have reached out to Rubyshoo and explained the situation, and tried to get them to listen and understand -- I do that from time to time, as MIRT can attest to -- but the signals I saw indicated that they did not seem interested in changing their behavior, so I decided not to waste my time.

As for the rest, it's all moot. I'm intimately familiar with the traditional forum moderation model because it's the system that DU used for the first ten years of its life, and running that system was... a challenge. It involved constant micromanagement, and it produced significantly worse results than the current system. Hundreds of alerts were sent every day, and every single one had to be discussed by a team of moderators who needed reach a consensus on each decision. A vast number of posts were deleted every day. Threads were locked constantly because too much fighting was taking place in the replies. As you note, we even had a moderator function that would delete entire sub-threads in one click. We had some members who would get fifty posts deleted in a week, let alone five in 90 days.

This happened because we tried as hard as we could to give everyone the benefit of the doubt. People who failed to follow the rules were given many chances, despite repeatedly showing us that they could not, or would not change their behavior. We did this even when those people were attacking and abusing the moderators and the Admins personally, both in private and in public. Especially in those situations, because we wanted to do our best to demonstrate that the system was fair and evenhanded and unbiased, and that we could tolerate fucking awful behavior and give people the benefit of the doubt, even when they repeatedly proved to us that they didn't deserve it.

At the end of that ten years, Skinner and I were completely and totally burned out, and if we had not introduced the Jury system (which took more than four years to develop, by the way, just so there's no confusion about how much thought we put into it), DU would probably have ceased to exist and I wouldn't be discussing this with you now.

So I don't think I can make it any plainer than that. Does the Jury system produce perfect results? No! There is no system of forum moderation that can produce perfect results. Sometimes you're going to see decisions that you disagree with. It can't be helped. But the Jury system works for DU by producing a more civil community with significantly fewer disruptions -- and just as importantly, it works for DU by allowing a single Administrator to keep the entire community humming along while avoiding burnout. That's why we're not going back to a traditional moderator system.

[highplainsdem]

39. I'm sorry. I hadn't realized it had become that nightmarish a situation for you, Skinner and the mods.

I'd never heard of a moderating situation as bad as this:

Hundreds of alerts were sent every day, and every single one had to be discussed by a team of moderators who needed reach a consensus on each decision. A vast number of posts were deleted every day. Threads were locked constantly because too much fighting was taking place in the replies. As you note, we even had a moderator function that would delete entire sub-threads in one click. We had some members who would get fifty posts deleted in a week, let alone five in 90 days.

I'd've probably banned members getting anywhere near that number of posts deleted in a week, after giving them a warning first. And I was used to mods being able to act without needing to discuss something with others, to hide or lock or edit posts (the posts always showing the editing info), or sometimes move them to another forum if posted in the wrong forum (not possible here with different numbering systems in each forum, though it would be really helpful for posts locked in LBN), etc. Any discussions in the mod forum were usually about really serious matters such as whether a member might have to be banned, though I'd make unilateral decisions as admin in the most egregious, urgent cases. And there was no separate MIRT, but any newcomers who were obvious trolls were booted immediately (by me when I was the admin) as soon as we became aware of them. There could also be discussions in the moderators forum over a member being unhappy with a moderating decision, though that didn't happen often.

Anyway, I would have found it impossible to cope with the situation you described, too.

I didn't reply to your message right away last night because I was trying to think of a way to combine the advantages of both juries and moderators, but I couldn't think of any.

And none of us want you to suffer burnout.

Again, I'm sorry.

[marble falls]

41. I get a good explanation of the alert on every jury I get called to and there are a lot of alerts lately ...

... Part of the reason it seems like it takes a while to fix MIs, is the way we do it: through committee. But it works well - fewer moderators going "off the rails" (I have, you've fixed a few for me), consistency in application of sanctions on the breaking of consistent rules is a good thing.

We'd all like to see little fixes, but I fear that over control and the complication involved with making them happen would one just plain lock up DU.

[ms liberty]

23. Kicking. We still have this problem, y'all.

Something needs to be done.

[Silent Type]

26. As long as it is demonstrably untrue. Opinions unfortunately are more difficult. And most of politics is opinion.

[Hekate]

28. Alternative facts are not a thing at DU -- and that's what we are talking about

[EarlG]

30. Unfortunately there is not a way to do this, at least not the way you're imagining it

I'll preface this by saying that while it seems like a no-brainer suggestion to disallow "untrue information," part of my job is to think of all the ways that something like this won't work. And I'm afraid there are lots of ways.

First, sending fact checks to Juries would be a bad idea, because the Jury System is almost the opposite of a fact checking machine. It is a subjective system, not an objective one. When we ask a Jury member, "Is this a personal attack?" we are asking them for their subjective opinion, because there is no way to objectively define a personal attack. So we take the aggregate of people's subjective opinions and use those to decide whether a post breaks a rule or not. We can't do this with facts.

Second, if we did send fact checks to Juries, most people would not be interested in taking part. The vast majority of DU members are willing to take a minute to give their opinion on whether a post that has been put in front of them violates a particular rule, but I cannot imagine that the vast majority of DU members would be willing to go and hunt down whether, say, a quote by some famous person that was posted on Twitter is actually true or not.

Third, if we sent fact checks to Juries, the system would be flooded with "Don't post untrue information" alerts on every single post that anybody was having an argument about. Imagine a post that says, for example, "The current situation in Gaza is Israel's fault." That's definitely going to get alerted on by somebody as "untrue information" -- so is the answer true or false? More to the point, will anyone be satisfied with the answer that the Jury gives? Or will it just set people off on a round of meta discussion, complaining that the Jury system is broken because it did -- or didn't -- remove that particular post?

Fourth, what about satire and humor? The Onion only ever posts untrue information, yet it is perfectly fine to post content from the Onion on DU. If we had a "Don't post untrue information" rule, then anyone who posted content from The Onion, or Andy Borowitz, or any other satire source, could have their post alerted on and removed.

(I'll mention here that if you're thinking, well, we could just add a satire exception to the rule, I'll say now that adding lots of explanatory text/verbiage and/or introducing conditions/exceptions to the rule would likely only serve to further confuse Juries and posters alike. Rules need to be as simple as possible to understand, follow, and adjudicate.)

Fifth, once the system gets flooded with alerts for "untrue information" and Juries are responsible for deciding what's true and what isn't, nobody will be satisfied with the results. Lots of people will have posts removed for posting "untrue information" when in fact what they posted was entirely debatable (this is a discussion forum after all). They will appeal up the chain, and then ultimately the role of fact-checker-in-chief will fall on me. I'm afraid I do not have the time (nor the inclination) to be responsible for dealing with dozens of fact checks every day, the vast majority of which will turn out to have been removed incorrectly because Juries were acting subjectively and not objectively.

Sixth, perhaps you're thinking that if Juries aren't the way to go about this, we could have a team of people whose job is to check the facts. Well, those folks would also get flooded with fact check requests on mostly subjective posts. It would be a significant amount of busywork for very little return. We'd have to marshal volunteers and create a new software system to handle fact check requests -- and even after all that, I would still end up being the person ultimately responsible for signing off on all the fact checking.

So what is the solution?

In my opinion, while the system that we already have is not perfect, it's better than anything else I've been able to come up with so far. Currently, we simply crowdsource the problem, organically. Rather than having an alert that sends a fact check request off to a small team of fact checkers -- which would be laborious and highly inefficient -- it makes way more sense to me when people just show up right there in the thread and say "Ahem, this is wrong, please correct it."

When I see an OP that seems too good to be true, and I scroll down and a number of the replies say something like "Just so you know, this quote is false -- here's the true story," or whatever, that's good enough for me to disregard the info in the OP, or at least double-check it. This system is fast, and makes use of all DUers (because anyone who happens to know the truth of a matter can just drop into a thread -- no special fact-checking team required). It is also public facing -- sometimes, even fact checkers get it wrong, and so it's fine to have a debate within threads over whether something is true or not. That debate does not need to take place behind closed doors.

This has obviously come up because recently we had a member who repeatedly posted false information and then refused to correct it when challenged -- even after the incorrect information had fooled dozens of people into recommending their posts.

To be clear, I do not consider individual instances of posting false information to be a serious issue. These days, anyone and everyone can be fooled by something they read or saw on the Internet, and in their excitement to share it, can post it without checking. If you do this by accident, I do not consider it to be a very big deal, and I do not want to punish people for it via the alert system (along with introducing all the problems described above).

But if you are informed that you've posted something that is essentially "fake news", you really should take the time to double-check it, and then correct it. And if you post "fake news" over and over again, but are not willing to edit or remove the false information even after being informed of the truth, then I consider that to be deliberate trolling.

To conclude, I'm not saying that there isn't a way to improve the current system. But I'm saying that it's not as simple as just adding a new rule and saying "problem solved." I'll continue to think on how we can improve things in this area, but at the same time I hope everyone understands that I have put quite a lot of thought into it already...

[highplainsdem]

32. Re what you said about fact-checking just via replies...

When I see an OP that seems too good to be true, and I scroll down and a number of the replies say something like "Just so you know, this quote is false -- here's the true story," or whatever, that's good enough for me to disregard the info in the OP, or at least double-check it.

It's good that you scroll down the replies, at least when something seems iffy, so you'll notice any corrections of the OP there.

But a lot of DUers apparently DON'T ever look beyond the OP before replying or reccing, so we can end up with OPs that are very high on the Greatest Threads list long, long after multiple people have corrected them and pleaded with them to correct or delete the OP.

And every single mistaken rec increases the odds that there will be more recs and that the.OP, with its incorrect information, will get copied elsewhere.

LBN automatically asks people posting to check for duplicates.

Twitter has a system where you're sometimes asked, when you want to like a tweet linking to an article, if you want to read the article first. (I can't recall if that's only for tweets from a magazine's account. I know there have been times I've run into that after reading the article when an individual, maybe the author, had linked to it, but I still wanted to like the magazine's tweet about it if I saw it later.)

If there is no way to quickly deal with misinformation here, a reminder to DUers to check for any corrections in the replies before responding/reccing could be very useful - whether that reminder is automated or pinned to the top of GD (where this problem occurs most often).

[EarlG]

34. There IS a quick way to deal with misinformation

It's the way I described -- post a rebuttal as a reply, with the correct information. If you arrive in a thread, and you know that the OP contains misinformation, there is no faster way to counter that misinformation than by simply posting a rebuttal right there and then.

Beyond that, I do agree that it is a problem when misinformation OPs manage to get high up on the Greatest page. But ultimately, despite the recent to-do over Rubyshoo's posts which prompted the calls to do something, I do not believe that this is a very big problem. It is quite rare for a misinformation OP to make it all the way up the Greatest Page without being corrected by its author, once the misinformation has been pointed out by people replying.

Nevertheless, for the rare occasions when a misinformation OP does sneak past and climb the Greatest page without the poster correcting it, we may be able to come up with a fairly straightforward solution.

[ms liberty]

38. Thanks for addressing this, and I see that Rubyshoo's been ppr'd

I noticed the lack of OP's from them as soon as I opened DU this morning.
I agree that it's not a problem right now in general, but it was a problem with that individual and to have allowed it to continue would have been to give the trolls a road map (the ones who weren't too stupid to read it). Thats only one of the reasons why we all need to be vigilant about it.
Being aware that this can happen should make all of us be more careful with our sources, our recc's and our comments.

[marble falls]

42. And don't forget: use that alert button as necessary.

[Ms. Toad]

43. How about a fact-checking team,

Like the MIR team, at least for the election season.

Maybe limit it to election-related posts, both to avoid things like Gaza, and because those are the posts that can do the most damage, if false.

[Skittles]

44. this makes sense to me

thank you