I've done a bit of work around reason codes/classifications. If you take away the option to give no reason, an actual reason becomes the new default reason and you lose any kind of meaningful analysis on the actual reasons that were being given. I've suggested this previously, but I think the best thing to do is to cap users daily downvotes based on tenure, or admins to do analysis around downvotes per user and remove people who don't seem to be able to grasp the concept of not contributing vs disagreement.
As I understand it, the more downvotes one casts over a certain period, the less weight they carry. The idea of shadow banning is absolutely terrible, and not in the cards here, I should hope.
Like capital punishment is a great tool for fighting murder. Sounds fine until you give it a moment’s thought, something nobody at Reddit itself has done. They make up for it, though, with a strict covenant of silence, so they’ve got that going for them. It’s ridiculously bad policy.
I'm not sure I follow. For true spammers, e.g. a bot account, shadowbans are a great moderation tool, since they don't allow the bot/spammer to know it's been banned. The problem comes when shadowbans are used for other purposes. As long as they're only used as a tool to fight true spam, I don't see the problem.
It’s not a moderation tool, for one thing. At Reddit, mods don’t shadowban, admins do. Anyone can ask for a user to be shadowbanned, including mods, but that’s still not a moderation tool.
What you’re suggesting is, to my mind, like saying capital punishment, if it was only ever carried out on actual murderers, would somehow fight murder itself. In the scenario you describe, it’s trivial for the spammer to detect and confirm that they’ve been shadowbanned. It can be and is, in fact, easily automated. It’s no more effective against determined spammers than capital punishment prevents crimes of passion in the heat of the moment.
In the complementary scenario, where some poor schlub gets shadowbanned and never knows it, they can participate for in some cases years without knowing. So no, shadowbans aren’t magically secret to spammers, but they can be to innocents. Imagine if summary capital punishment was doled out arbitrarily based on whispered guesses made by power mad paranoiacs, and the practice itself was forbidden to discuss. People might object to that policy, which along with ever sending a piece of mod mail, could subject you to that policy.
If you’re working security and someone gots to go, you put them outside. You don’t put them in a sound--proofed cell in the basement to keep an eye on them. Never do someone a minor injury.
Those are great suggestions. Although it may be a bit hard to do the analysis around the downvote one.
In particular, I wasn't lobbying for the "No reason option" to go away but to specifically have an alert come up right when the user hits the downvote button. That way it will potentially dissuade people from just spamming downvotes or not considering the full weight of the downvote button.
If you take away the option to give no reason, an actual reason becomes the new default reason and you lose any kind of meaningful analysis on the actual reasons that were being given.
What? If you force the user to provide a reason, then they will have to pick a real reason. You lose any meaningful analysis by including "no reason" because they are not forced to pick a reason. I think it's okay to have an Other option with a freeform result. You can use that to create other static reasons, if Other becomes popular.
There is nothing that will force them to pick a real reason, there is something that will force them to pick any reason. I've worked in note type/interaction classification analysis in a CRM system for more than one company any when you go from a system of not leaving notes as a habit, to forcing people to leave classifications, you find that some groups inevitably get set up as default options by users unless their classifications are being audited and managed.
By forcing people to leave value, It doesn't just mean you are getting nonsense data input into one category, it means you are also losing the true value of the category as it is not possible to distinguish the true values from the default entries (nonsense data).
I've done a bit of work around reason codes/classifications. If you take away the option to give no reason, an actual reason becomes the new default reason and you lose any kind of meaningful analysis on the actual reasons that were being given. I've suggested this previously, but I think the best thing to do is to cap users daily downvotes based on tenure, or admins to do analysis around downvotes per user and remove people who don't seem to be able to grasp the concept of not contributing vs disagreement.
The metered cap based on reputation is, subtly, already in place.
In the form of shadow banning downvotes (not downvoters)?
As I understand it, the more downvotes one casts over a certain period, the less weight they carry. The idea of shadow banning is absolutely terrible, and not in the cards here, I should hope.
Not shadow banning the user, but ignoring their downvotes. I don't like shadowbanning users either. Great system if it's already in place.
How about downvoting more than a certain limit could reduce their rep or exp points. As long as it was made clear to the downvoter.
Shadowbans are great tools for one use: fighting spammers. Aside from that I agree that shadowbans are a bad moderation tool.
Like capital punishment is a great tool for fighting murder. Sounds fine until you give it a moment’s thought, something nobody at Reddit itself has done. They make up for it, though, with a strict covenant of silence, so they’ve got that going for them. It’s ridiculously bad policy.
I'm not sure I follow. For true spammers, e.g. a bot account, shadowbans are a great moderation tool, since they don't allow the bot/spammer to know it's been banned. The problem comes when shadowbans are used for other purposes. As long as they're only used as a tool to fight true spam, I don't see the problem.
It’s not a moderation tool, for one thing. At Reddit, mods don’t shadowban, admins do. Anyone can ask for a user to be shadowbanned, including mods, but that’s still not a moderation tool.
What you’re suggesting is, to my mind, like saying capital punishment, if it was only ever carried out on actual murderers, would somehow fight murder itself. In the scenario you describe, it’s trivial for the spammer to detect and confirm that they’ve been shadowbanned. It can be and is, in fact, easily automated. It’s no more effective against determined spammers than capital punishment prevents crimes of passion in the heat of the moment.
In the complementary scenario, where some poor schlub gets shadowbanned and never knows it, they can participate for in some cases years without knowing. So no, shadowbans aren’t magically secret to spammers, but they can be to innocents. Imagine if summary capital punishment was doled out arbitrarily based on whispered guesses made by power mad paranoiacs, and the practice itself was forbidden to discuss. People might object to that policy, which along with ever sending a piece of mod mail, could subject you to that policy.
If you’re working security and someone gots to go, you put them outside. You don’t put them in a sound--proofed cell in the basement to keep an eye on them. Never do someone a minor injury.
Ah. That makes more sense to me. Thanks for the explanation.
Those are great suggestions. Although it may be a bit hard to do the analysis around the downvote one.
In particular, I wasn't lobbying for the "No reason option" to go away but to specifically have an alert come up right when the user hits the downvote button. That way it will potentially dissuade people from just spamming downvotes or not considering the full weight of the downvote button.
What? If you force the user to provide a reason, then they will have to pick a real reason. You lose any meaningful analysis by including "no reason" because they are not forced to pick a reason. I think it's okay to have an Other option with a freeform result. You can use that to create other static reasons, if Other becomes popular.
There is nothing that will force them to pick a real reason, there is something that will force them to pick any reason. I've worked in note type/interaction classification analysis in a CRM system for more than one company any when you go from a system of not leaving notes as a habit, to forcing people to leave classifications, you find that some groups inevitably get set up as default options by users unless their classifications are being audited and managed.
By forcing people to leave value, It doesn't just mean you are getting nonsense data input into one category, it means you are also losing the true value of the category as it is not possible to distinguish the true values from the default entries (nonsense data).