• 175 Posts
  • 9.48K Comments
Joined 2 years ago
cake
Cake day: June 19th, 2023

help-circle
  • If you want an oversimplified concept, it’s a fairly simple concept

    Effectiveness probability vs risk probability vs success probability.

    If you can succeed, will it do something useful, and what will it cost.

    Understand though, we’re all armchair strategizing here. Anyone that’s going to actually do anything isn’t sitting around bullshitting on lemmy. So it’s really moot, a circle jerk. That being said, apply the metrics.

    Example: shooting Trump. Low probability of success for the average citizen. High cost for everyone in the country. Won’t achieve much by itself. Violence is not a smart choice.

    Another example: using decentralized attacks against ice agents, with the goal of both reducing their numbers and pulling their focus away from current targets. Good chances of success; they’re easy targets with an organized local team backed by larger numbers giving cover and distraction. Effectiveness is high if successful, any decrease in gICEtapo is a net positive. Risks are high; very high. Local teams are likely to take casualties, but of more import, it changes the fight irrevocably towards civil war. Ergo, it would need to be a last ditch option from the perspective of your typical democrat that wants to preserve the system but change implementation.

    Everything comes at a price when you’re waging this kind of fight. Look at just the previous century, at the various revolutions before that. Any time the people are in a situation where their own nation is the enemy, blood is shed. And by enemy I mean actively acting with violence against the populace, or where the populace intends to take or retake governance of their own nation.

    Before any individual or group decides that it’s time to accept that fact, they have to be prepared to pay the cost. That cost is not static. The cost in lives alone can shift over days, much less weeks or months. Up or down. Look at d-day at the big end of things. The decision to launch the attack had to weigh losses vs gains in lives, as well as gains in territory and gains in progress vs the enemy. If they launched in the wrong weather, the balance shifts. At the wrong time of day, balance shifts

    But, most relevant to here and now is the opportunity to seriously dampen fascists before they consolidate power entirely. Could that be achieved in other ways at all? People like to argue over whether or not killing hitler before he came to power would have prevented the war entirely, whether it would have changed the timing and nature of the war. There’s no real answer to that. The only thing we can know for a fact is that the kind of people that engage in fascism do not stop on their own.

    Ugly truth. A revolution here in the US would not be against the people in charge. It will be against police first, then national guard, then standing military. It will also be against anyone and everyone that supports the current regime and anyone scared of the chaos civil war brings.

    Frankly, I don’t think the majority of just democrats have the will to do it. I think the left the actual left doesn’t have the will to do it. So it’s an uphill battle from the start.

    High chances of accelerating towards civil war as soon as the first deaths start dropping cops. High chances that it will be the excuse fascists are looking for to initiate a planned coup that eliminates the facade of democracy they hide behind currently. Even higher chances that people not directly involved will pay part of the price, no matter what happens after it starts.

    That’s a shit ton of risk, and a very high price to pay. So it would need to be a very good plan backed up with a support resistance network.

    Nobody wants that first shot fired. Nobody wants the death and misery it will bring.

    But will any lesser form of violence work? The fight is already asymmetric. The usual protest against armed cops methods are purely defensive so as to not escalate. How would you mount an offensive against a police force without bypassing their defenses? You can’t. Their defenses are good enough to handle the very less lethal tools they use. So even if you were on a level playing field with rubber bullets and tear gas and water cannons, they’ve spent your tax dollars on body armor, shields, armored vehicles, etc. So the playing field won’t be level at all, unless you use superior weaponry.

    There’s nothing better than what you’d be facing that isn’t lethal because the places you can do enough damage to take a riot cop down and out of the fight is more likely to be lethal or permanently disabling. Head shots, using explosives, shooting under shields at legs. If you want to take them down in numbers, there is zero chance that deaths will not occur.

    Maybe, maybe with enough numbers you could stampede and overwhelm, disarm and contain. But the closer you get, the more of your own get dropped, and the more the less lethal munitions creep into lethal territory.

    So, would it be better to strike when they aren’t massed? Well, yeah. If they’re off duty, you could conceivably just kneecap a bunch of them into inactivity. But you still face the escalation. But you’ve at least cleared some of the numbers. So you get similar risks, with similar outcomes. But you also need to have more skilled attackers. You can face roll a line of shields. But if you’re tracking down and taking out individuals, you run into access issues. Getting to them where they are, in a coordinated way. That takes more training than I have for damn sure. And I’m a nutter that’s fairly well trained.

    Plus, you gonna shoot the cop in front of their kid? Even non lethal, just injuring them with some close proximity beanbag rounds, can you pump a handful of those into someone’s chest while their mom watches? That reduces how many people will join in. It’s a much higher barrier of entry. And it takes more organization and planning, despite the seeming ease of getting the job done. I’m not capping anyone in front of their kid unless I have no other choice at all, or even just taking a bat to their knees (not that my crippled ass could pull it off to begin with, I’m talking about the barrier to entry here)

    So, when is it time to do that? Now. But nobody is ready. The radical left that’s armed isn’t organized. The democrats that might be organized, aren’t organized with that in mind, so it doesn’t count.

    But it’s not an impossible task. You get even a small core group in each city organizing decentralized attacks screened behind other events, you could do it. You could cripple police forces, seize their weaponry, and be ready for the inevitable. If you do it fast enough, you can maybe seize national guard depots after, if you know where they are and can get moving before they spool up.

    That slows response times if there aren’t already standing military forces in place and ready. There’s plans in place for that kind of attack though. The guard is aware that have a chance of being targeted as a resource, so it isn’t exactly a cake walk. Their armories aren’t easy to get into, but you could deny access to them via destructive means. Unless you have people inside, which changes things a lot. Good luck finding anyone on the inside if you don’t already have contacts though.

    So, that oversimplified equation of risk, effectiveness, success is only really useful for parsing the decision to take action.





  • I tend to ask how bad that worst act was. Some shit, ima judge your ass for even if it was only once, no matter how much you say you changed, because you might be faking the change to keep from being noticed.

    But the standard is pretty fucking high if it wasn’t personal to me. Like, if I saw you kick a dog that wasn’t even remotely aggressive, ima judge your ass, but be open to you changing. Eventually, if that’s the worst you’ve ever done and you at least pretend to have changed well enough to not repeat it, I’m not going to hold that over your head forever.

    But you kick my dog? Assuming I can’t get away with burying your ass, you better hope we never run across each other in the woods when we’re both 100, because it’ll be your last day on earth. No forgiveness, no benefit of the doubt.

    But really big shit? No, Hitler never gets to be forgiven, period. Yeah , he’s dead, but ima judge that motherfucker until my last conscious moment

    Now you brought up a specific example. It happens to be an example that I would absolutely never, ever trust the person again. Three reasons first, abuse of position. He already proved he will ignore professional ethics as well as general moral ones, as well as laws. That’s hard to change in people. Not impossible, but hard.

    Second, he got away with it. That means his regret is dubious. If he truly regretted it, why’s he divulging it that way instead of taking steps to make him less likely to be able to repeat things?

    Third, people that diddle kids don’t tend to stop. They just tend to pause. It’s impossible to nail down exact numbers because there’s some that never get caught. But the recidivism rate is high, and it is very rare that someone willing to do that can stop with just one act.

    So, no, I couldn’t believe that any good he did balances out there near certain fact of the only reason he didn’t diddle every single kid he could have was because he didn’t think he could get away with those. He absolutely would have fucked every single patient that met his victim profile if he could have.

    Also, it’s okay to say pedophile, the word police aren’t going to spank you.

    Tbh, I can’t think of any situation where I could weigh any amount of good acts against fucking a single child and that one bad act not taint everything else. And it doesn’t have to be literally fucking, I’m speaking figuratively here. There might be some act that was so irrelevant to the victim that it might not matter if the good was good enough, but I doubt it. I can’t think of anything at least.



  • That’s only a problem if abused.

    And it isn’t. The nature of the beast is such that there’s always admin oversight.

    More importantly, nobody can own a topic here. It doesn’t matter if one asshat abuses their position, there’s still 5 other c/bogantalk communities, or there can be.

    Having mod positions in multiple communities isn’t inherently abusive. Reddit conditioned us to think so, but it really isn’t the case.

    Nor am I personally willing to indulge my misanthropy to the extent of defaulting to the assumption that anyone moderating is on an inevitable path to douchebaggery. Which is separate from there being a severe lack of effectiveness to it when people have tried to go all god emperor over the last couple of years. Power mods that try it, fail, and become power mods over emptiness.

    If a given mod is both fair and effective, it doesn’t matter how many communities it is at all anyway. You’ll never know how many they moderate because there’s nothing to see but stable communities.

    It’s hunting for a problem that just isn’t here.


  • I mean, BPR, since you even admit you were blanket voting without any other criteria, then went back.

    That’s separate from the mod in specific power tripping though. Yeah, you got the reaction you wanted, but that particular mod is already well known to be batshit.

    Look folks, there’s always going to be edge cases where one of us is going to down vote persistent types of posts. I’ve done it, I’ve seen people that I know are reasonable, decent folks do it. But at some point, vote blitzing ceases to be a useful form of action and turns into just wagging dick to feel good about ourselves.

    So, yeah, now that mods can see blitzes easily, we can all expect to see more bans for it. And, inevitably, there’s going to be cases where it looks like a blitz, but wasn’t intentional. So we can look forward to posts here because of those errors.

    What’s the answer though? You can’t tell at a glance what a user’s intent is, or if they’re paying enough attention to realize they’re even voting on things in the same community. But, that kind of vote manipulation is also a fucking problem. Mods need some ability to keep their communities rolling along as intended. Lemmy, and even piefed, are pretty damn sparse with mod tools. Limited options to shape how a community is guided and help keep a vibe beyond just banning people.

    Having something other than gut instinct to use as a metric for keeping out bad actors is necessary. And the truth is that most people down voting multiple posts in a row aren’t there to participate in the community at all. They’re just reacting emotionally to titles and thumbnails.

    Which is fine! Nothing wrong with that, and it’s the best argument for down voting being an option because it’s great at filtering out people that just want to gripe about the surface of things. A down vote makes it so that less of them come in and grump at other people.

    But it does mean that mods are going to have to judge you based on that alone. Nobody has time to chase you down and ask your motivation.

    And, OP, just in case, it doesn’t matter what the community was, what the topic of the posts were. I’m not arguing over that side of things; I’d likely have down voted a few before having to decide if a report about the community to admins would do any good, or if I should just block it and move on. I’m talking about the reality of how voting as an aspect of threaded forums works, and the outcomes of it shape the nature of lemmy/fediverse activity.

    All of it is about balance. Any tool can be abused by mods, admins and users. The tools’ merits aren’t solely determined by their potential for abuse. The benefits have to be weighed. It has to factor in what other tools are available (or aren’t).

    My advice to mods? Use vote records sparingly. If you’re not intending to very rigidly screen for your community as a place of safety, chances are that it won’t help much. You’ll run yourself ragged trying to figure out who is and isn’t engaging in fuckery as opposed to fat fingering buttons, or is voting their conscience and just happen to be opposed to the subject matter, or genuinely think that a post or comment isn’t on topic and relevant/useful and think that about multiple posts.

    It’s a sucker’s bet. Save it for when you’re trying to get a new community rolling, or when your community is a vulnerable shelter.


  • Sure, but what I’m saying is that part of the problem with power mods is that they can’t do the task properly in the first place because the ability to moderate fairly and effectively is Sudbury directly relates to how much work and time a mod is willing/able to put into it. It’s trivial to moderate dozens of lemmy communities both fairly amd effectively when you only run into trouble on a given C/ maybe once a month due to low volumes of users.

    A lot of people start with the assumption that a moderator with a lot of subs/C/s under their management is there to fuck shit up from the beginning, and that just isn’t the case on average. Yeah, people can and do moderate to serve their egos sometimes. And it was very common on reddit to see people that didn’t start that way get that way.

    But this isn’t reddit. The user base is different, the people running instances are different, and vary in their oversight of mods. The mods are different too, so far. Go back and scroll this community. There’s assholes here and there, but it isn’t because they’ve swept up dozens of communities and hold the names hostage.

    It’s impossible to stay a power mod on lemmy for long because once you start being a douche in general, someone is going to another instance and start a new community and people leave yours. It’s happened. Multiple times. Look at ten forward, or the 196 communities.

    Even the dominance in numbers that .world has isn’t going to make communities there immune to being abandoned and the mod/s left with the crickets.

    Power mods here are an annoyance at most. So until the user base gets so big that even obscure little C/s that see posts maybe once a month are all getting high traffic, workload is a factor in whether or not moderating a high number of communities is viable as a way to abuse the position of moderator.


  • I respect that opinion, but I’ve seen too many cases where communities het as swamped by assholes that had a long standing pattern of being assholes. Not just reddit and lemmy as far as that goes, this stretches back to the nineties on forums and chat rooms.

    You can predict future behavior based on past behavior.

    Use the Nazi bar example. The bouncer sees this dude walking around the neighborhood with swastikas on, it doesn’t matter if the asshole is wearing a suit and pays the cover charge when he tries to come in, he’s still not going to be a customer you want.

    If I go around trolling vegans, and the mod of a vegan sub bans me only from their vegan sub, that’s legit. If I run around throwing slurs, and I get banned from communities for the groups I used slurs towards, fuck me, I deserve the bans. When it comes to bigotry in specific, I’d deserve to be banned everywhere.

    But it does need to be surgical, and appropriate to the degree of douchebaggery involved.

    As a tool, refusing to let people into a community/forum that will absolutely contribute nothing is a useful tool.



  • I don’t disagree about power mods. I’m saying that using an arbitrary number of communities won’t prevent that by itself. It’s a more complicated issue to address. A power mod could have two or three communities and if they’re the right ones, dominate a topic.

    That’s the problem, in my mind, that the limit would be arbitrary. No matter what number you pick, unless it’s one, you can’t totally eliminate power modding methodology from controlling discourse. And, like you said, it’s trivial to bypass to begin with.




  • I disagree.

    While I absolutely question the carnivore diet as a general thing, the way the community is handled has shifted from the way it started. Back in the beginning, I was in the same frame of thought you are, and said so directly in the community (eventually).

    But jet in specific, and other users, have shifted to better citations, and a more frequent way of presenting their opinions. I do not believe they engage in misinformation currently. If I believed so, I would be obligated by my personal ethical code to attempt to have admins shut it down. Can’t say it would succeed, but I’d have to try.

    Since jet in specific is very good about putting in disclaimers that people should approach the diet with care, I definitely can’t call it bad faith. Acknowledgement of the diet being limited in scope for the general population is good faith discussion by default.

    Again, I definitely disagree with the claims made by some of the people in linked videos. What they recommend doesn’t match current best practices, and is usually extrapolated from data that is specific to limited circumstances and applied generally, which is very flawed. There’s also non medical issues with attempting a pure carnivorous diet, but I doubt those would be relevant to this.

    As such, I still maintain that down voting everything willy nilly is a justifiable reason for bans, which is what this community is about.

    If you feel that the community is dangerous, harmful, or otherwise shouldn’t be accessible, throwing down votes at everything is not the appropriate method to address the issue. And that’s what the post was about, not whether or not the community should exist. If you genuinely feel that strongly about the dangers involved, you should be contacting the admins of the instance and explaining that to them.


  • Well, since it seems it was done in error, I suppose the issue is at an end. I would have said PTB though.

    I will still say that preemptive bans are a valid tool. They have to be applied surgically to prevent problems, but there absolutely are multiple reasons to ban someone from a community they haven’t yet interacted with.

    I’m not sure this one meets the standard though. Even if you had made the comment, or a similar one, seriously, I’m not sure I can see how that would make you likely to go trolling on a vegan community.

    Shit, I have actively taken the piss out of the asshole segment of the vegan population, and I don’t go to vegan communities and pull that bull. So I can’t see why a single comment would be enough of an indicator of future risk. And that’s when preemptive bans are a valid tool, when someone is saying or doing something that shows they’ll have a high probability of going to a community and cause problems. Not just because they hold a different opinion (no matter how wrong that opinion may or may not be).

    Which, despite it being resolved, that is why I would have said PTB to begin with.

    However, I’ll also say that Sunshine is really easy to get along with normally. Everyone has bad days and makes bad decisions sometimes, and I think that’s the case here. An otherwise very sweet person fucking up in a way that’s not the norm.


  • Nah, lemmy is way less intensive work on the mod side of things still. Most communities, you’ll go weeks without needing to do anything at all.

    This community here has a bias towards high traffic communities. The more traffic you get, the more reports you get, the more chances of someone doing stupid stuff that needs mod action. And, with a higher volume of mod actions, there’s a higher chance of even the most level headed mods fucking up in one was or another.

    Most of lemmy, one person could moderate a hundred communities and you’d never know that they were there at all.