Debunking Misinformation & Hysteria Surrounding the 1.19.1 Chat Reporting Update

  • Preface

    While I believe some criticism of the chat reporting system is founded, I've seen a fair amount of lies and straight spread about the chat reporting system. In this thread I seek to disprove some of these false claims and leave my personal opinion at the end. Bold are statements of fact and other text is just additional information.

    Forgery

    Cheaters or the server are unable to forge signed messages as you, without having access to your computer's filesystem.

    Context forgery no longer works (e.x. nodusclient/gaslight, video here); as the signature you send chat messages with is now based on the previous chat message. As such, if the real chat message signature is not based on the previous reported chat message (which in context forgery attacks would be malicious), your report will be thrown out and deemed as false. Here's a nice video with a visual explanation explaining why context forgery doesn't work anymore.

    Report Endpoint Spamming

    This endpoint will be heavily limited.

    False Positives/Negatives

    This issue exists with every moderation system, and isn't unique to Minecraft

    Breaking the premise of Anarchy servers

    Mojang are only going to be banning the worst of the worst and you shouldn't be saying racial slurs anyways, additionally Anarchy servers will most likely disable chat signatures with a plugin or jarmod.

    Mojang/Microsoft will spy on your chat messages

    The server has not been updated to send every chat message directly to Mojang, the only messages that are sent to Mojang are those the player reports to them.

    My Opinions

    While I'm not sure this was entirely needed or even asked for, I can see why Mojang and servers without enough manpower to moderate chat messages would want to implement this. This definitely won't impact the gameplay for me.


    Mojang have also repeatedly shown they're not going to back down just because a bunch of teenagers with nothing else to do join a Discord server and start screaming "oh noes this is the end of minecraft!!!" and spreading misinformation

    Edited once, last by Allink (July 29, 2022 at 9:49 AM).

  • While I'm not sure this was entirely needed or even asked for, I can see why Mojang and servers without enough manpower to moderate chat messages would want to implement this. This definitely won't impact the gameplay for me.


    Mojang have also repeatedly shown they're not going to back down just because a bunch of teenagers with nothing else to do join a Discord server and start screaming "oh noes this is the end of minecraft!!!" and spreading misinformation

    I'm in a similar boat in that it doesn't effect me much at all. But to be perfectly honest, I still see why this is an issue.

    The main fear is messages out of context or sarcasm taken seriously. Many popular personalities have been banned from social medias or games this way, and server admins are scared of getting attacked and banned. Report bots might even become a thing. Spambots already are, so it's very likely (The TF2 bot crisis showed bots can be coded to co-ordinate a vote-kick and get you VAC-banned). And it can't just be fixed by "being careful about what you say" since that's just the stupid argument people use about similar places like twitter where everyone is on edge or cancelling everyone. We don't want a game to become a literal warzone like that place.

    As for why you see why Mojang wants this, the reasoning is heavily flawed in their favour. They don't give a damn about that since if they did, it would be an optional choice for server owners. The main reason they did it? Save their own asses from angry parents and laws that protect children online (Some of the dumbest things because IMO children shouldn't be allowed to use the internet unmonitored and unrestricted anyway).

    According to quantum mechanics, unless something is observed, there is an equal chance of it both being and not being there. Hence it is said to be in superposition, until observed. However we are somehow fully certain, despite never having directly observed, that you indeed have no bitches. Your bitchlessness has broken the rules quantum mechanics had established. Indeed an impressive feat.

  • The main fear is messages out of context or sarcasm taken seriously

    I think Mojang stating that they would only ban people for the worst of the worst should counter-act this. And also, the context should be sent in a report. Additionally, if your friends are going to report you for a harmless joke, you have bad friends.

    Report bots might even become a thing.

    Then these accounts "report botting" will be banned from making further reports by Mojang, if I had to guess. Plus, like, ratelimiting exists.

    The main reason they did it? Save their own asses from angry parents and laws that protect children online (Some of the dumbest things because IMO children shouldn't be allowed to use the internet unmonitored and unrestricted anyway).

    Fair, but I'm lead to believe it was mainly a train of thought that started with Mojang implementing due to LiveOverflow creating this video, in which he shows how to modify message sent by a client with a proxy. Which, in all fairness, isn't a very practical exploit. E.x. "Oh, we just implemented chat signatures, that would be a good thing to add!".


    I think the whole controversy is in part due to Mojang's historically lax Minecraft-wide moderation on Java Edition. Whereas on Bedrock it's been the opposite of that for years.

  • I think Mojang stating that they would only ban people for the worst of the worst should counter-act this. And also, the context should be sent in a report. Additionally, if your friends are going to report you for a harmless joke, you have bad friends.

    The fact that your naïve enough to really believe that is kind of worrying. Dude, they banned the word "night" on bedrock edition because of the first three letters relating to a racial slur, and even if it wasn't intentional it's a huge show of how careless they are about the side-effects. Their "worst of the worst" standard is going to be scarily easy to reach. Also, I'm not on about friends, I'm on about public servers like this one. Like I mentioned, "don't say stupid shit" isn't an excuse, it basically saying "live every day in fear".

    Then these accounts "report botting" will be banned from making further reports by Mojang, if I had to guess. Plus, like, ratelimiting exists.

    It'll work for a while, but again, we're talking about a community full of hackers, modders and coders. They can and will find a way eventually.

    Fair, but I'm lead to believe it was mainly a train of thought that started with Mojang implementing due to LiveOverflow creating this video, in which he shows how to modify message sent by a client with a proxy. Which, in all fairness, isn't a very practical exploit. E.x. "Oh, we just implemented chat signatures, that would be a good thing to add!".

    They could've easily added a signature system without the reporting features along-side. And again, like I said, if they were really so rabid as to take moderation under their control, give us an official option to opt in or out.

    I think the whole controversy is in part due to Mojang's historically lax Minecraft-wide moderation on Java Edition. Whereas on Bedrock it's been the opposite of that for years.

    It's because they've proven with bedrock edition it's a horrible idea. The "night" example is one of many. And I know it's not an oversight and a sign of laziness or overdone censorship since they had enough of a brain to not ban "dick" since it counts as a name. Their blanket-banning words without thought, and many people report false bans on bedrock edition. Not to mention their censorship extends to private realms and single player and is locked to always be on unless edited externally. The worst part is a bedrock ban also locks you out of single player. So if you mess up online, you're basically banned from your own private worlds too.

    Imagine you get banned from public events for being a dickhead. That's warranted. But you come to your home to find out you've been evicted by the government (not your landlord or anything like that) for being rude at a unrelated location, are being forced to leave the country and can't ever come back. Sounds pretty unfair, right? Almost totalitarian?

    According to quantum mechanics, unless something is observed, there is an equal chance of it both being and not being there. Hence it is said to be in superposition, until observed. However we are somehow fully certain, despite never having directly observed, that you indeed have no bitches. Your bitchlessness has broken the rules quantum mechanics had established. Indeed an impressive feat.

  • Report bots won't do anything. The banning is manual as Mojang has stated.

    I was on about bots coded to mass-report users. Those are very likely to be used on the aforementioned out of context messages.

    The previous 5 messages are sent over to Mojang to ensure that this isn't the case, and even if that wasn't the case, spamming the n word isn't exactly "sarcasm"

    I doubt 5 messages are ever enough context when two friends are having banter, throwing in a few friendly insults, and some jackass comes in and reports them for "harassment and bullying" or some shit.

    According to quantum mechanics, unless something is observed, there is an equal chance of it both being and not being there. Hence it is said to be in superposition, until observed. However we are somehow fully certain, despite never having directly observed, that you indeed have no bitches. Your bitchlessness has broken the rules quantum mechanics had established. Indeed an impressive feat.

  • The fact that your naïve enough to really believe that is kind of worrying.

    Dude, they banned the word "night" on bedrock edition because of the first three letters relating to a racial slur, and even if it wasn't intentional it's a huge show of how careless they are about the side-effects.

    Bedrock and Java are entirely seperate and worked on by different teams. Also, try to hold back with the personal attacks please <3

    Also, I'm not on about friends, I'm on about public servers like this one. Like I mentioned, "don't say stupid shit" isn't an excuse, it basically saying "live every day in fear".

    Then, I guess, don't make satirical jokes that could be misconstrued as racist/something like that in public, like in real life?

    It'll work for a while, but again, we're talking about a community full of hackers, modders and coders. They can and will find a way eventually.

    Except ratelimiting isn't really a system that can be cheesed without hundreds of proxies... Even then, just block VPN IPs with Cloudflare or whatever Mojang use for their sessionservers. This problem also exists with every moderation system ever.

    They could've easily added a signature system without the reporting features along-side.

    Yes, they could've. But look at where we are now.

    It's because they've proven with bedrock edition it's a horrible idea. The "night" example is one of many. And I know it's not an oversight and a sign of laziness or overdone censorship since they had enough of a brain to not ban "dick" since it counts as a name. Their blanket-banning words without thought, and many people report false bans on bedrock edition. Not to mention their censorship extends to private realms and single player and is locked to always be on unless edited externally. The worst part is a bedrock ban also locks you out of single player. So if you mess up online, you're basically banned from your own private worlds too.

    And this is why I play Java instead of Bedrock, so I don't have to deal with the shit the Redmond team does.

    Imagine you get banned from public events for being a dickhead. That's warranted. But you come to your home to find out you've been evicted by the government (not your landlord or anything like that) for being rude at a unrelated location, are being forced to leave the country and can't ever come back. Sounds pretty unfair, right? Almost totalitarian?

    Singleplayer won't get banned on Java edition, which is what I'm talking about in this post.

    I was on about bots coded to mass-report users. Those are very likely to be used on the aforementioned out of context messages.

    How would a bot be able to differentiate between an ironic joke and unironic statement? Plus, if they make a large volume of false reports, or a large volume of reports within a certain timeframe, their accounts will almost certainly be punished.

    I doubt 5 messages are ever enough context when two friends are having banter, throwing in a few friendly insults, and some jackass comes in and reports them for "harassment and bullying" or some shit.

    Yeah, well, my school also had a "no banter" rule because it's difficult to distinguish between banter and actual bullying. It's much better to do it in private messages.

    Edited 2 times, last by Allink (July 28, 2022 at 10:49 PM).

  • Bedrock and Java are entirely seperate and worked on by different teams.

    They're owned by the same parent company at the end of the day, and their moderation team is more than likely directly controlled by Microsoft since it abide's by it's ToS. People forget that the teams are only separated for development of the game. Everything else? Microsoft.

    Then, I guess, don't make satirical jokes that could be misconstrued as racist/something like that in public, like in real life?

    Except ratelimiting isn't really a system that can be cheesed without hundreds of proxies... Even then, just block VPN IPs with Cloudflare or whatever Mojang use for their sessionservers. This problem also exists with every moderation system ever.

    Yeah, well, my school also had a "no banter" rule because it's difficult to distinguish between banter and actual bullying. It's much better to do it in private messages.

    All three of these fall under overly censoring or restricting people from just doing things they normally do. I doubt I'd get pelted with rocks if someone overheard me tell my best buddy to fuck off as a joke, MANY people use VPNs for safety or region locks more than they do for malicious intent, and banter isn't harmful at all, especially if a server has no rule against swearing (people are all at least 13+ on it) so in that case you're doing a 50/50 between if your gonna ban two friends having a go at it or a genuine bully situation.

    Singleplayer won't get banned on Java edition, which is what I'm talking about in this post.

    No one knows what the future holds. People already said that there was no way they'd do something like this in Java Edition because "They're different versions! And it never had it before!" but to everyones shock and surprise Microsoft did it. Remember, they own the game. They call the shots. If they did it there, they aren't behind doing it here.

    How would a bot be able to differentiate between an ironic joke and unironic statement? Plus, if they make a large volume of false reports, or a large volume of reports within a certain timeframe, their accounts will almost certainly be punished.

    1. Bots can be coded to recognise keywords and phrases. That can tie into recognising strings of messages that meet certain keyword-based criteria (e.g. several messages using words like "hate","fuck","kill". example: "Omg" "I hate you" "so much" "fuck you" "I'm going to kill". context in-game: Guy was killed by his friend using something like end crystals or respawn anchors. I know it's a specific situation, but there's many more likely situations you can run into.


    2. Mass reporting will lead to 1 of two things;

    • They either implement automation like Google does with YouTube or just hire people who don't care much at all due to how taxing it'll become to manually read and reply to these spams compared to genuine reports, which leads to more easier report spamming to get people banned.
    • They are more likely to ban YOU, the person reporting, if they think the issue is exaggerated. Which in turn will just make no one ever use the report function (besides bots), thus making it a useless feature they wasted resources on (technically a win for us, but still stupidity regardless).


    I doubt I'm gonna be able to convince you on why this is a horrible idea for Java Edition, so I'm gonna end this off by just saying one last thing;

    Regardless if you agree with the moderation or not, why the hell couldn't they ONLY ban people from only chat, signs, books and the likes instead of banning entirely from a whole subsection of gameplay?

    I don't need an answer to this because we all know it's because they don't care (and probably want you to buy the game again if you get banned).

    According to quantum mechanics, unless something is observed, there is an equal chance of it both being and not being there. Hence it is said to be in superposition, until observed. However we are somehow fully certain, despite never having directly observed, that you indeed have no bitches. Your bitchlessness has broken the rules quantum mechanics had established. Indeed an impressive feat.

  • Saying "fuck off" isn't bannable. Saying the n word, death threats or sexual harassment is. You don't "jokingly" sexually harass people with your best buddy, do you?

    All up for banning racial slurs and legitimate death threats/harassment. But if I said "I'm going to fuck you so hard" or "I'm going to murder you" to my friend and someone decided they wanted to use that to screw me over, they can (Refer to my statement about the bots using context to get you banned).

    And let's not forget people do say the n-word (non-hard R) just to refer to others casually. They'd suffer from this too. Like I said, the issue isn't I want to be free to harass people, it's that if I'm joking with someone I know it'll be taken out of context by third-parties.

    Just say it was just a prank

    Minecraft Sex Prank [GONE WRONG] [COPS CALLED] (They found my porn stash!)

    According to quantum mechanics, unless something is observed, there is an equal chance of it both being and not being there. Hence it is said to be in superposition, until observed. However we are somehow fully certain, despite never having directly observed, that you indeed have no bitches. Your bitchlessness has broken the rules quantum mechanics had established. Indeed an impressive feat.

  • They're owned by the same parent company at the end of the day, and their moderation team is more than likely directly controlled by Microsoft since it abide's by it's ToS. People forget that the teams are only separated for development of the game. Everything else? Microsoft.

    No shit, but there's no game-wide filter in Java...

    All three of these fall under overly censoring or restricting people from just doing things they normally do. I doubt I'd get pelted with rocks if someone overheard me tell my best buddy to fuck off as a joke, MANY people use VPNs for safety or region locks more than they do for malicious intent, and banter isn't harmful at all, especially if a server has no rule against swearing (people are all at least 13+ on it) so in that case you're doing a 50/50 between if your gonna ban two friends having a go at it or a genuine bully situation.

    If it's just "fuck off" then your report won't be accepted...

    1. Bots can be coded to recognise keywords and phrases. That can tie into recognising strings of messages that meet certain keyword-based criteria (e.g. several messages using words like "hate","fuck","kill". example: "Omg" "I hate you" "so much" "fuck you" "I'm going to kill". context in-game: Guy was killed by his friend using something like end crystals or respawn anchors. I know it's a specific situation, but there's many more likely situations you can run into.


    2. Mass reporting will lead to 1 of two things;

    They either implement automation like Google does with YouTube or just hire people who don't care much at all due to how taxing it'll become to manually read and reply to these spams compared to genuine reports, which leads to more easier report spamming to get people banned.
    They are more likely to ban YOU, the person reporting, if they think the issue is exaggerated. Which in turn will just make no one ever use the report function (besides bots), thus making it a useless feature they wasted resources on (technically a win for us, but still stupidity regardless).

    Yeah, of course they can, but there's more in a sentence than just words. There's grammar, spelling etc that can make it easier for a human to distinguish satirical content better than a machine could.

    If they think the issue is exaggerated, they won't ban you...

    Automation for chat reporting is going to lead to much more false positives/negatives. I don't think it's very realistic for Mojang to implement such a system...

    I doubt I'm gonna be able to convince you on why this is a horrible idea for Java Edition, so I'm gonna end this off by just saying one last thing;


    Regardless if you agree with the moderation or not, why the hell couldn't they ONLY ban people from only chat, signs, books and the likes instead of banning entirely from a whole subsection of gameplay?

    And I don't think I'll be able to convince you that Mojang/Microsoft haven't literally turned Minecraft into an Orwellian Nightmare. I honestly don't really care much about the chat reporting update, however I DO care about people spreading blatant misinformation (not you, just the points in the original post).

    Because banning a player from multiplayer is much easier and securer for Mojang than to ban players from all text inputs.

    Let me elaborate:

    Chat, signs, books, anvils etc are all controlled by the server. Mojang want game-wide bans. And Mojang controls the servers that allow you to connect to other servers in the first place (the authentication and session servers). If Mojang were to ban you from all text inputs, there's nothing stopping a server installing a mod/plugin that simply allows banned players to communicate through these methods, as if they weren't banned. However, Mojang has final say with the session/authentication servers and as such no servers can let you bypass your ban (except for using offline mode/a seperate authentication server).

    I don't need an answer to this because we all know it's because they don't care

    Your question was quite obviously not rhetorical. And no, I would say they do care about their game, since it's a source of profit...

    (and probably want you to buy the game again if you get banned)

    I would wager Mojang doesn't really want someone who's globally banned from their game to just buy another account and evade their ban.

    All up for banning racial slurs and legitimate death threats/harassment. But if I said "I'm going to fuck you so hard" or "I'm going to murder you" to my friend and someone decided they wanted to use that to screw me over, they can (Refer to my statement about the bots using context to get you banned).

    How exactly would Mojang, or any other party that moderates a game, differentiate between "I'm going to cut your corpse up into little pieces and ship it halfway across the world" between an actual imminent threat to someone's being?

    And let's not forget people do say the n-word (non-hard R) just to refer to others casually.

    Doesn't change the fact it's a slur...

    Like I said, the issue isn't I want to be free to harass people, it's that if I'm joking with someone I know it'll be taken out of context by third-parties.

    And yet Mojang send the context with every report...

  • I don't really care about the report system, I just wish Mojang did a better job at communicating with the player base. I'm sure a lot of hate came from the fact that they kind of dropped that in a snapshot, with only like one line talking about it and no prior warnings

    A fool's fool fools fools who foolishly accept the foolishness of a fool's fool.

  • I don't really care about the report system, I just wish Mojang did a better job at communicating with the player base. I'm sure a lot of hate came from the fact that they kind of dropped that in a snapshot, with only like one line talking about it and no prior warnings

    Yeah, I wholeheartedly agree with this statement.

    I don't think the system is necessary or was even wanted by anyone.

  • Context forgery no longer works (e.x. nodusclient/gaslight, video here); as the signature you send chat messages with is now based on the previous chat message. As such, if the real chat message signature is not based on the previous reported chat message (which in context forgery attacks would be malicious), your report will be thrown out and deemed as false. Here's a nice video with a visual explanation explaining why context forgery doesn't work anymore.

    External Content www.youtube.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.
    💀

  • Context forgery no longer works (e.x. nodusclient/gaslight, video here); as the signature you send chat messages with is now based on the previous chat message. As such, if the real chat message signature is not based on the previous reported chat message (which in context forgery attacks would be malicious), your report will be thrown out and deemed as false. Here's a nice video with a visual explanation explaining why context forgery doesn't work anymore.

    External Content www.youtube.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.
    💀

    >Entirely ignores the point being made and posts nodus video

    >Refuses to elaborate

    >Leaves

    ピバラ。