That’s great, thank you for this.
On another note about Chat, I know many would like this and it was mentioned above, are there plans to make it possible to know exactly what caused your text to be blocked? I sometimes try to respond to someone and the response is lengthy but absolutely nothing offensive (in language or content) or anything else against the TOS. I try to rephrase it and still get the same block.
Whatever your system is, it got many false flags for perfectly innocent messages from me.
One example from the past:
I just checked these blocked statements against our system, and it doesn’t appear to have any issue with them. I also don’t see these messages in your recent history. Do you recall when this was?
That other thread is from March 3rd, so it’s quite a while ago.
It’s possible we may have already seen it, and resolved the cause. We do try to keep a close eye on feedback regarding the rules, and regularly make adjustments to the system. If you encounter anything getting blocked that you don’t understand why, please do continue to report it to us so we can take a look at it.
We absolutely want this, but it’s been a lot more complicated than it seems on the surface. In order to do this, there needs to be additional back and forth communication within the chat system, and those small calls really add up quite fast. This is something we are still looking at methods of doing, and we have a bunch of other backend improvements well underway. Moderation, and chat services don’t often make for popular conversation topics, but it’s something we have internally been quite active about.
If I run into this issue again (I haven’t used game chat for months, I think), I try to remember to report this.
Perhaps you could integrate some one-click report feature in the client for this? Coming to the forum and creating a report while playing is quite a distraction, after all.
Example:
This message was automatically blocked due to a violation of our code of conduct. If you think that this is a false flag, right-click this message and select report from the context menu.
I fuc*in hate the profanity filter with a passion!
First of all, it makes the chat SOO SLOW, because all messages apparently have to first go to a 3rd party api and be checked, and after 2-3 seconds you either see your message, or a system message telling you it got blocked.
Then, even when I am whispering to my friend, we are talking in our native language, and god dammit, even then some messages get filtered??? Sometimes I am trying to send him the same message like five times, and every time I have to guess which is the “bad” word, so I can replace it
Unfortunately harassment in DM’s is one of the most common, and worst places for harassment. We’ve had this discussion around “conversations between friends”, but that’s where individuals typically try to go for some of the most vitriolic comments. In offering online communication, we have an obligation to do what we can to try to protect our players from online harassment and bullying, and DM’s are the worst place for this.
The chat system should not be this slow, and it sounds like this may actually be a network communication issue rather than the filter. Regarding blocked messages, the chat system is based on the language you’re set to. So if you’re set to English in the client, it’s going to read the message as English. This is one of the limitations we’re currently working on.
I apologize to EHG staff if this truly wasnt intentional and due to 3rd party “political censorship” . It is even possible you werent aware… So for that I apologize. I based my post off of the frustration i have experienced with Every other chat filter in game i have had problems with. It is BEYOND Frustrating when you are in a party, type a long message to your friend (not involving politics or Trump) and having it blocked, forcing you to type it again reworded only to find it is blocked again, its happened so many times it is instant shock somethings get hung up on chat filter and these arent even vulgar or inappropriate things… The system just blocks…
I am aware some people might use the word Trump in global to stirr up stuff but we always have the option to block those people.
My friend i was in a party with last night asked if i was going to watch the debates, i began to type “i dont know why Trump even accepted a debate with her” and it was insta-blocked… we began typing Biden, Kamala , Harris… those werent blocked… so it was extra frustration on top of the ALREADY frustrating chat filter problems…
Thank you EHG in advance for looking into these 3rd party caused problems… I apologize for the rudeness of my post, it was out of extreme frustration
Want to thank you for the clarification on these systems and communicating your positions!
You don’t need to. Use the in-game report tool. It will collect all the data, including, I’m assuming, the recent messages.
I would perhaps suggest to make online explicitly stated that it’s 18+ and then treat adults as adults meaning not censoring them at all. I don’t mind an opt out feature but it should be off by default in my eyes.
Regarding harassment: I take that word very seriously and will not tolerate an reality where being rude to someone or offending someone in a game who you don’t know and will never see again is labeled as that- if someone is being nasty…. **Block them **
Only thing that should be done is combating goldsellers and bots. Leave adults to be adults or kick them from the game until they are 18+. Not here to carebear for people online who will never know each other.
Your system allows blocking the messages of users.
If harassment happens that’s the option to go with for the user.
You’re not responsible for managing the emotional state of someone who can’t take a single sentence without breaking down, that’s was never your job and shouldn’t be the case currently either.
The notion of implementing a system which limits good actors to prevent bad actors is fine.
The notion of implementing a system which limits good actors to reduce the effort needed to prevent bad actors in a absolute minimal degree that’s already handled is not fine.
Your system is overzealous and out of place, it’s as simple as that sadly.
And if someone goes so far as to actively break the local law through your video games in some measure then inform the respective local authorities to handle it and remove the ability to chat from that specific account, that’s your only task to do there, you don’t have many obligations… going beyond those obligations is not helpful, it’s harmful.
I don’t understand you.
You are a valiant defender of user rights. Your points often include psychological perspectives that boil down to hurt expectations and how this makes the user feel bad. The reasonable, adult perspective has no relevance to you, but how people feel has.
Now we are talking about something where feelings get trampled on, where serious emotional damage can be done. People will feel insulted, harassed, discriminated, and marginalized by what was written. Blocking the offender might be the reasonable approach after the fact, but the damage is already done. The anxiety that may come now not knowing what the person might write about you in public chat might skyrocket.
What is now your point? “Get your shit together, behave like an adult, block them and move on”?
This is not only about law. This is about creating an environment that is as free of this stuff as possible. EHG might feel obliged to create a safe and friendly environment for all players. Informing local authorities will barely ever lead to anything if we are realistic.
I don’t know how good the system is at filtering out offensive/unwanted stuff, as I usually self-censor myself enough. Therefore, I only encountered false positives with my messages, I think. It were quite a few, but it was not so bad that I have considered the chat unusable.
EHG uses reported false positives to improve the system, that’s fine. But reporting false positives should be as easy as possible. @DJSamhein, btw, you are right, I forgot about the in-game report tool, but this tool is not that comfortable to use. Takes so much time that I would still consider it a barrier from the perspective of usability.
Those false positives are an inconvenience, but I am an adult. I can live with it, rephrase my perfectly innocent message, and move on.
I actually always had a positive experience using it. Just hit esc, select the tool, usually you don’t need to change the topic so you just write what you want to report and hit send and then close. Seems pretty straight-forward to me.
What would you change about it?
Chatmessage gets blocked and includes the info how to report a false positive. “Right-click THIS message and select Report from the context menu” or something like that.
No ‘ESC’, not overlaying windows, not selecting any category, writing a headline, what was wrong, etc.
Message is filtered, though, you don’t see it. And you don’t really need to type the message. The report tool should already send the message that was blocked, so just type that your message was wrongly blocked. I really had no problems using it in the past. Might just be a personal thing.
That being said, having a context menu to bug report these messages would be a nice QoL.
But you see the message that reports that your message was blocked. If you right-click that and select report, it should report the initial message as a possible false-flag. All those false-flag reports will be conveniently categorized for the devs. They don’t have to manually sift through general tickets and sort it into the false-flags sections.
A mod or dev goes through this collection at some point, accepting or rejecting them as false-positives. If my guess is right, the filter is based on a language model that tries to build context, so false and correct positives could be used as further training data.
First of all, that’s a misrepresentation of my stances.
There’s a difference between functional frustration (which for example is overcoming an obstacle) and non-functional frustration.
I’m generally advocating for situation to be adjusted where there’s non-functional frustration, those can’t always be avoided but need to be as much as possible.
Which leads to this:
You can always feel insulted, harassed, discriminated and marginalized.
This is the situation which can exist in any dispute and any communication.
Offending people is the cornerstone of speech, you’re by design trampling over others feelings the second you have a different opinion to someone, by design. That’s a given. Removing that aspect is removing all possibility for proper communication.
As an example: ‘Idiot’ is a word which is generally picked up by people as being defamatory, saying that someone lacks mental capacity.
Now we can showcase 2 examples of how it could be used and what the difference of the word is: ‘You are an idiot’ is in 90% talking down to someone. The leftover 10% are a factual statement if someone has a actual low mental capacity, that’s a descriptive term which has been coined ‘offensive’ by design despite actual existing real life examples of it existing.
A second one would be ‘You act like an idiot’ which is another version of saying ‘You’re out of line, rethink what you’re doing’.
The first is in - the majority of cases - not acceptable, the second though is in communication, given a former statement on the reasoning is given.
It’s one of the fundamental issues nearly world-wide with communication nowadays, overzealousness. The value of comments is derived from their quality of writing rather then the content inside the writing. Awfully worded sentences and arguments can have important insights while well described arguments can be utter bollocks.
Likely there’ll be more weight given to the second one rather then the first though when put side by side, which is a communication fallacy many fall under.
Yes, it is.
Taking away the ability to act for people though is nonsensical.
You have to accept that if you’re in any form of social environment you are always at risk to have your feelings hurt in some way.
Removing this option is the same as forbidding knives world wide because they theoretically could be used to kill someone. So lets have everyone cook without knives from now on since we can’t afford to risk that.
Obviously not a good approach.
It’s the same with communication. You need to give people the theoretical option to use any sort of tool in a proper manner and simply have consequences for those which use it in a malevolent way.
It’s called toxic positivity what you’re describing there.
That’s highly detrimental for people inside this environment, getting used to this causes the loss of ability to sustain yourself under criticism of any kind in the worst-case scenario, which fairly often happens.
That’s actually how a good portion of those people start to exist which break down under miniscule scrutiny, starting to act far out of line right away and escalating several stages as once for - the vast majority of people - miniscule details. The lack of ability to know how to accordingly react to such a situation causes it, which is a learning aspect, and learning this can only happen when you’re confronted with it and get the results of your actions reliably.
Mind you, what I’m talking about is the pure side-effect of filtering the possibilities of using words, obviously defamation or harassment itself is not something which is to be allowed. Trying to cull it at the root though causes more harm then good. You can only have a reactive stance to it, a proactive stance to inhibiting it is one of the worst things that can be done.
Depending on the false positive or the need to describe something at times can lead to the inability to communicate the meaning accordingly. Hence this is not a inconvenience but a outright limitation of communication.