I don’t know why that would be but it only happens on certain, random comments. Sometimes it’s if I mention a brand or company or institution of some kind, without any criticism or anything, a mainstream company, other times just completely random stuff that’s not even referencing anything in particular. Shadow deletion I guess, but I have no idea why.
As other people have said, you have to show us your posting history. Not summarizations of it, not characterizations of it, we have to see the literal literal text that’s causing the issue. Otherwise we can’t help you
You can, you just don’t want to. What is stopping you from helping me? At least be honest. If you had a solution for an alternative text script that worked, you could just provide it. I did my best to give examples already
None of your examples were direct text quotes so we could see what was triggering the algorithm.
I suspect what you consider common sense, and normal, is being flagged as offensive and off topic in the discussion. Because you refused to provide direct quotes That’s the most reasonable deduction for your readers to take away.
But sure I’ll give you good advice:
Use a large language model, like chat GPT, and have it form a comment, that is both helpful, supportive, thoughtful, engaging. Based on the thesis you want to convey.
That is, no matter what I say, you wouldn’t consider alternate text scripts regardless and would just try to find other ways to avoid comments getting deleted, which I know won’t work because it’s just like every few comments gets deleted for no reason and they’re all different kinds of comments. They can be helpful, supportive, thoughtful and engaging, and still get removed.
So you have no intention of helping me find a text script that would work under any situation, even when random comments are getting deleted? I’m just trying to understand what you’re saying.
ChatGPT wouldn’t work because random comments get deleted all the time, there’s no reason to think that auto generated ones wouldn’t either. There seems to be no identifiable pattern to why certain comments get deleted.
I’m genuinely giving you the best help to get your text to stay online.
If chat GPT generating a comment, with the helpful parameters I’ve provided you, would get deleted, then your account is already flagged. Make a new account.
Okay thanks, I’m sorry if I misinterpreted your comments. It’s hard to tell in this thread with a lot of people being mean to me and not wanting to help and just trying to find out what they’ve already decided that I’m saying something problematic when I’m really not. I might confuse someone who is actually trying to help so I apologise if that’s the case.
I don’t understand why if my account is flagged (which I don’t know why it would be), I can still make some comments but not others, and if I change the wording or separate them into different comments, they can sometimes work. And for completely innocuous stuff.
But to be clear, you don’t know of an alternate text script that might work? That would be really helpful.
I appreciate if you’re genuinely trying to help. But it seems like you’re just trying focus on other things to avoid actually answering what I’m asking, which is about alternate scripts, which I know is the only solution to my particular problem.
You’re trying to do homoglyph attacks. That is the wrong way to solve the problem. If you’re trying to dodge the automation which is removing your comments by just changing the characters, then you’re just perpetuating the problem.
Either stop using YouTube, or start making comments that don’t get removed.
Anything else will just get your account flagged even faster and harder as a spammer. Substituting a Spanish n for Latin n is a massive red flag to the algorithm. You will find other creative uses of Unicode making your account spam weight very high.
Interesting, so you’re saying if I use alternate homoglyphs as you describe them, the system will detect those and flag my account or something? Seems unfair of the algorithm to do that considering I’m already unable to make normal comments in a lot of cases, but of course the whole problem is the algorithm not working properly.
But I’m just wondering if it’s possible it wouldn’t detect those alternate characters and I could comment normally (using homoglyphs), since I’m already unable to, it might be worth a try. I guess the definition of spam can vary, but it’s not like I would be posting a lot of comments, just a regular amount, or even a low amount.
I don’t know what alternate text format works because the ones I’ve tried haven’t.
So for example, one that visually looks like the Latin/English alphabet, but which wouldn’t be detected as English characters because it’s in a different enough format.
Please see my other comment. To anybody who’s reading this conversation this is called homoglyph attack. If you use this on public forums, the algorithms will flag your account as a spammer, correctly.
I understand you’re saying the bots/algorithm might still delete it or flag me for it, but it would still be helpful to try it anyway since that’s already happening for me so it can’t really make it any worse than it already is. If you happen to know some of these “homoglyph” alternatives, that may be really helpful.
I don’t know why that would be but it only happens on certain, random comments. Sometimes it’s if I mention a brand or company or institution of some kind, without any criticism or anything, a mainstream company, other times just completely random stuff that’s not even referencing anything in particular. Shadow deletion I guess, but I have no idea why.
As other people have said, you have to show us your posting history. Not summarizations of it, not characterizations of it, we have to see the literal literal text that’s causing the issue. Otherwise we can’t help you
You can, you just don’t want to. What is stopping you from helping me? At least be honest. If you had a solution for an alternative text script that worked, you could just provide it. I did my best to give examples already
None of your examples were direct text quotes so we could see what was triggering the algorithm.
I suspect what you consider common sense, and normal, is being flagged as offensive and off topic in the discussion. Because you refused to provide direct quotes That’s the most reasonable deduction for your readers to take away.
But sure I’ll give you good advice:
Use a large language model, like chat GPT, and have it form a comment, that is both helpful, supportive, thoughtful, engaging. Based on the thesis you want to convey.
That is, no matter what I say, you wouldn’t consider alternate text scripts regardless and would just try to find other ways to avoid comments getting deleted, which I know won’t work because it’s just like every few comments gets deleted for no reason and they’re all different kinds of comments. They can be helpful, supportive, thoughtful and engaging, and still get removed.
So you have no intention of helping me find a text script that would work under any situation, even when random comments are getting deleted? I’m just trying to understand what you’re saying.
ChatGPT wouldn’t work because random comments get deleted all the time, there’s no reason to think that auto generated ones wouldn’t either. There seems to be no identifiable pattern to why certain comments get deleted.
I’m genuinely giving you the best help to get your text to stay online.
If chat GPT generating a comment, with the helpful parameters I’ve provided you, would get deleted, then your account is already flagged. Make a new account.
Okay thanks, I’m sorry if I misinterpreted your comments. It’s hard to tell in this thread with a lot of people being mean to me and not wanting to help and just trying to find out what they’ve already decided that I’m saying something problematic when I’m really not. I might confuse someone who is actually trying to help so I apologise if that’s the case.
I don’t understand why if my account is flagged (which I don’t know why it would be), I can still make some comments but not others, and if I change the wording or separate them into different comments, they can sometimes work. And for completely innocuous stuff.
But to be clear, you don’t know of an alternate text script that might work? That would be really helpful.
Chatgpt can rewrite your comment, generating your alternative text
I appreciate if you’re genuinely trying to help. But it seems like you’re just trying focus on other things to avoid actually answering what I’m asking, which is about alternate scripts, which I know is the only solution to my particular problem.
Chat GPT is generating alternative scripts for you.
If you already know the solution, just implement your idea. If you ask for help you’re going to get a diverse set of ideas to help you.
Sorry, I meant an alternative text format. I don’t think ChatGPT is capable of doing those. Like, variant text characters.
You’re trying to do homoglyph attacks. That is the wrong way to solve the problem. If you’re trying to dodge the automation which is removing your comments by just changing the characters, then you’re just perpetuating the problem.
Either stop using YouTube, or start making comments that don’t get removed.
Anything else will just get your account flagged even faster and harder as a spammer. Substituting a Spanish n for Latin n is a massive red flag to the algorithm. You will find other creative uses of Unicode making your account spam weight very high.
Interesting, so you’re saying if I use alternate homoglyphs as you describe them, the system will detect those and flag my account or something? Seems unfair of the algorithm to do that considering I’m already unable to make normal comments in a lot of cases, but of course the whole problem is the algorithm not working properly.
But I’m just wondering if it’s possible it wouldn’t detect those alternate characters and I could comment normally (using homoglyphs), since I’m already unable to, it might be worth a try. I guess the definition of spam can vary, but it’s not like I would be posting a lot of comments, just a regular amount, or even a low amount.
I don’t know what alternate text format works because the ones I’ve tried haven’t.
So for example, one that visually looks like the Latin/English alphabet, but which wouldn’t be detected as English characters because it’s in a different enough format.
Please see my other comment. To anybody who’s reading this conversation this is called homoglyph attack. If you use this on public forums, the algorithms will flag your account as a spammer, correctly.
But in this case I wouldn’t be spamming if I was just posting innocuous comments that happened to be formed with homoglyphs, right?
I understand you’re saying the bots/algorithm might still delete it or flag me for it, but it would still be helpful to try it anyway since that’s already happening for me so it can’t really make it any worse than it already is. If you happen to know some of these “homoglyph” alternatives, that may be really helpful.