Determine shadow ban status

[Noah670]

A clear metric should be given for every user to determine this status for greater transparency.

[randiaz95]

Theres actually a offensiveUserCounter.increment() that I found and tried to delete before in the PR here: #345

I don't know if this is the only place where something like that could happen, but it can be read from our settings page if we have a get endpoint for this specific user.

[rambda]

The only one that faithfully detected my search ban (not shadow ban) is https://taishin-miyamoto.com/ShadowBan/
My account is https://twitter.com/SmolderinCorpse
As you can see, I rarely talks in the past 10 years, I'm not toxic, I'm not a bot and I tweet sfw.

I think there should also be some transparency on the reasons for why users getting search/shadow banned. At least give users some instructions on how to get rid of the ban and how to regulate their own behavior, otherwise it lost the point of banning real people.

[leafstrat]

The only one that faithfully detected my search ban (not shadow ban) is https://taishin-miyamoto.com/ShadowBan/ My account is https://twitter.com/SmolderinCorpse As you can see, I rarely talks in the past 10 years, I'm not toxic, I'm not a bot and I tweet sfw.

I think there should also be some transparency on the reasons for why users getting search/shadow banned. At least give users some instructions on how to get rid of the ban and how to regulate their own behavior, otherwise it lost the point of banning real people.

I think you've misinterpreted the results from that website, understandably. It is confusing how they display it. I believe the green results you are seeing mean that you are 'safe' from that type of ban, not that it is effecting you.
I imagine the results would be red if it detected a ban was in place. I tried multiple accounts and they're all green.

[rambda]

The only one that faithfully detected my search ban (not shadow ban) is https://taishin-miyamoto.com/ShadowBan/ My account is https://twitter.com/SmolderinCorpse As you can see, I rarely talks in the past 10 years, I'm not toxic, I'm not a bot and I tweet sfw.
I think there should also be some transparency on the reasons for why users getting search/shadow banned. At least give users some instructions on how to get rid of the ban and how to regulate their own behavior, otherwise it lost the point of banning real people.

I think you've misinterpreted the results from that website, understandably. It is confusing how they display it. I believe the green results you are seeing mean that you are 'safe' from that type of ban, not that it is effecting you. I imagine the results would be red if it detected a ban was in place. I tried multiple accounts and they're all green.

Yesterday I was search banned. now I was unbanned.
The way I check search ban is, open firefox in private browsing and search "Fibonacci time signature", which is my latest tweet. I'm 100% sure I couldn't find my tweet yesterday.

Even so, I still have another account (I won't say the username) which got search banned that only posts selfies (sfw).

I've heard it's either because I changed my username recently or because I followed too many nsfw accounts, including models and artists (really, that's a reason?), but since there's no official advice , I have no way of knowing how to get rid of the search ban.

Edit: But since the algorithm unbanned my main account immediately, it means that it is reliable to some extent. Maybe the reason I got banned at the time was because I was using a VPN from another location or something.

[rambda]

https://taishin-miyamoto.com/ShadowBan/ does not even work for my account, LOL. Just says error, but your account works. It is all clear now, yeah.

But again. The actual secret notes on Guano is what makes the difference, like in viral content.

I thought it's pretty clear twitter admin can do that... I though everyone knew it.

I live in China. For me, it's hard to believe any social platform don't do that. I know what kind of place China is, don't remind me, pity me insted. But I have to also remind people, US is not an utopia...

If twitter shadow already bans people using the algorithm, why don't they do it manually too? Of course they do it.

[darkdevildeath]

@rambda

Happens all the time. There are several records.

I usually warn anyone who shares intimate photos with these tools. Even with P2P encryption it is possible for a user with elevated privileges to access the photos. The only thing that would really stop it are the supposed moral values ​​of whoever has control over it.
https://www.yahoo.com/entertainment/book-claims-facebook-fired-52-012010053.html

[rambda]

@rambda

Happens all the time. There are several records.

I usually warn anyone who shares intimate photos with these tools. Even with P2P encryption it is possible for a user with elevated privileges to access the photos. The only thing that would really stop it are the supposed moral values ​​of whoever has control over it. https://www.yahoo.com/entertainment/book-claims-facebook-fired-52-012010053.html

I know. This is no different from Weibo, QQ, WeChat, etc. Their admins also spy.
No one can supervise these twitter admins other than themselves. So it's impossible to stop them from prying into users privacy.
I use Telegram when I really need relatively higher privacy, relatively.

[rambda]

I said that I don't want to expose my selfie account, because apparently no one wants to do that.

Check @cawldha Success.

https://taishin-miyamoto.com/ShadowBan/

But I really wonder why?

Twitter rules:
Violent Speech, Violent & Hateful Entities, Child Sexual Exploitation, Abuse/Harassment, Hateful conduct, Perpetrators of Violent Attacks, Suicide, Sensitive media, Illegal or Certain Regulated Goods or Services, Private Information, Non-Consensual Nudity, Account Compromise, Platform Manipulation and Spam, Civic Integrity, Misleading and Deceptive Identities, Synthetic and Manipulated Media, Copyright and Trademark, Third-party advertising

Translate my tweets and read them, which rule did i break?

I still only had a search ban yesterday, and now I got shadow ban. And my only behavior today is to repost some life advice and reply some good tweeters, you can see it from my homepage.

POLITICS ALERT:

If you ask me what are these reposts and replies about, I can only say they are just some trans things, and I implicitly criticized CCP (instead of China) (with good will for my homeland). 🤔

I've noticed a recent controversy between trans groups and the twitter algorithm...but I was not interested. And I am not a US citizen and trans rights does not exist in my country. Fighting for it on twitter has little good to me. Most importantly, all Chinese trans tweeters that I know are not banned. So it was me who personally offended the algorithm.

As for CCP, I'm pretty sure twitter doesn't care... OR that's exactly what they care about? Maybe the algorithm can't read Chinese and thought I support CCP, so it banned me?

I really don't want to talk about minority rights and politics here, because it's GITHUB, but If politics is so important to twitter's algorithm, I have no choice.

---

Anyway, I think I gave a perfect example that, an ordinary user, without violating Twitter's rules at all, may be shadow banned for unknown reasons.

I repeat my advice again:
I think there should also be some transparency on the reasons for why users getting search/shadow banned. At least give users some instructions on how to get rid of the ban and how to regulate their own behavior, otherwise it lost the point of banning real people.

---

Edit:
OK I got unbanned in one day, but still search banned. why torturing me?
Edit 2:
No, every comment of mine was folded by twitter. Why? WHY?

[TolstoyDotCom]

The best way to determine if your replies are being censored is to create a test account using a browser (with all cookies cleared), preferably using a different IP. Then, load the pages you replied to. You'll probably be surprised: Twitter lies to millions of people each day via "shadow ordering".

[PaulNewton]

@Noah670 A clear metric should be given for every user to determine this status for greater transparency.

In what way would a first-party exposing such a metric NOT completely undermine the purpose of shadow banning for malicious accounts. Third party methods that try detect shadow-banning off a platform are irrelevant to the core of this type of problem.
Shadow banning is a moderation tool businesses can use to coral the efforts of potentially malicious accounts by making them continue to operate as if they are having an effect.
It is not a paradox that they are allowed to continue to use the platforms resources versus a complete ban to save resources, it's a tradeoff.
If a malicious account knows it's shadow banned it will make more new accounts using even more resources each time a full-ban happens using more service resources than if that account had continued being shadow-banned account ignorant it's wasting it's time.

[rambda]

@PaulNewton

A malicious account, 100% knows it's shadow banned, because it can simply request these shadow ban detection site.

Sadly, however, an ordinary account, which got mistakenly shadow banned, most likely does not know even know shadow banning exists.

Because, simply, even most people in China don't know shadow banning is a thing. It's so hypocritical, that it's beyond an average person's imagination.

[PaulNewton]

@rambda already preempted this, read:

Third party methods that try detect shadow-banning off a platform are irrelevant to the core of this type of problem.

The barrier of difficulty does not need to be removed because of an assumption of omniscient technical sophistication is given to potential attackers, not every malicious user is a coordinated troll farm with it's own developers.
That's how security works, you don't get rid of doors because some thieves have lockpicks, nor do you start handing out keys either.

[TolstoyDotCom]

Twitter does show some messages when they (almost always incorrectly) think you're misbehaving.

Several years ago I started a new account and put a few handles in the first tweet. They kept me from sending the tweet due to the # of handles.

More recently, I replied to someone with a link that had "idiotic" in it. That was a reference to a contest, not a person. Twitter showed me their "Most tweeters don't tweet things like this" screen, forcing me to tell them they're wrong and tweet it verbatim.

With shadowbanning and shadow ordering, Twitter is lying to millions of users each day. That's no way to run a company. Each and every day, Twitter treats millions of legitimate users as if they were spammers.

If Twitter were technically competent and opposed censorship, instead of doing that they'd evaluate the content of a user's tweets. I'm sure a technically competent company could identify outright commercial spam in almost all cases while not heavily impacting real users. Instead, Twitter bases that on their highly flawed algorithms and farms that out to unreliable parties. For an example of the latter, see DrPanMD. He's extremely pro-censorship and in one case he hid 99 out of the 100 replies to one of his tweets. Criticize him and he's likely to hide your reply or block you. Instead of holding that against him, Twitter holds his blocks against his critics.

Not only is Twitter technically incompetent, they have zero interest in open debate and protecting speech.