But when they do eventually figure it out, they end up hoping from one newly created account to another. Unless using new accounts are prohibitively expensive, and they usually aren't, then you'll end up just replacing one problem with another.
The correct way to moderate is to be civil and treat "low quality" users as human beings. Reserve inhuman bans for inhuman users (bots).
What do you do when those 'low quality' users are sending rape threats to specific users? Or stalking them in this scenario? The police aren't going to do anything because it's barely an actionable threat and hard to trace.
The entire reason shadowbans exist and have existed for decades is because there's an inherently unsolvable problem with certain users that extremely fixate on certain users or communities. If you think they can be treated by being civil then I don't think you've been a moderator for any extensive period.
Ban evasion was a thing long before shadow bans. I've seen persistent users disrupt IRC channels for years, maybe decades. You can't reason with or befriend somebody who holds a grudge that long. Moderators will just have to deal with the problem repeatedly. Shadow bans at least minimize recurrence by taking away the immediate feedback.
IRC generally has no barriers to account registration while Twitter generally requires a phone number. In the former bans are meaningless, in the latter they pose in an insurmountable hurdle to all but organized and funded campaigns. The real problem with shadowbans is "us" - humans. Shadowbans are a great tool for helping to minimize the impact of things like literal spammers and scammers, but it's also a tool that's great for minimizing political dissent, creating false impressions of the popularity (or not) of various concepts, or generally grossly manipulating and distorting public discourse.
As public online discourse becomes a more engrained part of normal society, shadowbans may even end up being an action that, in and of itself, end up being regulated and controlled. Shadowbans can help create a better and freer society for everybody, but they can also be used to create a dystopia. In some ways it's akin to a real world weapon. A firearm can be used to uphold and protected great values, but it can also be used to impose horrific ones.
> IRC generally has no barriers to account registration while Twitter generally requires a phone number. in the latter they pose in an insurmountable hurdle to all but organized and funded campaigns.
Since when is getting a new phone number that receive SMS an insurmountable expensive? There are literally dozens, perhaps hundreds of services selling this. Example [1] https://www.textmagic.com/sms-pricing/ seems to cost me 7c per text received with no account fees, and I only need it to receive a sign up message.
That is far from "organized and funded". People paid $8 to make troll accounts on Twitter when it was possible.
You are correct that burning a SIM card is not that expensive.
However, your example of a text capable VOIP number is not a good one. FAANGs require a real mobile number and these texting services (twilio, for instance) will not work.
All of this to say:
It’s a bit more expensive than you think. If you really care about succeeding, you’re going to need to burn a SIM card.
The correct way to moderate is to be civil and treat "low quality" users as human beings. Reserve inhuman bans for inhuman users (bots).