I'm still waiting for a new platform to take the world by storm: The freedom of Gab, with sane content filtering options to make it look as tame as reddit so by default users don't see all the fringe crap, but lets them if they really want, right to the legal edge.
People are not fringe will tend to abandon sites that embrace people who are, even tacitly. Sites that create havens for them will naturally tend to become sites that consist almost solely of them.
That's been the status quo up until now, but it doesn't have to be. We're moving towards orwellian style censorship at a breakneck pace with SV leading the charge.
Any violent thug can go spread their message in the town streets as long as they're not causing a disturbance. It doesn't happen very much, because it requires effort and it's sure to attract opposition. That opposition is necessary, like anti-bodies attacking an invading bacteria. By shutting out the harmful bacteria where anti-bodies can't reach them, they'll do nothing but grow and fester.
I think we have pretty good evidence in recent years that this is not how things really work. Those filters and algorithmic steering just provides a way to keep people in their bubble. What you’re describing is basically what YouTube was: allow anything legal but use sophisticated software to show people what they’ll respond to positively. The radicalization this led to is well documented at this point.
I’m not really in favor of what CloudFlare’s done here, but it’s really a fact of basic statistics that exposing the mainstream to well-made propaganda is far more dangerous than exposing radicals to the mainstream.
Imagine 1% of non-radicalized people exposed to radicalism become dangerous and 50% of radicalized people are de-radicalized by exposure to the mainstream (laughably optimistic). If you start with 100 million non-radicals and 500 thousand radicals, you have a 125% increase in radicalization.
This is how we now have so many flat earthers. When you have massive reach, it’s easy to grow your numbers.
> That's been the status quo up until now, but it doesn't have to be. We're moving towards orwellian style censorship at a breakneck pace with SV leading the charge.
Very few people actually want such a site arrangement. Believe it or not, I have no interest in interfacing with people who want to kill me unless they are open to honest and rational debate, which few are. One of the reasons I enjoy Hacker News is because I know won't have to deal with the type of people that will frequent the fringe elements of the style of site you are advocating for.
Also, I'd hardly call a business led moratorium on speech advocating terrorism and genocide "moving towards Orwellian style censorship". Absolutely nothing is stopping these individuals from standing up their own server and hosting this content themselves. They aren't banned by an ISP. Moreover, business owners should be allowed to express their opinions too, including who to do business with.
> with sane content filtering options to make it look as tame as reddit
People are really, really good at circumventing automated filters, while non automated filtering means subjecting people to horrific content for 8 hours a day.
I really want to be on the side of the broadest possible interpretation of free speech, but my experience leads me in a different direction. It's been my experience that corralling the fringe doesn't work because the people in that fringe are intent on spreading their message/content as far and wide as possible. Even against the wishes of the general populace of a given site like Reddit. That behaviour seems to spring from the same attention-seeking place as that of griefers like Goonswarm in EVE, whose stated goal was at one point to ruin as many people's fun in that game as possible.
So there's perhaps two parts to the freedom of speech argument space: One is that there are people who want to create/share/consume content that the general populace doesn't want to see. Then the second is that there are people who aren't acting in good faith, whose goal seems to be to ruin as much as possible.
The first group there will probably keep to themselves, manage their communities and cooperate with governmental authorities to remove illegal content or activities. Of course that doesn't take into account when those groups are acting against those authorities while still having the moral support of most of the population. For instance campaigning for marijuana legalization, which most people would say should be protected speech.
But then there's the second group, who hide within the first often enough, or who falsely act like the first because they want to destroy things because they seem to have no emotional investment in society. (in my opinion)
The second group thrives because no one wants to commit the manpower to track and manage them. I don't know how to fix that problem, I don't even know if it's possible. Heck there will be many who will argue it's not necessary at all.
1. Make users tag their content if it touches on certain topics and use harsh punishment if they fail to. If it's fringe content for example, make them tag it as such so default users will not see it unless they really want to. If someone doesn't know their ramblings are racist, they should be banned on incompetence grounds and not due to censorship.
2. Rethink real time posting. It's nice, but a 1 minute delay would be very beneficial. You could even conscript volunteer users, perhaps give them "free premium" accounts that act as sentinels for all new content so they can see all new content in real time, like a mod of sorts, and can flag inappropriately tagged content as such so it's kicked back to the poster as "You forgot to add a violence tag".
Essentially the only people who should be de-platformed are those who are either banned for legal reasons or because they're simply too incompetent or unwilling to tag their content with appropriate fringe tags.
Sane defaults. Option to not see filtered content, but see a note saying "these many comments were hidden" (and a way to see them). Both AI-based filtering (e.g. for porn & gore), and user-moderated lists (e.g. a left-moderated list of "hateful neo-Nazis" and a right-moderated list of "crazy leftists").
