I disagree, you don't need a "rock-solid strategy for moderating" "user-supplied content", you just need a user base that is by and large not inciting violence, particularly against government officials.

If I run some hobbyist forum where say 1 in 1000 users is posting hate speech or calling for violence, as long as I have less-than-rock-solid moderation that can clean up some of their posts and ban the worst offenders based on user reports, I wouldn't be too worried. It's also in my interest to keep the community healthy.

Now, if 20% or 50% of my users are posting hate speech and calling for violence, yeah I would be worried, but I would be less worried about moderation, and more worried about why my hobby attracts extremists. Might call for some self reflection.

It sounds like you have done your risk analysis and settled on an balance of risk and moderation effort that is acceptable to you. That sounds fair to me.

> Now, if 20% or 50% of my users are posting hate speech and calling for violence, yeah I would be worried, but I would be less worried about moderation, and more worried about why my hobby attracts extremists. Might call for some self reflection.

LOL! I think this is actually the right take! Definitely also applies to your social network...

Just a thought....

Maybe it doesn't attract extremists. Maybe extremists just happened to randomly pick your site as the latest site to use for communication.

Just theorizing here, but assume that instead of hosting their own content, web sites with comments are hijacked to host extremists propaganda/plans/violent event, with a pointer to the next web site to use (not IF) but when the current one gets shut down.

The only way to avoid this is to moderate the comments before they are allowed to be displayed. And I'm betting there are a thousand+ web sites that are run by amateurs where this is the last thing on their minds.

Hopefully the fact that scaling and searching would be impossible makes this a non-issue.

Sure but AWS isn’t run by robots (yet!). It’s run by people who understand context and intent. I also very seriously doubt that Parler was turned off without any notice. If a company was getting abused like that, I think AWS would try to help them.

winning argument right here

No, the lesson of Parler is not that you need to have

> ... a rock-solid strategy for moderating ...

The lesson is you need to have a rock-solid `policy` that content which advocates violence is impermissible.

Unfortunately I cannot reference Parlor's actual Community Guidelines (https://legal.parler.com/documents/guidelines.pdf) since the site is down for some reason. :)

However, based on what is described here (https://theconversation.com/parler-what-you-need-to-know-abo...) `there are policies against “fighting words” and “threats of harm”. This includes “a threat of or advocating for violation against an individual or group”.`

Based on recent examples of Parlor posts it seems like the Parlor moderators were not adequately enforcing this policy.

Pretty sure Parlor had no official moderators.

Just the community self-moderating themselves.

"a rock-solid strategy for moderating it"

You mean, like Facebook and Twitter?