3 ms·
It's impossible to believe that Facebook, Twitter, iPhones, and Google weren't in major ways used by the perpetuators and organizers of the assault on the capit
by 99_00 6y ago
It's impossible to believe that Facebook, Twitter, iPhones, and Google weren't in major ways used by the perpetuators and organizers of the assault on the capital to facilitate their criminal conspiracy and actions.
Selective enforcement of rules is being used to against political positions and, let's face it, competing platforms.
>Although parlor was only used for hate,
I don't think that is in evidence.
- temp8964 6y agoAnytime you look at the trending hashtags on Twitter they are always divisive and hatred. Even when unify was trending, the top twitters and top comments are the ones against unification, like "we don't unify with nazis".
- tonfreed 6y agoWhat's considered a Nazi just keeps gradually expanding. Twitter seems at that point where any white male is a Nazi, and any woman that loses weight is enabling the white males so are therefore Nazis as well. I'm curious to see what happens when Biden refuses to meet with any of the BLM radicals and the more senior Democratic members of congress reign in the young ones like AOC, these people have outlived their usefulness now that Trump's out 3and I don't think they're going to like what comes next.
- BoorishBears 6y agoSite A has a TOS that bans hate speech Site B is just like Site A, but with less reach. However, their TOS allows hate speech as long as it's not tortious So on Site A a comment spamming nig**r will get you banned, and on Site B it is a-ok as long as it doesn't mention anyone's name. - Site B should be expected to have more hate speech. Additionally even if Site A gets hate speech, it's providers are more likely to give it leeway since it's not by design that hate speech appeared. Meanwhile Site B, by the very virtue of it's design becomes a lightning rod for hate speech not allowed on Site A, and it is understandable providers are not on board with it.
- toomim 6y agoParler does ban hate speech within the rule banning "Any direct and very personal insult with the intention of stirring and upsetting the recipient". https://www.techdirt.com/articles/20200625/16303844790/just-like-every-other-platform-parler-will-take-down-content-face-impossible-content-moderation-choices.shtml https://www.techdirt.com/articles/20200625/16303844790/just-...
- BoorishBears 6y agoParler does not ban hate speech. Parler bans "tortious" speech, meaning speech that would open someone up to a civil tort, think defamation. In fact, Parler later switched to just using the simplified wording of "we won't be used as a tool in a civil tort" - My example is not tortious, it is not targeting a person in a direct and personal manner. There is _a lot_ of hate speech that is not tortious, and does not fall under "direct and very personal" Spamming nig**r in a public forum is upsetting to some readers, but not covered under the fighting words doctrine they tried to invoke nor their guidelines. Saying "I hate these damn <insert race or religion> they should all go back home" is the same. - And like I mentioned above, your article is out of date, their latest guidelines before going down no longer even bothered with breaking this out and all of what was quoted no longer existed. Instead it was all replaced with > Parler will not knowingly allow itself to be used as a tool for crime, civil torts, or other unlawful acts. You can confirm this by googling this URL (which us obviously down if you try to visit it directly) "https://legal.parler.com/documents/guidelines.pdf https://legal.parler.com/documents/guidelines.pdf" The latest revision was on 12/04