Right. So a somewhat rough analysis would be: sample a range of "average" tweets on both platform and somehow aggregate an average "hate" value.
What I'm getting at is, I understand Parler is generally a right-leaning platform, and therefore the types of "hate" will be right-leaning. Twitter is a generally left-leaning platform, so I would expect their type of hate to be generally left-leaning. So I think a tweet about storming the capitol isn't good evidence that Twitter is better. Because, for exmaple, perhaps a violent Antifa tweet would be left alone on Twitter but moderated in Parler. Perhaps some ML bot can quantity sentiment of tweets.
What I'm getting at is, I understand Parler is generally a right-leaning platform, and therefore the types of "hate" will be right-leaning. Twitter is a generally left-leaning platform, so I would expect their type of hate to be generally left-leaning. So I think a tweet about storming the capitol isn't good evidence that Twitter is better. Because, for exmaple, perhaps a violent Antifa tweet would be left alone on Twitter but moderated in Parler. Perhaps some ML bot can quantity sentiment of tweets.