"The identity of the reviewers will remain anonymous, unless they choose otherwise — as happens now."
(Also "support for public review [being] inversely correlated with reviewing experience" means inexperienced reviewers are more likely to support it. Not less.)
As for the anonymous part, that's why I wrote "with style-detection and deanonymization tools". If the Internet could find Shia Labeouf's flag in a day [1], could they find a reviewer based on their writing?
[1] https://www.dailydot.com/unclick/4chan-shia-labeouf-secret-l...
You don’t have to actively demonetize your reviews, you just have to prime your audience ahead of time that your ~~election~~ publication was stolen and they’ll do the rest.
Though, we do eventually need to have a conversation about deanonymization of online accounts. That's not a thing we want to be done so easily