> lock down content in the face of its increased value as private ML training data That's easy to fix: Give out API keys; moderator tools, accessibility tools

> lock down content in the face of its increased value as private ML training data That's easy to fix: Give out API keys; moderator tools, accessibility tools and third-party readers get free API keys. ML training data scrapers have to pay. No need to go to war with the community over that. Reddit has been constantly nagging users to use their App, so I really don't think it's about ML training data. It's about monetizing users by forcing them into their App.