2) They're different because they usually store and distribute her data somewhere other than twitter. If she deletes a thread, that copy persists. If people interact with that copy, she doesn't get to see or manage that. If it's through twitter's API, she can see interaction with her content and block people if they interact abusively. And third party clients generally don't profit off of individual threads.
https://bobbin.herokuapp.com/ is a compiler that is much more like a third party client; it all goes through twitter's API, and in follow ups she said it sounds fine.
She used to get requests to turn her threads into blog posts; now she does not.
Also, she didn’t seem to necessarily want money from them, but somewhat similarly to GPL’d open source projects, doesn’t like scrapers making money off of her work. Plenty of software devs license their work similarly to that as well, and it sounds reasonable to me.
reply