I think the author is on to something here, but isn't there a bridge across the moat re: scraping? Sure the API made things convenient, and scraping content wi

I think the author is on to something here, but isn't there a bridge across the moat re: scraping? Sure the API made things convenient, and scraping content will be a bit of an arms race, but scraping public postings that don't require a login to view still seems like a bridge over the moat. It's tempting to refer to the recent victory of LinkedIn over HiQ, but there's an important distinction in that ruling: It pertained to use of logged in accounts and did not explicitly overturn the prior ruling that public-facing data was fair game. For now it's all still unknown territory. I would be skeptical of anyone who adamantly affirms a position that scraping for training is allowed/not-allowed because it is not yet settled law in the US where these companies are headquartered or the rest of the world where they operate.