How are developers handling reliable web scraping at scale?