This will eventually mean anyone engaged in getting data off the web for machine processing (i.e. scraping) is probably looking at needing to run headless brows

This will eventually mean anyone engaged in getting data off the web for machine processing (i.e. scraping) is probably looking at needing to run headless browsers and using OCR to get the information out in future. I kind of wonder if Google etc. do this already to extract information based on layout. Somewhere along the way we need a better way to exchange data between websites before this comes along and sets us back by a couple of decades. The semantic web stuff (sadly) hasn't really worked, but something of that ilk is needed to get us to the future.