Google catalogues everything
Recently, I searched Google images for a topic and found a matching image. I was surprised to observe that its URL domain was https://encrypted-tbn0.gstatic.com/
I then copied path from the thumbnail results page and used Google's "Search by image" function using the URL and it returned a single match that turned out to be a PDF. Opening the file showed the image but it had been cropped when the PDF was designed (it was a marketing brochure). Whilst the Google copy was a thumbnail of the full size image.
So in its unrelenting indexing of everything it finds, it appears Google deconstructs PDFs into their constituent parts and catalogues them discretely.
:/
Biting the hand that feeds IT
