Google is a private company. While I find their legal takeover of other's copyrights highly questionable, I can't blame them for building a database.

The problem is that they'll have no competition.

There's also a concern as you note with the discarding of paratexts when a corporate entity decides to digitize books, whether we're talking about dust jackets, watermarks, variant paper stocks, marginalia, end-papers, broadsides, etc.

Those huge warehouses that will become obsolete will also house many unknown rare and precious books that are still in general stacks. You can go into any old library and putz around the old dusty shelves, and you'll find unbelievably rare books.