Comprehending object storage and data access

Two part question.

First I understand cloud object storage is not the same as the traditional idea of a disk drive or a file system (‘block storage’). In object storage the files become “objects” that are monolithic rather than readable/searchable? Is that the right idea?

Ok and now the really important question: Can you explain how I can use object storage for a large dataset so that it isn’t just basically an archive? That is, I want to use object storage very interactively. I would like to do this from the cloud… but also over the internet. I feel like this is supposed to be possible but I need to really get how to make it happen; such that I’m sure we’re doing it correctly and it is efficient.