TFW you realize how many hundreds of billions have been made by first convincing people that structured data stores are too hard to set up and maintain, then 5-10 years later convincing the same people that unstructured data stores are not conducive to fast and reliable analysis. Rinse, Repeat.
Comments
In the late 80s and early 90s, we got hard drives and started collecting lots of files.
1996: Ralph Kimball publishes the first edition of the Data Warehouse Toolkit, and suddenly we decide relational databases are cool now.
(obviously this is a hugely reductive summary that leaves a ton of out, but the cyclical pattern of each approach solving the problems of the other and reintroducing their own is pretty impressive to see go on this long)