My first job doing actual database programming was around 30 years ago, writing a dBase III+ system that sucked information from a commercial time charge billing system and produced much nicer and richer reports.
That’s ETL. The E and L were easy, the T took a bit of interesting work (setting the third bit of the last byte of a number to 1 to represent the field being a negative value?!). But when did the term start being bandied about in SQL Server as common nomenclature? At a guess, I’d say when Data Transformation Services became a part of the package, whenever that was.
Then a decade or so later, I wrote a system in Microsoft Access (probably 2.0) that allowed multiple users to coordinate individual databases so that each user could see other people’s information, but couldn’t change it.
That was sneaker net replication. Everything was copied on to floppies (some data sets were so minuscule back then!) and it worked reliably. I have no idea when replication entered the database lexicon, but I certainly didn’t use the term back then. We had SQL Server, but this group was isolated from our network because of reasons, so I had to provide the service through some unconventional means. And it worked.
So what’s next? What are we doing today, that we think is a one-off process, which will become a TLA or known term in a couple of decades? I have no idea, I hope to be fully retired and enjoying the German countryside then.