On a related note: anybody on here want to work on some kind of innovative enterprise database startup? (I don't have any ideas, I just want to work on one because I see a lot of potential).
Yes, they have a very robust database. But what they've been selling to enterprises for at least 20 years is better thought of as ... schema.
When you buy Oracle Financials (or whatever modules / pieces / etc.) what you're buying is "how to set up and run your business." All of the non-physical parts of your business operations are now entities inside Oracle SW (and, as it happens, stored in the Oracle RDBMS, too, but that's practically speaking irrelevant).
Once you're "on" Oracle, or SAP, or whatever, every invoice you send out exists only because it's in that software. The software is your system of record, your single point of truth for your whole enterprise. You might think you are physically a human employee or a box of nuts but if you're not entered into Oracle with a row ID, you don't exist.
So, yeah, Oracle has an RDBMS. But what they really have is the "operating system" or the "schema" of a large enterprise which otherwise would have to rely upon a vast pyramid of specialist operational knowledge (bureaucrats / paper pushers / functionaries) in order to know how invoices got approved, sent, dunned, paid, reconciled, etc., or how orders got quoted, approved, picked, packed, shipped, etc.
(Not meant to sound condescending; I certainly wouldn't have known had I not spent some time seeing Oracle deployed in real life. It was certainly super-instructive for me to have spent a summer during college working for a midsize enterprise on Oracle, in a role kind of bridging ops and IT, and seeing what it really did. No SELECT statements involved.)
What are some promising open source efforts out there to create such a universal schema?
(I'm thinking along the lines of "The Cathedral and the Bazaar")
Relational databases have had [technically] open-source alternatives from the beginning, you're right. But they were also always playing catch-up to commercial databases. Oracle supported SQL long before Ingres, making them the lead product at the very beginning, and their services and support (and superior marketing power) kept them at the forefront.
Corporate or military investment is usually what drives cutting edge technology, which is adopted by necessity.
> anybody on here want to work on some kind of innovative enterprise database startup? (I don't have any ideas, I just want to work on one because I see a lot of potential)
Find out what a huge corporate product is lacking, make that, and then get them to buy you.
On your point about Military / Govt applications. This Larry biography:
"The Difference Between God and Larry Ellison: *God Doesn't Think He's Larry Ellison"
This book is an insanely amazing read. Apparently, the CIA was one of their first clients, and Oracle was named after a failed CIA project (the book hints that Larry might actually be CIA)