Very interesting editorial, Kendra. You've mentioned a lot that I've never heard of before, such as SQL Provision or other tooling to mask/de-identify production data. I've never worked anywhere, where developers had a local copy of the database. I'm familiar with the concept, but reputation only. In my current, state government job, I see either a mix of production/test databases, or production/test/development databases. The last of which is the oddest because the development database is shared so I see little difference between a shared test database and a shared development database, with the exception that the test database is one that users are supposed to run tests against.
On my team my boss is the only one who can make changes to the database. That is a very strict policy he's made. At best we can only make suggestions. And he makes those changes through TOAD, which as far as I know doesn't concern itself with testing different indices configurations looking for something that performs better.
At this point, whether a local dev database is better than a shared one is academic, to me. What is the proof of it?
Kindest Regards, Rod Connect with me on LinkedIn.