> As a general rule, I have found out that every time there is something that doesn't make sense ("smart people who want to make money should have figured out a solution to this!"), the answer is government interference with free trade.
Sigh. I guess that goes along with "government is evil in all sorts of ways".
Simply put, Oracle was the best product 20 years ago. Open Source DB projects were toys that would lose your data. So companies and orgs (and in our case, the govt) used what works.
We developed on that. That was what was available, and we used it and paid the licensing costs. Well, lockin is a thing; for us and our customers. If we were to spend the time making a branch to use $other_DB, we need to properly test that as well. Do stored procedures work? Is time/date working right? Is there a close mapping of privileges? Is the DB appropriately audited? Are there active devs on it?
All of those are YES for oracle. And its fuzzy about others. I'm sure we could defend using Postgres or other first tier open source DBs.... but how much work would it be to migrate? And for those customers who use Oracle elsewhere in their infra (multiple schemas for different webapps), they are unable to switch. They already have the DBAs in oracle. So, we'd have to support Oracle even after leaving.
It's a lose-lose. But I'm sure painting it in the libertarian "government interference with free trade" fits your worldview.
Because when I read
> "As a general rule, I have found out that every time there is something that doesn't make sense ("smart people who want to make money should have figured out a solution to this!")"
I see that the more likely solution is that I don't have enough information, and the surface answer is inadequate. There's a reason smarter people couldn't figure it out, and it's not because of pithy sayings. It's because the solution is a magnitude or 2 more complex than what it looks like.
Oracle weren't the only game in town then though. Sybase and db2 were around, widely deployed and enterprise worthy at the time. The sales and lockin tactics which earned oracle notoriety were not unique either but they did that part 'better'.
Sigh. I guess that goes along with "government is evil in all sorts of ways".
Simply put, Oracle was the best product 20 years ago. Open Source DB projects were toys that would lose your data. So companies and orgs (and in our case, the govt) used what works.
We developed on that. That was what was available, and we used it and paid the licensing costs. Well, lockin is a thing; for us and our customers. If we were to spend the time making a branch to use $other_DB, we need to properly test that as well. Do stored procedures work? Is time/date working right? Is there a close mapping of privileges? Is the DB appropriately audited? Are there active devs on it?
All of those are YES for oracle. And its fuzzy about others. I'm sure we could defend using Postgres or other first tier open source DBs.... but how much work would it be to migrate? And for those customers who use Oracle elsewhere in their infra (multiple schemas for different webapps), they are unable to switch. They already have the DBAs in oracle. So, we'd have to support Oracle even after leaving.
It's a lose-lose. But I'm sure painting it in the libertarian "government interference with free trade" fits your worldview.
Because when I read
> "As a general rule, I have found out that every time there is something that doesn't make sense ("smart people who want to make money should have figured out a solution to this!")"
I see that the more likely solution is that I don't have enough information, and the surface answer is inadequate. There's a reason smarter people couldn't figure it out, and it's not because of pithy sayings. It's because the solution is a magnitude or 2 more complex than what it looks like.