I recall turning in a report about Amazon’s use of Oracle as its core database. The client, a bank type operation, was delighted that zippy Amazon had the common sense to use a name brand database. For the bank types, recognizable names used to be indicators of wise technological decisions.
I read “Amazon: DROP DATABASE Oracle; INSERT Our New Fast Cheap MySQL Clone.” Assume the write up is spot on, Amazon and Oracle have fallen out of love or at least beefy payments from Amazon for the sort of old Oracle data management system. This comment becomes quite interesting to me:
“This old-world relational database software is very expensive,” Jassy [Amazon tech VP] said. “They’re proprietary. There’s a high level of lock-in. And they’ve got punitive licensing terms, not just allowing very little flexibility in moving to the cloud the way customers want, but also in the auditing and fining of their customers.”
Several thoughts flitted through my mind as I kept one eye on the Philae gizmo:
- Amazon’s move, if it proves successful, may allow Mr. Bezos to mount a more serious attack on the enterprise market. Bad news for Oracle and possibly good news for those who want to save some Oracle bucks and trim the number of Oracle DBAs on the payroll
- Encourage outfits that offer enterprise cloud solutions. Will Amazon snap up some of the enterprise services and put the squeeze on Google and Microsoft?
- Trigger another round of database wars. Confusion and marketing hype often add a bit of spice to the Codd fest
- Cause concern among the commercial, proprietary NoSQL outfits. Think of MarkLogic and its ilk trying to respond to an Amazon package designed to make a 20 something developer jump up and down.
Interesting move by the digital WalMart.
Stephen E Arnold, November 14, 2014
Phi Beta Iota: The CIA’s $600M deal with Amazon, like the CIA’s seeding of Google early on, make sense in this context. The good news is that the surveillance state is realizing it cannot do what it wants to do unless it gets serious about open source everything engineering. The bad news is that any combination of Amazon-CIA is as bad for good people and open information as has been the Google-CIA combination. It will take 20 years for this initiative to show tangible results useful to the public, if at all. Far better to trash both Amazon and CIA now (as well as Google and Oracle) and create a new Autonomous Internet inclusive of open databases along with open analytics and all the other opens, from now.