This is just part of Larry Ellison's talk about cloud computing, it was made to keep shareholders happy by downplaying the hype for cloud computing, it's not about the technology itself.
Larry Ellison doesn't like cloud computing because despite all the benefits of this advancement it would turn his companies business models upside down. Oracle would lose large amounts of money because they rely on massive licensing deals and expensive maintenance (assuming they go SaaS).
They have developed an independent unit to build out their on-demand efforts, just as Innovators Dilemma recommends when faced with a disruptive technology.
But the market is still to new and if Oracle made a major shift toward it, it would cannibalize a large portion of their profits. Therefore dropping the share price.
A big part of cloud computing is the idea of SaaS where they charge pay-per-user monthly or based on usage. This is driving costs down and the software vendor is responsible for upgrades and maintenance (a major part of Oracles revenue).
He has been quoted saying that there's no money in the on-demand model. These enterprise software companies had massive profit margins in the 90s with the licensing model.
It's in Oracles interest to slow the transition towards cloud computing and SaaS. Although it is overhyped, like all new technology.
The idea behind this is that every company buys their own Oracle-based grid. This means lots-of-money for them.
If everyone switches to "the cloud" with non-relational data stores, that's not so good for them. (Even if "the cloud" is Oracle-backed, they still only sell one copy.) They have made a lot of money convincing people that their product is magical, and they don't want to lose out on that.
You know, back in the day, Oracle did exactly the same thing themselves - they released their product on IBM mainframes because no-one would take them seriously otherwise. They even had to leverage a consulting contract with the CIA to claim to be "government contractors" to convince IBM to sell them ones.
By no means did they invent this strategy, nor are they the only ones to use it, they're not even the only database vendor. MySQL relies equally on "magic". Selling on your technological merits doesn't work. Look at Sybase and Informix.
Since when did the cloud mean non-relational data stores? For that matter, since when did the cloud mean anything concrete * ? I've heard cloud computing describe more ideas (all of them ancient) than almost any other term in computing. Yes, the cloud is yet another instance of "What's old is new again."
* other than "it's running on someone else's machine"
Is he really railing against cloud computing, or instead the rush to label everything cloud computing? It sounds like he embraces the model, he just doesn't like the dopey name (and inclusion of everything internet-related under the cloud computing umbrella).
same could be said of "web 2.0" or "information super highway" which were similarly overly-hyped and overly-applied.
Larry Ellison doesn't like cloud computing because despite all the benefits of this advancement it would turn his companies business models upside down. Oracle would lose large amounts of money because they rely on massive licensing deals and expensive maintenance (assuming they go SaaS).
They have developed an independent unit to build out their on-demand efforts, just as Innovators Dilemma recommends when faced with a disruptive technology.
But the market is still to new and if Oracle made a major shift toward it, it would cannibalize a large portion of their profits. Therefore dropping the share price.