Speaking at Oracle OpenWorld Tokyo Ellison also confirmed that the company had enquired about acquiring open source database vendor MySQL AB and denied that Oracle’s recent open source acquisitions were designed to harm its rival.
There’s a lot of romantic notions about open source – that just from the air these developers contribute and don’t charge, said Ellison. There’s this idea that because it’s open source people who work in Radio Shack develop the software for free, it’s just not true.
Open source becomes successful when major industrial corporations invest heavily in that open source product, he continued. Every open source product that has become tremendously successful became successful because of huge dollar investments from commercial IT operations like IBM and Oracle and Intel and others.
Ellison has a point in that many of the major open source projects, such as Linux and Apache, have been pushed into the mainstream thanks to their adoption by the likes of IBM, and that major corporations employee the majority of open source developers.
His comments do not do justice to the vast number of smaller open source vendors which also contribute heavily to open source software, however, and Linux distributors – and Linus Torvalds et al – might be interested to learn that Red Hat didn’t make Linux: IBM made Linux, Intel made Linux, Oracle made Linux, according to Ellison.
We have many more developers on Linux than Red Hat, he added, pointing out that the Redwood Shores, California-based company Oracle Cluster File System to enable Linux to scale across enterprise clusters.
Ellison also cast doubt on the potential for open source software to pose a radical threat to the proprietary software market. I don’t think open source will replace traditional software. I think open source will in some areas replace traditional software, he said.
There is no good open source business intelligence product as yet, Ellison said as an example. There are huge gaps in open source, it will be a long time before open source becomes popular for what we call mission critical database applications.
That said, Ellison has recently been talking about Oracle an aggressive approach towards open source, and he added that the company will continue to make open source acquisitions following its purchase of Innobase Oy in October 2005 and Sleepycat Software Inc in February.
Having said that, we’re a great believer in open source, we’re a great believer in Apache. You saw us buy Sleepycat, you saw us buy InnoDB, so I think you’ll see Oracle choosing to participate in open source in those places where we think open source can win, but I don’t think there’ll be an open source ERP system for a very long time.
The acquisitions of InnoDB and Sleepycat have been seen in some quarters as tactical moves against MySQL, given that InnoDB was the open source database company’s storage engine of choice and it was rumored to be working on an alternative with Sleepycat.
Ellison denied the acquisition were made to harm MySQL but did confirm reports that Oracle had looked at MySQL as a potential acquisition. We’ve spoken to them, we’ve spoken to almost everyone, he joked.
Are we interested? It’s a tiny company. I think the revenue for MySQL is between $30m and $40m, Oracle’s revenue next year: $15bn, he added. We bought some open source stuff, we bought Sleepycat, which is the most popular, it’s actually much more popular than MySQL.
It’s the most popular embedded database, the fact that it was open source didn’t really matter, it’s a valuable business for us to be in, the embedded database business, Ellison added. I think you’ll see us buy companies that make business sense and add technology. MySQL: never say never, but it’s not top of our list.
