When Oracle's chief executive Larry Ellison announced the company's Unbreakable Linux 2.0 strategy at Oracle OpenWorld in October, it seemed likely that the database and applications giant was poised to muscle Red Hat out of its dominant position in the Linux support market or snap up the company at a bargain price.
Three months on and the first indications are that the plan has failed on both counts. It might be early days, but download figures of 9000 for the first 30 days (revealed by Oracle's president Chuck Phillips during the company's recent earnings call) must be disappointing.
It should be noted that customers do not have to download the Unbreakable Linux code to become an Oracle customer, they can also simply sign on for Oracle to support their Red Hat installations.
However, one of the claimed benefits of Unbreakable Linux is that it includes Oracle's own bug fixes, and the download figures are an indication that there is a low level of interest in Oracle's code.
As ComputerWire recently noted, Novell enjoyed 325 000 downloads of its SUSE Linux Enterprise distribution in its first 28 days, while Red Hat's non-commercial Fedora Core 6 had an average of 12 500 installations per day in its first month.
While these figures are for both server and desktop downloads, Novell's figure of 170 000 for SUSE Linux Enterprise Server also puts Oracle's server-only Unbreakable Linux in the shade.
Of course, downloads are good, but paying customers are better, and converting one to the other is notoriously difficult in the open source model.
As an indication, open source customer relationship management software firm SugarCRM recently announced 1000 paying customers from over 1 million downloads.
These figures are not unusual, even for the most successful open source projects. Open source database firm MySQL has a conversion rate of roughly 1 in 1000 for example, while Marc Fleury, senior vice president and general manager of Red Hat's JBoss division, recently revealed that "on average we only monetise 3% of our user base for JBoss and roughly 10% for Linux."
Even allowing for adoption by Oracle's existing customers resulting in an initially higher rate of adoption, it seems likely that Unbreakable Linux would follow a similar download conversion rate.
Assuming that Unbreakable Linux enjoys the same sort of conversion rate, it is possible to predict that in its first 30 days Oracle currently gathered at best 90 and at worst only nine customers for the Unbreakable Linux program from those downloads.
ComputerWire speculated at the time of the Unbreakable Linux announcement that perhaps the ultimate aim was not to attract Linux customers but to drive down Red Hat's share price and make the Linux company a more attractive acquisition target.
If that was the plan, it has backfired. While initial investor fears wiped over $1bn from Red Hat's share price, investor confidence returned as decent third-quarter results saw shares leap above the $19,51 they were trading at before Ellison's announcement. Red Hat's shares closed 2006 at $23.
While it is still early days for the Unbreakable Linux plan, Red Hat's strong third-quarter performance indicates that it has had limited short-term impact on the Linux market leader, while the download statistics also indicate that the long-term impact is also likely to be negligible.
Oracle's history can boast a number of success stories as well as much-hyped failures. The company will have to work hard in 2007 if Unbreakable Linux is to follow in the footsteps of the Oracle Database, rather than Raw Iron.