The Minix 3 codebase first of all has much fewer bugs because of its size alone (3,000 lines compared to six million for Linux code), Tanenbaum said in his presentation. When questioned by Linux Magazine/Linux Pro Online about whether Minix's growing codebase may eventually end up with the same bug rate, Tanenbaum replied that it was more important to consider the type of bugs, which brought him back to his original argument against the Linux kernel. His view is that drivers and other services should not be poking around in the kernel and would thereby have less of a chance of endangering the system. Overall, he said, performance speed is less significant than reliability: "If someone had a choice between a two- and three-GigaHertz processor, they would take the slower one if it would run more stably over many years."

.

Tanenbaum's argument is practically the same as he's had for the past 20 years. His main track talk at FOSDEM (which one insider revealed to Linux Magazine/Linux Pro Online as having originally been handed in as a Lightning Talk) was therefore less of surprise. More interesting is that the computer science professor has recently found a community for his free BSD-licensed Minix 3 project. It was clear that he was having just as much fun as his audience at his talk, which was laced with humor and graphic elements and which was followed by a 20-minute question-and-answer and photo session.

Born in 1944 in New York, Tanenbaum currently teaches at the Free University of Amsterdam, where since 2008 his research has been under the aegis of a EU grant for design and implementation of particularly stable operating systems. Minix, which he started in the 1980s, is apparently the reference project. Since its third version, the microkernel operating system has been open source under a BSD license. Tanenbaum invites developers to visit the project page to encourage porting of programs and libraries. As immediate tasks he named live operating system updates without reboots and distributing Minix processes on many operating system kernels. The project icon is the raccoon ("At some point every project is an animal," says Tanenbaum), because it is small and clever and "because it eats bugs."

Tanenbaum had written an earlier guest article on Minix for Linux Magazine Online. Whoever might be questioning the sense of a stable yet little distributed operating system that is currently serving primarily research might find some answers in a recent Slashdot discussion.