As anyone who has used Linux in the past knows, the dot-zero kernel announcement serves only one key purpose: it gets more people to test the new kernel, especially early adopters who don't often download interim kernels. These are folks who the kernel group wants running and testing the 2.4 kernel, to find bugs now so they can be fixed for the real releases of 2.4.
It's what Linus meant, in all seriousness, when he said that he released 2.4.0 to the general public because "things don't get better from having the same people test it over and over again." In a later notice, talking about the process for submitting fixes for 2.4, he made the point even clearer:
"In some cases people seem to think that 'since 2.4.x is out now, we can relax, go party, and generally goof off'--not so."
The release of 2.4.0 is not a giant technical leap from the 2.4.0 pre-release, or the 2.3 releases before that. In the world of open source, the development process is more of a continuum than a series of milestones. Unlike in the proprietary world, your access to works-in-progress is not limited to what the developer wants you to see.
Work on 2.4 has not slowed because of the release. Linus may be resting, but as I write this just days after the 2.4.0 came out, 2.4.1 is being readied for release. Some heavily desired features of the new system, such as direct support for the ReiserFS journaling file system, were not ready for 2.4.0 but are expected in a later release.
Don't expect to find 2.4 as the default kernel in packaged distributions such as Red Hat or Caldera this quarter. Most vendors didn't trust the 2.2 kernel for production systems until it was at about 2.2.12 or so. Development on 2.2 didn't stop while 2.4 was under development--the latest 2.2 update came out only two days before 2.4.0. Until 2.4 is ready for production use and becomes the primary release in commercial distributions, most users will continue to be happy with 2.2. And don't be surprised if you see more updates of 2.2.
Now don't get me wrong, there are some significant features introduced in 2.4: better support for USB and other hot-plug devices, better support for Itanium and other CPU architectures, better scalability upward (into the enterprise) and downward (onto embedded devices), and more flexible firewall capabilities.
But these facilities have been around for weeks, even months, in the 2.3 development kernels. Nobody who really follows Linux is being taken by surprise, which is another reason why you don't see much rejoicing among real developers and users. If you really want to follow what's happening with the Linux kernel, read the weekly recaps in Kernel Traffic.
As far as I can see, the hoopla comes primarily from analysts and others who took wishful projections, turned them into deadlines, then started whining when said "deadlines" weren't met. One might argue that the attainment of such fabricated milestones offers an excuse to call attention to Linux's advances. But let's face it, Linux has been receiving a lot of attention as a usable platform. It was plenty good before 2.4.0 came out, and it will be plenty good until 2.4 is ready to be put in a commercial distribution.
So consider the source when you encounter Linux kernel release hype. Anyone who thinks that 2.4.0 is a really big deal--good or bad--hasn't really been following it all that well.
Do you think the release of 2.4.0 is significant? Tell Evan in the TalkBack below or in the ZDNet Linux Forum. Or write to Evan directly at evan@starnix.com.
