It's official, kernel version 3.0 is on its way. Linus confirmed this both in an announcement and by checking in the first 3.0 release candidate. The new branch is not expected to include any major milestones in terms of features, but it may stimulate developers to adopt policies of cruft removal and modernization.

Linus had mooted the idea of skipping 2.8 (stable branches use even numbers) in favor of 3.0 a couple of weeks ago on the kernel mailing list.

On the forums, reaction to the news has been mixed, with some wondering if Linus is falling prey to the current trend of large major version number jumps to give the impression of significant progress. But, that raises the question: is it worse to have a small version number jump for a new release that breaks everything, or a large version increase that turns out not to be hugely different from the previous release. It's a point that Linus touches on in his announcement:

[...]we are very much *not* doing a KDE-4 or a Gnome-3 here. No breakage, no special scary new features, nothing at all like that. [...] So no ABI changes, no API changes, no magical new features - just steady plodding progress.

His justification for the version jump is two-fold. First, the 2.6.x branch was around for an unusually long time. 2.6 has been with us since 2003, and version 2 itself began way back in 1996. Second, it is symbolic because, as the Linux kernel is 20 years old this year, it is now entering its third decade.

So, it's not a ground-breaking release, and the message seems to be: everything's going great, prepare for more of the same.

The original announcement on the Linux Kernel Mailing List.