Andrew Morton, the lead maintainer of the Linux kernel v2.6, has not confirmed this statistically, but he feels bugs are being added to the Linux kernel faster than they are being fixed. It's quite possible that is true, considering the explosive growth of kernel additions.
This echos concerns of power users that have been voiced in various forums for the last year or so. To a regular user, like me, it shows up in the fact that in the 4 months since I installed my Linux system, I have had to revise the kernel 4 times to fix problems. (This is not as bad as it sounds, as such revisions most often amount to simply downloading a new kernel image, installing it, and then rebooting to put it in service.)
While such speculation in public surely does not help Linux' image with those not familiar with open development processes, this sort of discussion and the resultant feedback is an important part of those processes that have given us Linux in the first place. Even if you are not a Linux fan or user, it might profit you to follow this discussion and see how it's worked out. The process is often not pretty, but it is effective.
Jack
Jack,
ReplyDeleteI would agree about the "bugginess" of some of the recent kernels, especially the 2.6.16.x editions. My Compaq laptop was dead stable on the 2.6.15.x kernels but the .16.x ones have a BIG issue with older machines and ACPI - causes a kernel panic on initial boot.
For now I've revered back, filed bug reports with debug info and all that, and test each new .16.x release. No fix as of the latest kernel for Red Hat FC4. On the other hand, the Compaq desktop is running just dandy with 2.6.16-1.2096 on FC5.
Bill_B