Date: Tue, 28 Mar 2000 11:36:34 +0200 From: Martin Cracauer <cracauer@cons.org> To: freebsd-stable@freebsd.org Subject: Let 3.x die ASAP? Message-ID: <20000328113633.A28085@cons.org>
next in thread | raw e-mail | index | archive | help
A bit of a controvesical question, for sure. I am so impressed with the results of upgrading my more important machines to 4.0 or 4-stable that I could imagine not to produce any more 3.x releases. Let us consider what happens when we release 3.5 before or at the same time as 4.1: - People will publish reviews based on 3.5 instead of 4.1. As usual, an unadaequate number of them running on SMP machines. - People will run into inferious hardware support, especially they will have trouble with large IDE disks and/or IDE PCI controller. Similar issues for Laptop users. - People will continue to run with inferiour NFS, Linux emulator etc. The same applies to networking fixes against attacks, especially DOS. - People will choose 3.5 over 4.1 and will later go through an 3.x->4.x upgrade for no good reason. - The remaining issues in 4.x (especially hardware) aren't addressed as fast. There will be some issues with "unclean" usage that the core userbase doesn't use that don't come up until the masses jump on them. For me, the real reason is that I now hate the latest 3.x after seeing the improvements. It is near to junk, IMHO. Several machine of mine had gone instable while moving from 3.[01] to 3.4-stable and I suspected hardware trouble. I didn't debug it because I didn't want to mess with 3.x anymore. Now after upgrading to 4.x the same machines are rock-stable again. It is my impression that the 3.x branch lost the required testing when the core committers moved to 4.0-current. Higher releases of 3.x are just not polished/tested enough anymore, beside the undoubted concrete bug fixes. Speaking of testing by committers, I also hate to have machines for three branches running to develop bugfixes that apply to more than -current. In fact my last 3.x machine (besides a 3.1 machine that doesn't have the IDE troubles I have with 3.4-stable) is now a production machine of my employer that I'd like to upgrade as well. I couldn't think of a good reason to delay that upgrade for 3 or 6 months if my boss asked me. When i upgrade, I had to commit userland bugfixes to 3.x with only testing them on a newer branch. Kernel bugfixes would be impossible unless I find someone to test them for me. I assume that I'm not alone here. That many bug-fixing committers will not move with 3 machines along the current branches and that 3.5 will not be a worthy successor to our previous *.5 releases. Opinions? Martin -- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Martin Cracauer <cracauer@cons.org> http://www.cons.org/cracauer/ Tel.: (private) +4940 5221829 Fax.: (private) +4940 5228536 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000328113633.A28085>