"Etch and a half" ( was Re: Bugfix/hardware support updates to stable releases?)
The idea (mentioned in the prior thread) of having an "Etch and a half" release with an updated X/kernel/installer sounds EXACTLY like what I was hinting at. Backports are great, but having a supported, Debian-tested release that Debian can give to users with new/exotic hardware (which has better support in newer kernels/X releases) would be much better. I've been wrestling with Etch this summer on my MacBook (first generation!), and such a release would probably eliminate 99% of my complaints. Many Debian users are probably in the same boat - they don't mind having an old GNOME/Emacs/coreutils, but do mind being stuck with a kernel/X/installer that doesn't fully support their hardware.
Anyway, I'm curious - is this still a legitimate consideration within Debian? If it were to be done, it would have to be December/Januaryish (any later and it would be too close to Lenny, unless of course Lenny is late). I figure that this would be a new "branch" - much in the same sense as sarge, etch, lenny, and sid are. Thus, one wouldn't HAVE to upgrade, but new users and anyone standing to benefit from a new X/kernel (and possibly some other bugfixes) would want to consider using the new "etch and a half". I think this idea would definitely benefit Debian - more people would be able to use it, and it would remain quite stable while still supporting the latest hardware.
Obviously, there are issues with the whole plan. For one, Debian would effectively have one more development branch - at least while any such "and-a-half" releases are developed. Effectively, this may end up looking like FreeBSD if such releases ever became common - albeit with less updates to the "-stable" branch. Furthermore, that would mean one more branch to support with security updates, upgrades to lenny (or lenny+1 if a "lenny-and a half" or a "lenny-and-two-thirds" are ever made), etc. These are all legitimate concerns. However, I think it's worth consideration - in my mind, this would fix Debian's greatest flaw in a way that's much better than the Ubuntuish "we must release the full distro every 6 months" approach.
I'm curious to hear what everybody thinks regarding this whole concept, and I'd love to see Debian seriously consider this.
P.S. I know I am not a Debian developer, and I don't even have a package in the archive. I totally understand if you feel like it's a bit overreaching for me to bring things like this up. Rest assured that I appreciate everything Debian does, and I don't mean to detract from that AT ALL.
I will say that I'd definitely test any "and-a-half" release if this ever came to fruition, and I'd work to help make it work out as I could. Heck, I've already been trying to work on making my own "and-a-half" to make Etch work fully on my MacBook :)