Ever since Ubuntu shipped its first long-term-support release, the 6.06 Dapper Drake (one of my all-time favorites by the way), the distro's LTS editions have enjoyed three years of support on the desktop and five years on the server.

Now Canonical is extending desktop support for the upcoming 12.04 LTS (to be named Precise Pangolin) to a full five years on both the desktop and server, making the release that much more compelling for enterprise users and others (like myself) who might not necessarily stick with the release for the full five years but want the option of doing so.

It makes the quality and stability of this next release that much more important, as SABDFL Mark Shuttleworth enumerated in a blog post yesterday.

Read the rest of this post

I don't have anything against Ubuntu 11.10, or the Unity interface. I might even like it. I'm getting so used to the way my Android phone works that I'm open to new desktop paradigms/metaphors.

But Debian Squeeze with Backports + sundry extras is running so well, and I remember so very well how I regretted my move away from Debian Lenny in 2009.

If Ubuntu doesn't fall off the track into "every damn thing is new" crazy for 12.10, then I'll be interested. It needs to work.

I like it when Jono Bacon, community manager for Ubuntu, can focus on the cool things the Linux distribution is doing and not just the problems with the project. Not that those problems should be ignored, but if Ubuntu doesn't have the goods, what's the point?

To that effect, Jono offers a detailed screen-by-screen look at the upcoming 11.10 release on his blog.

Seen above is a smallish grab from Jono of what happens when you use alt-tab to switch between applications.

Read the rest of this post

I'm doing an update today on my daughter's Ubuntu 10.04 LTS-running Gateway Solo 1450, the 2002-era laptop that I upgraded from 8.04 in a not-seamless but doable operation for someone with a bit of experience in these matters.

I've done a lot of upgrades. I'd say maybe half were successful. That doesn't say much for upgrades. But when it comes to Ubuntu upgrades, I can generally make them work with a bit of Googling.

I've been hard on Ubuntu 10.04 over the life of the release. (I could find links, but I'm just going to keep writing.) While the UI changes in 11.04 (GNOME giving way to Unity) are bigger, I thought the changes from 9.10 to 10.04 were too huge and unproven for an LTS release. My opinion was and is that 10.04 needed to be 9.10 with bug fixes and not a total reworking of the GNOME theme with buttons on the other side of the screen and lots of unproven, slightly broken Ubuntu-coded (or -ordered) enhancements.

Read the rest of this post

In the comments to my article on Debian’s Mozilla team offering newer Iceweasel builds, I eventually wound around to an idea that I believe would provide an enormous benefit to Ubuntu users:

There should be an official Ubuntu LTS Backports repository.

I see a lot of value in the Ubuntu long-term-support releases, but they’re pretty much treated by the project as regular six-month releases with a longer support life.

Read the rest of this post

How's that for an incendiary headline? Before I continue, here's how I got here:

First I tested the Ubuntu 11.04 Natty Alpha image, which I thought way too raw for a release two months and barely two weeks away, with the Unity layer on top of GNOME barely functional.

Then I tried a pre-alpha of Fedora 15, due May 10, a full 10 days after Ubuntu Natty, and found that while I didn't seem to be running GNOME Shell, it was GNOME 2.91.6 and pretty much worked as normal, and anchored a live system that was functional and responsive, though pretty darn ugly.

The comments on both entries were mostly written by Ubuntu defenders, telling me how wrong I was to judge Ubuntu by this alpha image, how it was going to rock at release time, and how I should a) do some research, b) screw my head on right and c) basically realize that Ubuntu can do no wrong, so shut the f* up.

I'm not a Ubuntu basher.

I always say that Ubuntu holds a very important position in the world of free, open-source software as well as in the wider world of all computer operating systems and environments, and for that reason I hold the Ubuntu project and its corporate entity Canonical to a higher standard than other software projects and companies.

I even run Ubuntu, albeit on one computer and not some or all (but certainly not none).

And I went back into my past entries and found a couple of reviews of previous Ubuntu alpha releases that … actually were functional, and Ubuntu Natty at this point in time running a desktop window manager (is that what it is?), Unity, that is untried, barely tested and not terribly functional does not bode well for a release in under three months time.

Here is my comment from a thread in LXer that explains the reason NOT why I'm anti-Ubuntu but why I see a frantically waving red flag in the path Ubuntu is taking toward its bleeding-edge push for new technologies in what people are expecting to actually use on their desktops:

What I'm saying is if you can't deliver basic functionality in an alpha release of the distribution you hope to deliver to end users in two months time, maybe you should consider holding the troublesome features until the next release so you can provide a better user experience. Fedora did this with systemd, which they could have put in Fedora 14 but instead chose to hold for Fedora 15. Debian is extremely conservative as to what they'll put in a Stable release. Testing is frozen very early, and development then focuses on eliminating bugs in those frozen packages. And yes, Gnome Shell has had considerably more development than has Unity. Along with that "edge" in development, Fedora is releasing AFTER Ubuntu. Fedora gets a reputation for being too "bleeding edge," and I am among those who have been burned by changes in Fedora mid-release. I left Fedora after F14 when I couldn't get my video to work. Now with the same hardware, F15 displays perfectly. And I don't think it's anything Fedora did; the bugs were fixed upstream. I'm still not happy that Ubuntu made a big deal about pulling from Debian Testing instead of Unstable to create the 10.04 LTS, yet they pushed many new or newish features/services such as Ubuntu One and the Me Menu which clearly could benefit from a lot more development before going into a release that is supposed to last three years on the desktop. More care and more conservative package choice should be the guiding principles behind a release with such a long support life. I'm sure that Ubuntu One and the Me Menu features have been improved for 10.10 and will be even more polished in 11.04, but that leaves LTS users to either turn off the features or be forced to jump on the six-month cycle to get better versions. Pulling from Debian Testing is just lip service if you're shoving a bunch of stuff on top of it that has not been through as careful a development process. I want to like Ubuntu, I still use it on one machine, and I support many of the project's goals. But when Fedora seems more conservative in its releases, you know there's something that's not quite kosher. I don't think newbie users are well-served by such raw software. I hope I'm wrong and Ubuntu 11.04 turns out to be a rock-solid, fast and functional release that gives those new to Linux the minimum of trouble. Of course, there's always Mint …

I'm on the record as thinking the Ubuntu 6.06 and 8.04 LTS releases were great ones, but I'm not as happy with 10.04 LTS (though that's the version I run on my remaining Ubuntu machine), which was advertised as conservative in the way packages were pulled from Debian Testing rather than the usual Unstable but which ran off the rails by incorporating features added to the Debian base and other upstream packages by Ubuntu developers that were in no way ready for what I think a long-term-support release should be.

And I fear that Unity is another piece of software for which Ubuntu is both the upstream and downstream, with what smells to me like a mandate to release in distribution form before GNOME 3 and GNOME Shell is allowed to get any traction via other distros such as Fedora and Mint.

Question for you: Do you think there will be a Canonical-supported "spin" of Ubuntu featuring GNOME 3 and GNOME Shell? GNOMEbuntu? Or will it be Linux Mint?