Why Major Non-Ubuntu Distributions Need to Step up their Game
Listening to a review of the new Ubuntu release I could not help but notice the amount of hype Shuttleworth’s little distribution can generate. Can you feel it? The buzz is orders of magnitude greater than with any new major distro release. I’ve criticized Ubuntu in the past, but there is no denying that Ubuntu is a milestone in desktop Linux and has done a great deal of good by making Linux adoption easier for the masses.
I decided to once again examine the Distrowatch distribution rankings. While these are just a very rough estimates based on site analytics, they give us a relatively good picture of the current state in GNU/Linux land. In this article I would like to highlight a few distributions that have, to put it bluntly, left me completely confused as to where the projects are heading.
My grudge with Fedora Linux has a lot to do with the fact that I constantly need to keep reminding myself the project does not have the same mission statement as most major distributions. A lot of innovative (or frustrating) components, such as the Network Manager and PulseAudio have been introduced to the Linux ecosystem by the Fedora Project. Being on the bleeding edge and the testbed for Red Hat, the Fedora developer community definitely tries hard to produce a stable version of the Fedora Desktop.
I tested every single release since Fedora 9 on three different computers. RPM is a lot better now, but still light-years away from APT. The OS crashes frequently. Fedora 12 constantly reminded me of updates that I already applied several times. The developer community tends to waste time on irrelevant philosophical software choices (Gnote anyone?). Fedora has become the distribution for only the most persistent users that don’t mind using what is in essence a product in perpetual beta. Competitors on the other hand use Fedora as a source of new software packages, integrating them into a better base, solidifying their dominance over the original makers. Being a testbed means your goal is to provide a stable product in the end. Well, at least that is what they claim Fedora to be. I remain unconvinced. If your product is too unstable, it needs more testing. If you believe you have tested the distribution thoroughly and many people still experience massive problems, it means something is wrong with the software base. There is no such thing as a stable testbed.
Mandriva Linux, actually its parent company, just recovered from some major financial problems. I do not know what the future holds for the once leading desktop distribution – it is always nice to see some cool stuff coming from the French developer community, though I cannot help but chuckle when podcast hosts or bloggers struggle to find a reason to actually use Mandriva (simply not being Ubuntu is not a valid reason), except for a good KDE experience, though I’m not so sure if Kubuntu lags far behind. Mandriva needs something like they had with Metisse, something that screams This is Mandriva and this is why we are unique. Despite the corporate problems 2009 was a great year for Mandriva, it is much more stable than it used to be. Unfortunately the WOW factor fades away after a few days of usage and one starts thinking why they didn’t simply stick with Ubuntu/Debian, since some specific features, e.g. urpmi and the USB installation tools leave a lot to be desired. Mandriva’s APT-powered counterpart PCLinuxOS has returned to a more frequent release schedule recently, which gives me hope (if they stick to their guns) because I view it as an overall better OS – together they might have a shot at world domination.
(Open)SUSE‘s German roots are still noticeable. The big, honkin’ distro comes loaded with different desktops. Although YaST had me swearing at the computer screen many times for hours, it is surprisingly stable. The goals of thsi project are “to make openSUSE the easiest Linux for anyone to obtain and the most widely used Linux distribution; leverage open source collaboration to make openSUSE the world’s most usable Linux distribution and desktop environment for new and experienced Linux users”.
A lot of the OpenSUSE architecture causes fragmentation in the Linux world, which makes command-line skills (i.e. using zypper, another take on RPM) harder to transfer. Another important improvement would be to slim-down the distribution a bit. This goal was partially realized with the OpenSUSE Build Service, but the process should be made easier for new users. NimbleX‘s customization UI seems like a good example.
The Linux distributions mentioned above are not going anywhere. They sometimes seem to be going nowhere. While Ubuntu, despite its flaws, is creating the widest desktop audience Linux has ever seen, the runners-up struggle to capture the flag for second place. If nothing revolutionary is introduced to the base or the desktop environment of the four mentioned distributions that Ubuntu or Mint cannot take and redistribute in a nicer package, they might go the way of OpenSolaris.
update: Mandriva is going to be forked, which means the community version might go in a completely different direction!
Related articles
- Mandriva Rises From The Ashes As Mageia (lockergnome.com)
- Welcome Mageia! (nowwhatthe.blogspot.com)
Piss-weak Panda, Lame Lemur, Ordinary Octopus, Newbie Numbat … whatever the code-name is this year, it’s still just the same shit, different year. Upgrade treadmills are for lusers who want to pay for the privilege. A new release is altogether too yawn-worthy to even notice – outside of the dedicated GNU/Linux press you wouldn’t know a new release of Ubuntu had even taken place.
Ubuntu isn’t suitable for everyone, even if you happen to like it, and it provides nothing of worth for me. There’s room in the world for many different takes on the same general problem and it would be pointless for all solutions to aim for the same narrow target that ubuntu does.
Here is my opinions on the matter :-
I have tried all major (and many non major – like yoper, etc) distro’s over the last 8 years (started on Mandrake Linux) and now I use just 2 – Ubuntu and ArchLinux.
I use Ubuntu for a ready out of the box distro and to ensure that i’m not missing anything in the DIY distro Arch.
I prefer Arch as it is the most up to date ‘stable’ distro – I have tried every version of Fedora (since fc1) – the longest it has stayed on my system is a few days – fedora is just not suitable for a desktop (far far too buggy) – Arch which mostly has even newer packages seems so much more stable than FC.
Opensuse I have always thought looks more professional that others (including Ubuntu – arguably until 10.04) and I really like the centralised control panel (like pclinuxos/mandrake), but zypper is just not as good (in speed and realibilty) – if I add external repos to opensuse (which you NEED to in order to have a useable system) it always occanioally breaks..
Archlinux fills the Gentoo ‘market’ in a way although far more up to date, it is also very easy to compile software with your own CFLAGS and later versions, etc – although obviously arch is not for everybody.
I totally agreed with mcox. My system has Ubuntu and Arch. I use Arch most of the time, Ubuntu is too fancy and does not suite by taste. I installed Fedora 13 (which seems stable but a bit out of date and the security update is too slow compare to Arch and Ubuntu). And the preupgrade process is terrible. I also install Gentoo but can get it to work. X can only be run as root but not as normal user, I still don’t get it till now.
Sorry, but I don’t see Fedora competing with Ubuntu. In my opinion, and that’s why I use it, Fedora is not about the distribution as a final “product” or something like that, the point about Fedora is a community (meh, I said the C word
) of contributors and free software enthusiasts who try to make free software better (read: try new stuff, fix bugs). If you’re willing to dedicate time to improve software and can live with the occasional annoyance after updating, that’s the distro you want. It’s just not competing with Ubuntu, because the philosophy behind the distribution is completely different. The releases are mostly just showcases to let users try what has been achieved with improving the software and fixing bugs. That’s probably why you see it as a testbed, because you go and use it with the same attitude you’d have when using Ubuntu. (in my opinion though, Fedora is heading a similar route lately, and I don’t really like that, but that doesn’t fit in here right now 
)
I’m not saying your blog post is bad, I liked reading it, I’m just trying to say that you’re comparing two things with an unfair measure here.
Same goes for other distributions as well, some people mentioned Arch for example.
As for Mandriva, I don’t know much about them, but it looks like they failed due to unrelated things (their company going bankrupt for whatever reason) and openSUSE may not have the market share that Ubuntu has, but is still a nice distribution that easily competes with Ubuntu and has their own way of doing things and some neat stuff that Ubuntu doesn’t, e.g. the excellent KDE integration or YaST.
[...] Why Major Non-Ubuntu Distributions Need to Step up their Game (internetling.com) [...]
Listening to a review of the new Ubuntu release I could not help but notice the amount of hype Shuttleworth’s little distribution can generate. Can you feel it? The buzz is orders of magnitude greater than with any new major distro release
I decided to once again examine the Distrowatch distribution rankings. While these are just a very rough estimates based on site analytics, they give us a relatively good picture of the current state in GNU/Linux land. In this article I would like to highlight a few distributions that have, to put it bluntly, left me completely confused as to where the projects are heading. ı am read thankss