Some Early Adopters Stung By Ubuntu's Karmic Koala

Norsefire writes to mention a Register piece reporting that early adopters are having a tough time with Karmic Koala, Ubuntu's latest release. "Ubuntu 9.10 is causing outrage and frustration, with early adopters wishing they'd stuck with previous versions of the Linux distro. Blank and flickering screens, failure to recognize hard drives, defaulting to the old 2.6.28 Linux kernel, and failure to get encryption running are taking their toll, as early adopters turn to the web for answers and log fresh bug reports in Ubuntu forums." What has been your experience if you've moved to Karmic?

Comments

Professionalism

[sopssa]

Just imagine the amount of bashers if the news would had read;

Windows 7 is causing outrage and frustration, with early adopters wishing they'd stuck with previous versions of the Windows. Blank and flickering screens, failure to recognize hard drives, defaulting to the old kernel, and failure to get encryption running are taking their toll, as early adopters turn to the web for answers and log fresh bug reports in Windows forums.

This again comes from the fact that both Windows and Mac OS X releases are properly tested and maintained and tend to be in more professional quality.

But why don't the Linux distros go to same lenghts? It shouldn't be impossible, unless of course, commercial projects are maintained more professionally.

Re:Professionalism

[db32]

The irony is too good...

Flagging this as "Troll" for being critical of how Linux distros don't get the same levels of QA testing isn't exactly demonstrating great professionalism...

Re:Professionalism

[Anonymous Coward]

part of the reason is that community are the testers. you never should move to using a new release as soon as its out.

Re:Professionalism

[characterZer0]

But why don't the Linux distros go to same lenghts?

Debian does go through great lengths, and people complained that the time between releases was too long.

Then they switched to Ubuntu.

Re:Professionalism

[Shotgun]

How much did those users pay for their copy of Karmic?

Yes, it does make a difference. If I pay for a finished product, I expect it to be finished. If someone hands me a CD and says try this, I will try it, but not get upset if it doesn't work out perfectly.

In this society that we call open source, we fully understand that Canonical doesn't have the resources to run large test labs. We also know that we get the product for free, and can ban together with a large cadre of like-minded folks to fix problems that we do find. Most Ubuntu releases are initially full of problems. They tend to dissipate much quicker than your first Service Pack that you'll get from the behemoth that HAS charged you enough to do some proper engineering and testing.

Re:Professionalism

[MrSenile]

Funny, and all this time I thought the majority of the reason people use Windows is because it comes pre-installed on all the computers that they buy. The fact it 'just works' is because the people who make the systems have already prebuilt the systems with a tested and verified image that works on that specific hardware being purchased.

If they did the same with Linux, which some distributions do, then it would 'just work' as well.

I mean, get real. Can you imagine grandma getting a barebone system and installing Windows 7/Vista/Xp from cd, then having to search the internet for the drivers required for the hardware that isn't automatically recognized?

Pretty much the same headache grandma would have looking for any missing linux drivers, and funny enough, in a bare-bone install, linux is likely to support more out of the box than Windows. Go figure.

So for the 'lack of tinkering', you have to thank Microsoft and their excellent marketing division and their stranglehold on the hardware corporation.

Cheers.

Re:Professionalism

[Angst Badger]

Right. That's why I'm still running XP on the Windows side of the box and have no plans whatsoever -- nor really any motivation -- to upgrade to Windows 7. And the next box will likely be running XP under virtualization. Odds are that it will be quite some time before there is any significant Windows software that won't run under XP.

You can bandy the word "professionalism" around with all of its varied meanings and hope that no one here is literate enough to call you on it -- this is Slashdot, after all -- but the fact of the matter is that the only relevant aspect of professionalism here is the amount of money involved. When you're running a multi-billion dollar company, you can afford to test your software on a wide variety of machines with a large QA staff to run the whole exercise. Microsoft and Apple have the billions; Canonical does not.

All that said, there are any number of free software packages out there that are polished and refined and blow away their commercial competitors, so it plainly can be done. On the other hand, an operating system and all of its associated software is a lot more complicated than any single application, so testing it thoroughly has got to be a daunting task. Moreover, the risk and effort involved in downloading the latest Firefox beta is much less than downloading and installing an operating system beta, so there are probably a lot more testers for apps than OS distributions. Still, the last couple of Ubuntu releases have had non-trivial problems, and for a distribution that prides itself on stability, this definitely should serve as a wakeup call to the folks at Canonical.

In the end, though, I'll take a rough start on an Ubuntu point revision over the "professionalism" of Windows Vista and, for that matter, the rough start that many people have reported with Windows 7. And while I'll grant you that OS X is a polished product, several OS X releases have had noteworthy issues, and that doesn't even begin to cover the primitive suckware that passed for the MacOS pre-OS X. Modern operating system development is hard. Neither commercial nor free OS producers do it as well as we'd like. Even so, how much do you want to bet that there are fixes for the problems with Ubuntu 9.10 a good six months to a year before Microsoft issues its first service pack for Windows 7?

Re:Professionalism

[Korin43]

And then I thought Ubuntu was too slow, so I switched to Arch (rolling release) and it's more stable. That may seem strange to you, but only if you don't know how Ubuntu/Debian defines "stable". Stable on Debian and Ubuntu means old. If it was release a year ago, it's stable. Who cares if sound doesn't work on your computer, at least it's stable! Who cares if pidgin-facebookchat crashes every couple minutes, in Debian-land it's stable (this is a particularly interesting case because pidgin-facebookchat was added right after the project started, and then Ubuntu arbitrarily stopped adding new versions to the repos even though the plugin still isn't done, so every release adds to the stability). Mozilla release are remarkably stable and always contain security updates.. but sorry, Firefox 3.5 wasn't old enough until this release. Every version of the nvidia drivers add more stability, but I think we'll stick with the old versions.. you know.. because they're old.

And that's not to say that sticking with old versions is always bad, it's just that the method of deciding what's stable is literally "is it old?". Why not test things and then update, instead of arbitrarily picking a version and declaring it to be stable? Or keep track of projects that release safe code and give them 2 weeks to make sure there's no horrible bugs, and then update (like what exactly is the reason for holding back Firefox and Pidgin?).

Re:Professionalism

[nurb432]

Well, then they don't want windows either. The fact windows needs 'tinkering' as you put it is why there is an entire PC support industry in the first place.

Re:Professionalism

[tweek]

NetworkManager is a piece of shit w.r.t wireless. I've read every fucking thread out on various mailing lists and the author simply says "It's the driver's fault" despite the same problem happening across the board to multiple users of different cards.

The biggest problem is the stupid fucking background scanning it does. What happens is that when NetworkManager gets a wild hair up its ass and decides it time to scan for more networks, your wireless NIC will disassociate from the current AP until the scanning is over. God forbid there happens to be one shitty AP somewhere at the edge of your range and it takes too long to respond. Your connection is toast and you have to re-associate but meanwhile you've just lost connectivity for 2 minutes. Hope you didn't need that download anytime soon or that you remembered to screen that SSH session to a production server. Any machine I use that has wireless, is running WICD now instead of NetworkManager ( http://wicd.sourceforge.net/ )

I love Ubuntu. Honestly the only problems I've ever had were with the switch to PulseAudio. I grew out of tinkering with my distros a LONG time ago. I need my machine to work so I can work. I did a fresh install of Karmic and moved my home partition stuff around this time. The ONLY problem I had was with PulseAudio and my Audigy card ( https://bugs.launchpad.net/bugs/467732 ).

indeed

[pele]

me being one of the early adopters that got stung
I haven't seen so many bugs and reboots since the days of windows 95

Re:indeed

[Eil]

*sigh* We see these kinds of articles on every major new release of Ubuntu/Fedora/Windows/OSX. This is NOT news. When you're swapping out major parts of your OS and applications, things are bound to break. I'm not an Ubuntu fanboy or anything, but this kind of stuff gets on my nerves. To everyone who claims they were "stung" by this update, I have two questions:

1) Did you bother to test the new release at any point during its 6-month development cycle? The alpha and beta builds are available as a Live CD well ahead of the final release, it's a trivial matter to burn a copy, stick it in your machine, and give it a test run.

2) If stability is important to you (and I assume it is by the use of the word "stung"), why did you upgrade anyway? If I'm not mistaken, Karmic is not even an LTS release.

To provide a counter-example, I have 5 machines under my control that have been running Ubuntu for years. Out of those, NONE have ever had a problem upgrading to any version of Ubuntu, even Karmic.

Encrypted home folders, a balanced look...

[solevita]

In fairness, it does sound like the failure of a single individual to get their home folder encryption running was picked up by El Reg and blown up out of all proportion. Flickering screens? Yes, I saw that, but it was fixed by a fresh install rather than an upgrade.

There are some niggling bugs and lack of polish, but this isn't anything like Canonical Vista, despite what some people are hyping.

The plural of anecdote isn't data

[quanticle]

As long as we're trading unsubstantiated anecdotes, let me say that my experience with Karmic Koala has been perfectly smooth. I have it running natively on one machine and inside a VirtualBox VM on another, and in both instances both the install process and the system as a whole have worked very satisfyingly.

Re:Great

[MichaelSmith]

The two boot screens look sparse and cold to me. I wondered if Mark Shuttleworth was paying people back for the complaints about his "human" color scheme. The GDM window looks ugly to me. I definitely want the old one back.

Re:Release cycles?

[nine-times]

There are a couple things that I'm inclined to point out. First, the article is basically saying, "some people had trouble, some people were unimpressed." It's hardly a scientific study of the quality of the OS. Sometimes the complainers are the most vocal, and the people who are happy sit quietly.

But more important, just a bit of advice for anyone who got burned by the upgrade and are upset: if your computer is important to you, don't be an early adopter. Just because a new version of your OS comes out doesn't mean you need to upgrade right away. Sit and wait to hear what people say about it, and wait for some of the kinks to get ironed out.

I'm not making excuses. Yeah, sure, it'd be better if Canonical would make sure that every release was perfect right out of the gate, but still, exercise some common sense. If you've been doing this for any amount of time, you should know better by now, especially since it has happened with pretty much every single OS. When Vista was released, it was a buggy POS. Yes, I used it. They cleaned it up well enough, but it wasn't any good when it was released. I forget which release of OSX it was (maybe 10.3?), but one of them erased your external hard drives if they were connected when you installed the new OS. That made it really fun if you had just backed up your data to an external hard drive in preparation for the upgrade. And I think it was FreeBSD 5 where everyone was complaining about how crappy it was for months after release.

Whatever system it is, you just can't trust blindly that they'll have it in perfect working order on day 1. If you want to be an early adopter, great, you get to help work out the kinks. Otherwise, give it at least a month or two.

Re:I got a bit stung

[diamondsw]

Wait, did you say everything went smoothly except you didn't have sound or video ?

That right there is why Linux hasn't gone mainstream.

Re:My experience

[Daishiman]

Guess you've never had a Windows install crap out of the blue or become noxiously saturated with garbage at book. I admit that the quality of releases in Ubuntu hasn't been as good as Windows during the timeframe I've used it. Nontheless, I've always been able to fix stuff in Linux, while I've had to reinstall Windows from scratch many more times.

Re:Release cycles?

[Jezza]

Mine is a disaster. Now am I dropping Ubuntu? No, I'll drop back to 9.04 (I have all the data - I've been around the block enough times to not make that mistake). However, I might look at Red Hat if the problems aren't resolved quickly.

And here's the advantage of Linux - I can move to another supplier, I'm not locked in.

Re:Release cycles?

[reallocate]

So... Linux is not ready for the real world?

Rants replacing Bug reports?

[Jerry]

I've been using Linux for 11 years. Before Linux captured 10+% of the desktop market share (according to Ballmer himself!) most of the community was technically oriented and ranting wasn't that common. We understood that those doing the developing were VOLUNTEERS and the best way to help them was to post BUG reports filled with details of the bug that the developer could use to resolve the bug and fix it. IOW, the users were the testers. We understood that and agreed to it. We were patient and our patience was rewarded.

Now, we have a generation of users who don't appreciate or care that most of the developers are still volunteers. These users don't care that they get the OS, the desktop and tens of thousands of high quality apps for free. Even worse, they don't want to take the time to take notes of the problem they think they are having and file factual bug reports at application's bugzilla site. What they will take time to do is write rants in blogs and news groups. Rants that are devoid of facts or knowledge but long on flames and vituperations. Thankfully, most developers know about these kinds of "Penguins" and ignore them. What else can they do? The rants rarely contain useful information and the developer doesn't have the time to search the countless blogs and forums for rants about his software. If he did he wouldn't get any developing done and he'd get discouraged and quit, which would make Microsoft happy,

To make matters worse, many ranters are serial ranters. They aren't satisfied with ranting in a single forum or blog. They visit as many as the can and post essentially the same rant in all of them. This makes the ranter appear to be part of a larger movement when, in fact, he is not. There were several ranters in the KDE4 dustup that were identified as serial ranters, and for a year and a half you could track them through the Linux sites as they dropped one rant after another. If someone called them on the topic of a rant they'd switch topics in their next rant. It didn't matter. The purpose was to destroy KDE4, if possible, and force developers back to KDE 3.5.x. The ranters were totally ignorant of the technical issues and reasons why KDE was redesigned from the bottom up.

The examples of stupid rants are almost endless. One ranter registered on a forum just to make his first post a rant against KDE 4.2.1 because "IT didn't have a way to change the menu structure to KDE 3.5.10's." Read the documentation? NO! It takes too much time and he's much too important to do such trival stuff. Ask a question on the forum instead of ranting for his first post? NO! He's not about to humiliate himself by asking a newbie question.

So, he rants. The first reply states "right click on the K-Gear menu icon and select "Convert to classic menu".

Now, everybody knows that not only is he a mindless ranter, he is also an idiot.

The problem is that his subject line appears in some Google search of "Problems with Ubuntu" and adds at least one count, or more if the rant is picked up by multiple blogs, to the number of users supposedly having trouble with Kubuntu (or Ubuntu). Someone takes the results of that search and extrapolates it into a story about how "Some Early Adopters Stung By Kbuntu's Karmic Koala".

Meanwhile, my Kubuntu Karmic 9.10 instalation on my Sony VAIO VGN-FW140E/H notebook with an Intel GM45 video chip continues to hum like the perfect combination that it is. Did I say that I checked the compatibility of my notebook with Linux before I installed Linux on it?

Re:Release cycles?

[nametaken]

If you want a Debian that stable, use Debian. :)

Bad statistics - GIGO

[AliasMarlowe]

The Register failed to notice the text in red boldface on that ubuntuforums.org page which states:

"*** Disclaimer for those willing to analyse this poll ***
Most of users voting here are users with issues.
Users with painless experience are not likely to come here."

The statistics derived by The Register are thus invalid, and probably quite wrong, being from a nonrepresentative self-selected subset of Karmic installations or upgrades. Here's another nonrepresentative data set: I have installed or upgraded 4 PCs from Jaunty to Karmic at home (2 upgrade 32bit, 1 upgrade 64bit, 1 conversion 32bit to 64bit). All went flawlessly, even the migration of user accounts and reinstallation of applications (including commercial paid-for apps) on the 32bit to 64bit reinstallation. Being a self-selected non-representative dataset, would that entitle me to proclaim that every Karmic upgrade or installation was flawless? Obviously such a conclusion would be unfounded, and so are those of The Register.

It's tricky to get reliable statistics on Ubuntu installations. According to an unofficial monitor on the official torrent tracker, there were over 16 million torrent downloads as of today http://spreadubuntu.neomenlo.org/. The number of direct downloads from the servers is unknown, and the average number of installations per download is also unknown. BTW, I've uploaded more than 60GB on these two torrents in the last several days from home, and the upload rate is still humming along (I limit each of the torrents to below 1Mbit/sec upload).

It's also tricky to get reliable statistics on Ubuntu installation problems. The forum mentioned by The Register probably has only a fraction of those with problems, and that came to about 1400 as of yesterday. Comparing this number to the number of torrent downloads would give 1 in 10,000 but that would also be an example of bad statistics, since both of the numbers are incomplete to an unknown extent or nonrepresentative to an unknown extent.

Systematically incomplete nonrepresentative data produces incorrect statistics. It's the old adage: GIGO.

Re:Release cycles?

[TheLink]

5 years? LTS is just for 3 years.

And even then it's still just a gamble by Ubuntu and the users. Or should I say a hopeless dream.

Because not all the developers involved in building the software in the LTS release work for Ubuntu. Ubuntu can't force them to fix bugs even if they are critical, and worse it's even harder to convince them to _backport_ fixes to some old version.

So what actually happens with "LTS" (or most Linux Distros) is it gradually gets less and less supported over the years. The developers just say "Bug? Try the latest version and get back to me"[1], and if the latest version just doesn't quite fit with "LTS" you're stuck with the options of living with the bug or heading to uncharted territory.

With a server system it's usually not such a big problem since you don't tend to change the software and hardware much. But for a desktop system - you might wish to change your vidcard, soundcard, printer, network card or harddrive (to SSD with TRIM for example) within that 3 years. And if the support happens to only be in the latest and greatest Linux kernel, good luck getting it backported to your "LTS" kernel.

Or say the developer totally revamps the architecture of something lets call it XYZ - you could end up with a split - old XYZ for old stuff new XYZ for the latest stuff - but your LTS GUI might not be fully compatible with the latest XYZ for some stupid reason. You grumble and the GUI developers say "try the latest version". So now you have new XYZ and new GUI on your "LTS" distro, which kind of defeats the purpose right?

In contrast, Windows 2000 and XP have actually got better and better supported over the years - more and more drivers were released that wouldn't BSOD the system, more and more software released that didn't require Administator privileges to run (or even install - many games and apps nowadays install fine without requiring admin). Yes support for Win2K is dropping, but that's after way more than 3 measly years.

[1] In my experience the developers too often say "WONTFIX" or "WORKSFORME" even if the behavior is broken. Good luck spending a fair bit of time convincing the developer that its broken and should be fixed. Yeah it's free software, so I'm happy that it mostly works as it is, but still...

I think too many of the bug reports are going directly to a developer. I think they should go to someone like a project manager (with a clue). The project manager can then coerce the developer to "fix this", or just ignore the bug (dupe or user error) and not have the developer even know of the report. Or group a bunch of reports into one bug, or split a report into a bunch of bugs.