How Hard could fixing ‘Hardy’ be?
Ubuntu 8.04 is coming on the 24th of April 2008. As is typically the case in the past few years, many millions of Ubuntu users across the globe are either waiting or, in some cases, trying out the pre-release software, reporting and fixing bugs.
Beta 1 of the upcoming version of this popular linux distribution came out a few days ago and, sadly, despite the fact that 8.04 is an LTS (Long Term Support) release, it seems that it’s going to be — by far — one of the buggiest releases ever to come out bearing the name ‘Ubuntu’. In my tests with ‘Beta 1′, there are hundreds of major bugs involving the brand new (and largely unstable) gvfs subsystem for GNOME, application compatibility with PulseAudio and general system behaviour, responsiveness and stability. Having said all that, Ubuntu 8.04, at least in its first Beta is significantly faster than its predecessor.
Having less than a month before the ‘final’ comes out begs the question: is Ubuntu 8.04 going to be delayed so that it matches 7.10 in terms of stability (remember this is supposed to be a Long Term Support release), or is it going to be released on time with a x.xx.1 version coming later in the year? Why has Canonical abandoned the ‘few things at a time, done properly’ approach that it long had with Ubuntu since the beginning? While desktop linux is still sorely lacking the usability foundations that would enable it to compete with OS X, stability was never one of its problems; unless ‘Beta 1′ represented a codebase quite far from what Canonical is working on for release — sadly this is not the case — Ubuntu 8.04 is probably going to be a feature-rich update, but a huge disappointment in the areas where linux traditionally shines.
I have only tried the KDE 4 flavored kubuntu and I found very enjoyable! However it is quiet buggy, as you point out, but I attributed this to the “will eat your children” state of KDE 4. The fact that it is much faster is due to the 2.6.24 kernel and the much more responsive gnome 2.22. It is a policy of Canonical to release Ubuntu a month after the Gnome release, so I expect that most bugs will be dealt with by then. I am not so sure about pulseaudio. It might be safer to judge a RC.
In my huble opinion the most dissapointing lacking feature is the inability to upgrade directly from the previous LTS 6.06. I cannot see how one will avoid a bad experience going from 6.06 to 8.04 step by step. It is safer and less time consuming to install from scratch.
As far as usability goes, it lies entirely within the view of the user. As a long time windows user I really hated OS X in my first encounters with it in the university. I found ubuntu barely usable ar first, but enjoyable. Nowdays I do not think I will move away from debian for some time.
In my view, KDE 4.0 is not an end-user release. It’s a developer preview release which, even compared to absolutely unfinished OS releases (Mac OS X 10.0 and 10.1 come to mind) seems like a joke for daily use. I haven’t (and wouldn’t) use it as a basis for any criticism of Ubuntu at this stage. Maybe in 2 years it’ll be a proper release (4.2) but right now it’s only a showcase of libraries and a kludge of a desktop shell.
Canonical have strayed from their schedules in the past and they could do so again, if need be. Especially when the release is going to be an LTS, I would think a delay would be in order. I don’t think that given the number and severity of the existing bugs allows for any usable production-grade release in 30 days.
The lack of a straightforward upgrade path from 6.06LTS that you mention is absolutely critical, yet it goes beyond the point of the article; there are hundreds of points that bug me with Ubuntu (no pun intended) that I haven’t touched in this — admittedly very short — post.
Usability is definitely not absolutely subjective. There are numerous methodologies to evaluating the usability of software and there are loads of areas where desktop linux currently lags behind the other major Operating Systems. By many of these accounts, irrespectively of what you think of OS X, it wins hands down. Keep in mind that until 10.4 many of the technologies and APIs in OS X were work in progress, in the same way GNOME and KDE are works in progress. Still, Apple managed to provide a respectable feature to bug ratio to keep most people content.
In any case, at the end of the day, for anyone comfortable with Debian I don’t think there’s any point discussing usability: unfortunately, software conditioning among the power user/developer population is a sufficiently crippling factor in their perception of usable software. If that weren’t the case Linux would probably be the most usable OS around today, free from the fundamental flaws and deficiencies it’s still carrying from yesteryear and that don’t get fixed precisely because a significant part of the linux world (yet a tiny part of the general population) keeps ignoring them.
Now, that I reread my previous comment it does not make much sense. There should be a new paragraph beginning “The fact that *Ubuntu* is much faster”, in the first paragraph.
Also I think I agree with your last paragraph. Linux still is “by geeks, for mostly geeks”. In my opinion this is not a failure. However I do see Linux grabbing a part (most probably the low end) of the OS market in the next few years.
Since you have an “apple biased” open mind about OS’s you might find this article by ers for the switch to 64 bits interesting.
http://catb.org/~esr/writings/world-domination/world-domination-201.html
Also, please do not take the last sentence as parenting, or “advising”, i just thought you might find it interesting, in case you do not know about it.
Well, the last part of this sentence is not at all related to the first. Linux-based solutions are appearing everywhere, and rightfully so. I use and promote them whenever possible. This doesn’t mean that it’s good to just accept the deficiencies of desktop linux solutions and stand still while the rest of the world evolves. This, I fear, is a failure and one of the greatest ones affecting linux today.
Yes, I’ve read it. Amusing, yet not entirely flawed. Esr, like many ‘geeks’ only see the tip of the iceberg when it comes to proposing ’stuff that needs fixing’. In any case, I stopped considering esr ‘interesting’ about 10 years ago and never looked back. For the record: I do not, in any way, shape or form, consider myself biased wrt any company or platform. I’ve used more than 15 operating systems since the 1980s. I use linux and bsds daily, as much as I use Mac OS X (I stopped using Windows about 7 years ago after more than a decade of use).
Yet it would be foolish to ignore the realities of the ageing desktop computing industry and comfortably sit in front of a ‘geek’ OS: Despite my criticism of Apple and its policies (just search for Apple on this blog and read some of the posts here), I have to be objective and acknowledge that given Microsoft’s incapability in creating a secure, modern and efficient OS and the chaos of linux and the ‘pure’ FOSS world, OS X stands as probably the best solution for the vast majority of people out there right now, be they developers or users. Note: This has nothing to do with Apple, ideology, the trends or Macintosh computers per se. It doesn’t, in other words, have anything to do with most of what ESR writes in World Domination 201. I would probably be the first to embrace, develop for and promote an equally solid, consistent and forward-looking FOSS desktop OS if it ever appeared — I still do all three for linux, despite it being a kludge. Sadly neither the organisation, technical foundations or will exists right now within FOSS for anything matching the above description even at the slightest. Hopefully it will.
Well, the last part of this sentence is not at all related to the first.
(I do not really know the allowed html tags in the comments). What I meant to say is that despite linux being “by geeks for geeks” I see it grabing part of the low end of the OS market, such as older PC’s or ultra-mobile laptops when the cost of Vista or XP is a significant part of the laptop value.
About esr, I just found this specific article interesting, even though I disagree with part of it. I understand that you judge things from an entirely different point of view. I hope you did not receive my comment as a “fanboism” critisism.
As “for the chaos of linux and the “pure” FOSS world”, some people see chaos, others see choice. It might be that none of the choices in the FOSS world can compete with the desktop experience of OS X. Certainly the average person would not dare search through the “chaos” to find the software that matches her needs. Configuring it is most probably a nightmare for most people.
I also agree with you that the FOSS community very frequently lacks the focus to produce a forward-looking desktop. In a way though the FOSS community has very different goals from either Apple or Microsoft.
To get back on topic, I see Ubuntu as an extremely interesting project that introduces new people to the linux world. Not all stay. This is not going to be a ground breaking release for Ubuntu, nor the “year of the linux desktop”. I trust Canonical will release a stable and usable LTS, as they did in 2006. Maybe you are right and they will postpone the release date.
This is a flawed, albeit very common argument. You can have choice without the ‘chaos’ that I suggest exists in FOSS. Similarly, you can have ‘chaos’ without choice too. They are independent of one another.
The FOSS community has no concrete goals, and that is part of the magic of FOSS. I suppose with a change in mentality FOSS could become much much better than OS X or any other OS. Sadly, experience teaches that this is very unlikely, exactly because people choose to ignore all that’s wrong with FOSS and keep extolling all that’s right.
More to the point, Ubuntu — and most desktop-oriented distributions — should at the very least try to reach parity with Windows or OS X. More often than not, they claim they’ve exceeded both in features, usability, etc.. This may be true in some areas, but it’s definitely not in most. In any case, Ubuntu, like GNOME are not about ‘revolutionary’ products, they are about ‘evolutionary’ iterations of ever-improving, usable FOSS. I doubt that Hardy is going to be as ‘usable’ as it should be, irrespectively of Canonical’s intentions. Hopefully, I’m wrong and it’ll be a great release or even a late great release perhaps. Thanks for your comments.