Mark Shuttleworth writes on his blog: “I’m of the opinion that Ubuntu could not exist without Debian. So it’s absolutely my intention to see that Ubuntu is a constructive part of the broader Debian landscape. It’s vital that Ubuntu help to sustain and grow Debian, because it’s the breadth and strength of Debian which make up the ‘shoulders of greatness’ on which we in the Ubuntu community stand when we reach for the stars.”
because it’s the breadth and strength of Debian which make up the ‘shoulders of greatness’ …
Oy, I like Ubuntu but their preachy ‘we are the world…’ marketing style is grating. Anyone agree?
Agreed
It’s my opinion that no distribution is much better or worse in the open source world. Technologies will flow from on distribution to another. So their marketing sometimes puts me off. Look at novell or redhat that have been contributing substantially to the OSS.
I still think there leaves much to be desired though. The flow between certain fundamental efforts just isn’t where it could be. I think alot more upstream bug fixing needs it occur on the path of the multitude of derivative distros that exist.
no, i don’t, they are just proud of what they had acomplished.
no, i don’t, they are just proud of what they had acomplished.
Destroy the debian project ?
It either destroys itself or won’t get destroyed by anything. All the rest is whining and laying the blame on somebody else.
Agreed. I like the distro just fine, but the whole “let’s all be shirtless shiny happy people, unite the world, and sit around in a cirlce and sing koombuya” is totally lame.
fortunately they have an option option to change a backdrop for ya
“Oy, I like Ubuntu but their preachy ‘we are the world…’ marketing style is grating. Anyone agree?”
Then just ignore it, like you do with the marketing for all other products you use on a daily basis.
I doesnt bother me personally though.
It doesn’t because I only see it on the CD jacket and front promotional material on their website. If it was embedded all through the wiki, forums, and gui, I’d be sick of it in less than a day.
Clippy says, “I see you keep trying to kill me, but I’m still here. Don’t you want to look up ‘how to kill Clippy’ on MSN Search? You won’t find anything there on the subject, but I’ll show you plenty of maybe-related advertising. But you can pay for support subscription from Microsoft Developer Network and maybe find out how there!”
I guess the level of annoyance might be relative to your exposure.
I enjoyed reading what he had to say. There are writers and then there are good writers. It is obvious he falls in the latter category.
… he discovers Debian. Makes me wonder why? Suddenly a spark of reality? I mean it’s ok as long as it isn’t just for the media. If he _really_ thinks this way it would be a great opportunity for him to start fixing the somewhat tense relation between Ubuntu and Debian. Just for a start.
What do you mean ‘all of a sudden’? He used to develop for Debian. He’s said stuff like this since Ubuntu started. The only people who think Ubuntu doesn’t acknowledge the importance of Debian are those who aren’t paying attention.
START fixing the relationship? The somewhat tense relationship between Ubuntu and Debian has grown up DESPITE his efforts. Shuttleworth genuinely seems to see Ubuntu as merely an offshoot of Debian, whose specialty is adapting Debian into a high quality end-user desktop, and as such wants to give all his modifications and improvements back to the parent system, Debian.
Whether or not all this will actually WORK… that’s the real question, especially if Canonical decides Ubuntu needs to go in a different direction that Debian REALLY doesn’t like, so Ubuntu can be profitable.
He writes well and thinks clearly, but I’m not sure what his conclusions are. Perhaps he is being tactful and polite and keeping his feelings to himself, apart from observing that “a little introspection” would be beneficial for Debian. He praises Sid as the best thing about Debian – it is hard to disagree, imho – but what might this mean for Testing and Stable? If you want to ask “what now for Debian?” you will find Debian praised in this article but it doesn’t really answer the question.
My own worry is that a project the size of Debian needs a very large number of skilled developers. If those developers start to move elsewhere, or fail to join in the first place, then the scope of the project will be at risk and in that event change will pretty well be forced on it.
Sid is development, Testing is the test whether it works for stable or not, Stable is stable. My desktop is Testing, my servers are Stable. And no, they are not Ubuntu – for a very good reason.
You cannot really compare Canonical to the Debian project: The first is a commercial entity that now wants to get some money back from Ubuntu. The latter is a non-commercial project, not a company. Therefore it cannot be run like a company – thank God!
Sorry, what was your reason for not using Ubuntu, then? You hint that it’s ideological, but I’m not really sure. I’m not bashing your choice of Debian any more than Mark is bashing Debian itself; without Debian, there’d be no Ubuntu. It just sounds like you’re a bit anti-Ubuntu, owing to being a bit anti-Canonical, and I can’t figure out why on either point. Since Canonical has promised that Ubuntu will always be free (as in speech and beer), how do you figure that they “now want to get some money back from Ubuntu”?
Canonical is a business, and is pouring money into Ubuntu. Unless they’re just going to spend Shuttleworth’s fortune and then go home, Canonical is going to have to find a source of continued revenue so they can continue to pay the developers, print and mail CDs at no cost to the buyer, etc etc.
He praises Sid as the best thing about Debian – it is hard to disagree, imho – but what might this mean for Testing and Stable?
This is the real point buried within Mark’s slick pandering. He thinks the Debian project should focus on Sid and let third-party distributors produce releases focused on a variety of conflicting objectives. Debian should be the ultimate developer’s playground, where code matters and politics don’t. The burden of producing end-user releases that comply with their community principles distracts from the more essential task of maintaining the world’s most extensive library of free software packages.
In Mark’s view of the world, Debian is not an end-user system. It should be easy to install and configure just about any software to do just about anything, but it doesn’t need to do anything in particular well out-of-the-box. Ubuntu and other projects produce systems that are purpose-built for a variety of usage profiles.
The reason why Ubuntu rubs the Debian community the wrong way is that they don’t want to be an extension of Sid, Testing, or Stable. They want to be essentially an alternate Testing/Stable branch. There is nothing inherently wrong with this approach, and the Debian community should realize that they suck at release management anyway. Debian’s value is in Sid, and they should focus on that.
Many Debian developers would be happier using one of the Ubuntu-based distributions anyway, using a chrooted Sid install for development purposes.
This is the real point buried within Mark’s slick pandering. He thinks the Debian project should focus on Sid […] In Mark’s view of the world, Debian is not an end-user system.
This is also where Mark Shuttleworth is most clearly wrong. Debian is widely used for servers and Debian has constantly shown significant growth on that sector. In fact, I think there are much, much more Debian servers out there than there are Debian desktops. And these servers don’t use Sid, they use Debian stable.
Ubuntu, on the other hand, has yet to show that people trust it as server. So far Ubuntu has “only” been home users’ favourite desktop, showing number of page hits at DistroWatch and some media attention.
http://news.netcraft.com/archives/2005/03/14/fedora_makes_rapid_pro…
http://news.netcraft.com/archives/2005/12/05/strong_growth_for_debi…
You’ve actually been playing the logical equivalent of broken telephone. Mark used SID as an example of one of the many places where Debian does things right. “moleskine” speculated that he might be trying the send a subtle message about the usefulness of SID versus Testing/Stable. You assumed his speculation as fact and stated as fact that he thinks that Debian should give up on Testing/Stable and focus on SID and let third-parties produce “real distros”. You then use this stated “fact” to show why Debian is offended — after all Mark basically said that all the hard stabilization work of Debian’s developers is useless and does not produce and end-user system. Now “da_Chicken” has used your stated “facts” and used statistics to prove the Mark is wrong.
Let’s take a step back and see what Mark actually said. It boils down to this:
* Ubuntu depends on Debian’s health.
* Ubuntu and Debian have different goals
* Ubuntu has a narrowly focused goal and cannot handle the breadth that Debian does without becoming less focused and less good.
* Debian has an extremely broad “global OS” goal and cannot become more focused and tailored without alienating a large portion of the Debian user base and without.
* Ubuntu’s narrow focus benefits from a more hierarchal model that tries to reduce conflicts.
* Debian’s universal focus benefits from an inclusive democratic model that allows conflicts since that allow for common ground to be estabilished for a broad range of conflicting goals. Debian is likely not the best in any one area, but because of its consensus model, it is one of the most balanced that satisfies the widest group of people.
* He praises Debian on a few areas, including SID.
* He hopes Matt will return to work with Debian.
* He thinks Debian would benefit from looking at itself, and presumably make some reforms like use more of Ubuntu, Maemo, and other’s work, but does not wish to say more (from his previous writings, it seems because he does not feel his is qualified to give solid prescriptions).
Okay. That’s a summary of his whole blog entry. Do you find anything anti-Debian in it or anything that does not look reasonable (even if you disagree with it?)
Edited 2006-09-11 13:35
Mark Shuttleworth’s blog entry compares Debian to a “Tibetan Plateau” and Ubuntu to a “cluster of peaks” that depend on the plateau. These metaphors are open to many kinds of interpretations but I’d say that Debian has grown a few peaks of its own, apparently without Mark Shuttleworth noticing it.
When Mark Shuttleworth started Ubuntu, Debian was struggling to put out a stable release (Sarge) that had been delayed many times and also the speed of development in Debian Sid suffered from that. It was a perfect opportunity for a derivative distro like Ubuntu to establish itself as an improved version of Debian (for desktop use at least). Debian (stable) still used the old installer (a set of floppies), XFree86, old versions of almost everything while Ubuntu used the development version of Debian’s new installer, plus it had XOrg and the latest GNOME.
But things are different now. I don’t think that Ubuntu has too many advantages left when compared to Debian’s desktop offering — this applies especially to Debian testing and unstable. So I think that Mark Shuttleworth’s metaphors are not quite as adequate as they would have been a couple of years ago. Debian is not a flat plateau any more — now its peaks reach almost the hights of Ubuntu’s peaks. This means that Ubuntu and Debian have partially overlapping goals and that they can now compete with each other. (Is it possible that Mark Shuttleworth has already realized this and that this might be the actual meaning of the “conflicting goals” that he warns about?) In fact, I believe that this competition could prove out to be more advantageous for Ubuntu than the arrangement where Debian would just stay as a “warehouse” of packages for derivative distributions (or a “plateau”). There is absolutely nothing wrong with a bit of friendly competition. 😉
I see what he means. I had the same thoughts about Gentoo once (before a just gave up).
Gentoo also was about the joy of.. well gentoo
but then the “higher ups” started to worry about things such as QA and releases and a lot of other things that most “users” at the time just didn’t care about (I remember that Daniel dissapeard around this time too, haven’t got a clue if it’s related though). And then it the focus shifted towards beeing a great system “for something” in contrast to just beeing a great system to build and improve upon.
I think there is a real need for a distribution that is nothing else but a play ground for development, play and resarch. I’ve preached about Conary and rPath beeing on to something here with the integration of distributed version control and distributed dependency managment.
Debian, like any institution or product, cannot be all things to all people. It can also not be perfect for one group at the expense of another.
Well, maybe Debian cannot be perfect for every possible group of users but it does a pretty good job in trying to be a very flexible general purpose distro that can be transformed with minimal effort to serve just about any purpose that you can imagine. Debian has a crystal clear goal to be the “universal operating system”, so it doesn’t really need Mark Shuttleworth’s advices in setting its goals.
Ubuntu has a narrower target group — Windows users who want to switch to GNU/Linux — and Ubuntu tries to make Debian easier for this target group by removing choices. Ubuntu offers a pre-selected desktop and a pre-selected set of applications. Debian’s installer gives more choices, at the price of being slightly more difficult. So far Ubuntu has still remained flexible enough to attract also more experienced GNU/Linux users but these are not really Ubuntu’s target group. If Ubuntu could stand on its own without a need to build every release on a snapshot of Debian Sid, Ubuntu would very quickly remove this flexibility because it only confuses Ubuntu’s main target group.
Debian’s installer has “tasks” that users can select when they want to modify the “universal operating system” for some specific purpose with minimal effort. You can select different kinds of server tasks or you can select the desktop task — selecting the laptop task adds laptop support to your Debian installation. These tasks are also improved all the time and they get more refined with every Debian release. I believe that some users will be amazed how polished the desktop task, for instance, will be in the upcoming stable Debian release, called Etch. 😉
a recent survey found that something like 76% of Debian users run Sid, while only something like 6% of Ubuntu users run the equivalent beta code
The survey must be flawed. I think that most Debian installations are servers and then you’d definitely want to use Debian stable, not Sid. Not that Debian would be a bad choice as a desktop — my desktop is Debian testing with some packages from unstable (I set the apt-pinning after an example I found from “man apt_preferences”). Tracking Debian testing allows you to dodge most bugs that appear in Debian Sid, and tracking also Debian Sid with apt-pinning allows you to get some newer packages and the occasional missing dependencies from Sid while most of your system is Debian testing. Oh, and Debian testing is supported by the security team although the Ubuntu website claims that unreleased versions of Debian don’t provide security fixes.
No, I’m sorry I don’t think the following is true:
“Ubuntu has a narrower target group — Windows users who want to switch to GNU/Linux”
First of all, if these Windows users want to switch to GNU/Linux then they’re more likely to want to try other things, including looking at other Linux distributions.
Perhaps you mean Ubuntu tries to appeals to Windows users who may not want to switch. Even this isn’t true. Ubuntu makes it clear: Ubuntu wants to be GNU/Linux for human beings.
Quoting their website: “a clear focus on the user and usability (it should “Just Work”, TM)”. They don’t try to be the universal operating system, but they do depend on the universal operating system, debian.
I used to use nothing but Debian but now I use Ubuntu on my main desktop because things “Just Work” (in the way I want) and it is easier to use. I could, easily, use Debian sid, but, I’d rather not waste time setting stuff up when Ubuntu does things pretty much perfectly anyway. Except sound. I hate sound in Ubuntu (and GNU/Linux in general!)
I still use Debian for some of my machines, and all of my servers – because Debian rocks. Ubuntu rocks too, but it meets my needs better on the desktop.
By the way, I can’t wait for Debian Etch! I love Ubuntu, and I love Debian. I wish that these Debian dev’s and Ubuntu dev’s would stop hating the other side and love both operating systems.
(Oops, did I just become a hippy? :S)
Ubuntu makes it clear: Ubuntu wants to be GNU/Linux for human beings.
Quoting their website: “a clear focus on the user and usability (it should “Just Work”, TM)”. They don’t try to be the universal operating system, but they do depend on the universal operating system, debian.
Don’t you see any contradiction between the statements that Ubuntu wants to be “GNU/Linux for human beings” and that Ubuntu doesn’t want to be “the universal operating system”? “GNU/Linux for human beings” is the kind of description that embraces all possible target audiences and all possible purposes. It doesn’t exclude anything and, in practice, this description becomes synonymous with “the universal operating system”.
But Debian is already “the universal operating system” and it’s also pretty easy to use. How can Ubuntu make Debian even easier (for its narrower target audience — the new GNU/Linux users)? How can Ubuntu make things “Just Work”? Have you thought about this? My perception is that Ubuntu tries to make Debian easier by removing some choices that its main target group rarely needs. This removal of choices makes Ubuntu less flexible and Ubuntu’s pre-selected set of packages are actually hardwired to the meta-package that provides the Ubuntu desktop. Users can of course remove this meta-package if they want. This removal would clean the system from a heavy baggage of extra cruft (like international fonts and laptop stuff that desktop users don’t need) but it would also mean that users have to say goodbye to some Ubuntu goodness (artwork & desktop polish) that comes hardwired with this meta-package.
So, I’m sorry but I think that my point remains valid. Ubuntu’s target group is new GNU/Linux users (even if Ubuntu defines its target group non-exclusively as “human beings”) and Ubuntu tries to make Debian easier by removing confusing choices, which makes Ubuntu less flexible than Debian. Luckily for the more experienced users, Ubuntu cannot get rid of much of Debian’s flexibility because every Ubuntu release is built on top of a snapshot of Debian Sid. But if Ubuntu was to detach itself from Debian and stand on its own, Ubuntu would very quickly become so inflexible that it would drive more experienced GNU/Linux users away because experienced users aren’t really Ubuntu’s target group.
Yes Debian is universal operating system, like Ubuntu. But Ubuntu main goal is simplicity and it does it well. So if you want desktop platform that “just works” choose Ubuntu, if you want more choises choose Debian or Ubuntu server.
P.S. Yes, what you can do in Debian you can do it in Ubuntu, it depends on your knowledge.
Edited 2006-09-10 10:34
I think the problem lies in defining the term “GNU/Linux for human beings”. You define it as a “kind of description that embraces all possible target audiences and all possible purposes”.
Well, this (your) definition is not the definition Mark Shuttleworth has in mind.
You are right, however, when you say that Ubuntu’s main audience is new GNU/Linux users. And that’s what the “human being” part means by Ubuntu’s definition. Computer users with few (or even none) experience using computers. Not the seasoned hackers, not the enthusiastic geek crowd, but the simple and unexperienced “human beings” out there who have never gotten into programming or electronics.
Ubuntu doesn’t want to be a “universal operating system”. Mark and his fellas never intended Ubuntu to be a Universal operating system. Ubuntu’s intentions are not to cover (as you define a “universal operating system”) “all possible target audiences and all possible purposes”.
You are also right in calling Debian GNU/Linux a “universal operating system.” Heck, in my opinion Debian is _the_ universal operating system. Debian is my first choice for embedded systems and servers. And if I’ll have to manage a corporate desktop environment ever again, Debian will my fist (and almost only) candidate.
But since Debian is a “universal operating system”, it’s also a “jack of all trades”. And when it comes to the home desktop, Ubuntu delivers, exactly because of it’s focus, because of it’s limitations, a better user experience.
Developers and hackers will, of course, choose GNU/Debian over Ubuntu on they desktop systems, since Debian offers exactly the thing they need (a full fledged swiss army knife) whereas Ubuntu would maybe only get into their way.
To sum it up:
– While Ubuntu calls itself a “GNU/Linux for human being” that doesn’t imply they want to cover every possible purpose or/and all possible target audiences.
– Debian is a (maybe _the_) universal operating system.
– Ubuntu is not, it focus is to be easy to use and easy to manage. It’s a very good choice for the hassle-free home desktop.
– Experienced users prefer Debian for it’s fexibility.
Edited 2006-09-10 10:45
that’s what the “human being” part means by Ubuntu’s definition. Computer users with few (or even none) experience using computers. Not the seasoned hackers, not the enthusiastic geek crowd, but the simple and unexperienced “human beings” out there who have never gotten into programming or electronics.
Eh, seasoned hackers and enthusiastic geek crowd are not human beings??? That’s kind of a rude thing to say. 😉
– Experienced users prefer Debian for it’s fexibility.
I started with Linux in 1999 and my first foray was Redhat in school… At home, I played with Debian “potato” released in 2000. After having some problems and not finding anything in the manual pages or the documentation, I asked the Debian mailinglist and intelligent question. The responses I got were childish and immature flames from elitest pricks. That jaded me permanently from the “debian” experience.
Never again did I touch a debian based distribution until the release of Ubuntu Hoary Hedgehog and simply fell in love. I don’t want to spend hours tweaking things, I just want them to, “Just Work TM”. As a poweruser, Ubuntu makes a perfect desktop. CentOS makes a perfect server for me.
Keep in mind that I am (by trade), a Linux systems admin and I write things like this:
http://www.digitalprognosis.com/opensource/faster-dapper.sh.txt
Do yourself a favor. Next time your hunting for a sysadmin job, make sure you hide that script and disavow all knowledge of it’s existence. Your not going to impress anybody with a tweak script that sets NOPASSWD for sudo and turns your Ubuntu box into an indecure piece of shit like Linspire.
Well if you could actually read it, you would notice that it sets it back at the end of the script 🙂
Maybe you should try to understand things before you blindly flame them. sudo caches credentials for a small period so you don’t have to enter the password every single time. That script does sudo apt-get dist-upgrade which can take a very long time on slow connections.
Maybe you should get a clue…
1. You’re an idiot, because no it doesn’t. Check your f–king code.
2. You’re an idiot because it wouldn’t even be necessary. Once you initiate the first /bin/sh using sudo scriptname everything spawned from that shell, ie. the remainder of the script, runs as root.
1. You’re an idiot, because no it doesn’t. Check your f–king code.
Before you make yourself look more ignorant than you have ALREADY PROVEN, take a look at line 120. If you are blind, it states this:
sudo sed -ie ‘/NOPASSWD/s/NOPASSWD: //’ /etc/sudoers
For those of you that don’t understand this, this takes the NOPASSWD statement and replaces it entirely with null. No, I didn’t all of the sudden change the file on my site because of your statements…
2. You’re an idiot because it wouldn’t even be necessary. Once you initiate the first /bin/sh using sudo scriptname everything spawned from that shell, ie. the remainder of the script, runs as root.
This wasn’t supposed to be run with sudo. The script runs sudo by itsself. Being an elitest prick doesn’t help anything or anyone.
Notice by reading my response to #1 that you were indeed wrong. You were wrong, and that is a fact. End of story.
If someone kills your script (I checked, they can) midway through execution, won’t the system be left with sudo set to ‘nopasswd’?
Anyway, if your script requires the user to give a password for sudo the first time (when the command setting to NOPASSWD is run), why not just have THEM run your script as sudo, and run everything assuming you’re root?
I didn’t test it, but I’m pretty sure sudo would be inherited by all processes spawned by your script, so you wouldn’t have to worry about sudo timing out and sitting at a prompt until someone notices. Correct me if I’m wrong.
I mean, yes this would cause problems if the user forgot to run the script as ‘sudo’, but you could check that quickly by trying to create (and then remove) a file in, say, /etc; if that fails you exit with an error saying the script must be run with sudo.
I would also log completion messages AFTER you’ve successfully changed the setting. I’m assuming that the logging command is less likely to fail than the configuring command…
Ubuntu is a very nice distro for now. But if you are trying to hit the enterprice then you need a lot more then a warpjump to distrowatch. Debian has proven itself. I’m not so sure for ubuntu …
I agree. They need, I think, to focus more on central manageability in addition to “It just work”ing. I hope thats what I see developing with Edubuntu
I have to agree. But for desktop, it is the best distro i’ve ever used for a long time.
Simple to use, light to update, and less hellish than rpm based distros.
For the last 4 years. Development and progress of tools in the dev world demands me keeping more current.
Graphic Design works also requires me to use Sid and since I can manage my own system from custom kernels to custom debs built for my needs to managing my own systems I don’t need Stable on my system.
More to the point, doing Cocoon 2 development work is not a reality in Stable unless I break a bunch of it.
Stating that one distribution is no better or worse then another is a complete disregard of reality. At a base level, all systems are a Linux kernel + gnu/open source tools. Sometimes proprietary apps are added as well. What make distributions stand out are installation ease, support availability, commercial application support, and integration between applications (just to name a few). Try showing a non-technical person the installation and configuration of Ubuntu verses Gentoo, there is a significant difference.
Strange, people were always screaming he wasn’t giving debian enough credits for their work, even when he does he gets criticised for not doing so. I think he’s rather fair to debian and explains why he didn’t become part of it and started his own distro.
The reason why he chose to base Ubuntu on Debian is because Debian has already done 90% of the work. If Debian is weak, that would mean he would now have to do all the work and pay through the nose for it too. Mark is a business man and he knows very well that a strong Debian means a strong Ubuntu.
What you say is true, but it actually goes beyond this.
Imagine if tomorrow all Debian programmers and users gave up Debian and moved to Ubuntu and imagine that the Ubuntu infastructure was able to absorb them all quickly (virtually impossible, but let’s suppose its the case).
The first thing that would happen is that all the diversity of Debian would try to assert itself within Ubuntu. Ubuntu could no longer keep its laser sharp focus. Schedules would slip. Conflicts would arise. The friendliness of the Ubuntu forums would diminish. In short, it would be very bad for Ubuntu.
Personally, I think there are some things that Debian can learn from Ubuntu (e.g. making SID releases every 2 months the concept of making every nth release of SID the base for Debian testing, then Stable) and there are some reforms unrelated to Ubuntu that would benefit Debian, but Debian has one thing that Ubuntu can only dream of achieving. It was with Linux since almost the beginning and survived several changes in focus and crises. Think of how different the computer world was in 1993. Debian endured where countless other Linux distros were abandoned and has grown stronger with time. I don’t know if the next 10 years will change as dramatically as the last, but if Ubuntu is around then (and I strongly believe it will be), it will be because of its Debian heritage (which will undoubtably still around).
Personally, I think there are some things that Debian can learn from Ubuntu (e.g. making SID releases every 2 months the concept of making every nth release of SID the base for Debian testing, then Stable)
Debian testing is usually less than a month behind Debian Sid. I’d also like Debian to adopt from Ubuntu the practice of making frequent “alpha” releases, snapshots of the development branch. This would give people a better idea where Debian’s development currently proceeds and it would banish the idea that Debian has older packages than Ubuntu — you can easily get this idea if you only look at the stable Debian releases.
But I wouldn’t base these “alpha” releases on Debian Sid, I would base them on Debian testing. AFAIK, the “testing” branch was introduced because Sid tends to be quite broken for some time after every stable Debian release (also Ubuntu’s development branch is usually quite broken immediately after a stable Ubuntu release). Debian testing should ideally be ready for use at all times, which should make it a good basis for “alpha” snapshot releases.
Unfortunately also Debian testing is a moving target and it has occasionally missing dependencies but these can usually be resolved by apt-pinning both Debian testing and Sid so that most of the system stays testing while the missing depends are fetched from Sid. (See “man apt_preferences” for more information on apt-pinning.) Apt-pinning testing and Sid also allows users to easily update some select packages to Sid versions if they want to. This flexibility of mixing packages from different development branches is one of the features that is available in Debian but not in Ubuntu. Of course, this might be too confusing to less experienced users who seem to be Ubuntu’s main target group. 😉
I think you’ve got it somewhat skewed. Mark Shuttleworth was a Debian developer (and I think recently reactivated his developership in order to vote in their elections, not that that actually works in his favor) who wanted to bring Linux to the teeming masses.
So, what distribution would he base it on? Well, he was a Debian developer so he not only likes Debian enough to work for it, but he also knows the system (pieces at least) very well. It’s actually fairly logical that he a.) would base his new distribution on Debian and b.) probably not try to bite the hand that fed/feeds him, so to speak.
This alone shouldn’t really be a problem, I mean, Knoppix is made by a Debian Developer named Klaus Knopper, and other helpers. That said, Knoppix is (at least at the moment) not in the spotlight, and Knoppix was never intended to do all the things Debian can do.
That said, I can see why Debian users are particularly suspicious and distrustful of Ubuntu- Debian has very well-defined and highly political stances on free software, even at odds with the GNU people; now there’s Ubuntu taking their work in a big flashy way, and trying to develop for them when Ubuntu clearly doesn’t follow Debian’s strict structure and rules. (essentially, trying to do Debian’s work for Debian without having to follow, and therefore rendering meaningless, all of Debian’s organization)
The problem is basically politics.
(I tend to wonder whether the problem is solely Ubuntu seeming to take credit for Debian’s work, or Ubuntu attempting to become Debian’s new development process too.)
And before you ask, I’m pretty it would violate the GPL to require that anyone who uses Debian’s stuff play by their rules and organization.
> Quoting their website: “a clear focus on the user
> and usability (it should “Just Work”, TM)”. They
> don’t try to be the universal operating system, but
> they do depend on the universal operating system,
> debian.
I agree completely.
This is why I use Ubuntu. For my home computer, I like to fiddle around sometimes but I don’t want to *have* to fiddle around to get a shiny desktop setup – I want that to just work. I’ve not used Debian, I always used RedHat (derivatives) or SuSE before. But Ubuntu looked like a good compromise.
My day job involves kernel hacking and I haven’t used Windows on my desktop in years, so I don’t really *need* things to be simple – I just like to leave the complicated stuff at work 🙂
Of course, Debian probably suits my needs almost as well! However, I was attracted by the desktop focus of Ubuntu, and the frequent releases with new versions of my most important packages.
Shuttleworth has never been out to get Debian, he’s gone overboard trying to play nice ever since the very start. Compared to any of the other children Debian distributions, Ubuntu has been the most reciprocal by far yet has seen the most wrath from annoyed Debian users. Why? Because Ubuntu has been so incredibly successful, stealing away users and Debian developers in a way that Linspire, Xandros and the like never did.
In a way, Ubuntu is becoming a victim to its own success, because as Shuttleworth rightly points out his distribution relies on a strong Debian to continue to exist. If the folks at Debian cannot keep their own house in order and lose developers like mjg59 because of internal political fights, Ubuntu loses too.
Shuttleworth is now in the unenviable position of having to try shepard the Debian community in addition to his company’s own community. A thankless job if there ever was one, it’s amazing what a benefactor he has been to the community despite the bile that gets constantly thrown in his direction.
Ubuntu and Debian in a bit of a spat. We may be making the slow march towards Slackware after all! 😀
kinda off topic but ubuntu related…as ubuntu comes with gnome by default,im trying to get used to it without using apt-get to install other gui’s…whats driving me nuts about gnome is the menu highlite text the comes on when you hover over a menu item…is there a way to turn this annoying text off?
I don’t think the problem is really about politics, but about (understandable) jealousy. It was never Debian’s goal to “achieve world domination” and now that a Debian fork seems to be going in that direction, it’s too bad, from some Debian people’s point of view, that the Debian name and signature isn’t on it. After all, there is no issue between Debian and Linspire or Xandros, for example.
I’d be jealous too, I think, if I were a Debian developer.
So, the GPL is not meant for jealous people. Debian is a great project, but in a world where the Linux hype is coming from new users embracing easy-to-install distro’s, it is simply a result of Debian’s lack of interest in making a simple, easy and powerful, Suse-style graphical installer (and no questions about screen resolution, please
, that Debian can’t profit from Linux desktop growth. The only way to change that for Debian is to embrace such an installer after all. I think they’re working on that, so they might try to launch a little hype around it.
If that’s really what they want.
And I’m sorry, I just can’t find anything evil here in Shuttleworth’s attitude.
And I think Ubuntu was USING Debian’s new installer (currently in beta)… there was an article on that recently.
Debian is most perfect and soooo mature. I believe now it’s Ubuntu time, IMO because Ubuntu has highest growth-stage rate than any other distro ever.
Maybe debian is too old and overwheighted, huh? 😉
[Pardon me, if my english sucks]
So does forking peoples projects that are still actively being worked on which is something apparent to almost everyone but Shuttleworth.
If only they would let me give more than one +1 to a post. That is the truest comment so far.
“So does forking peoples projects that are still actively being worked on which is something apparent to almost everyone but Shuttleworth.”
I guess it’s not apparent to all those other Debian-based distributions either. Or the Debian project itself, since as it says on their website: “Debian welcomes and encourages organisations that want to develop new distributions based on Debian.”
http://www.debian.org/misc/children-distros
You make an interesting point there, of all those distros based on debian there seems to be animosity only on this one fork. How odd.
In my opinion, Ubunto achieved objectives that Debian never did, getting windows users to try a linux platform and give them a chance to get into the open source world
The Idea of saying that Ubunto ows Debian a favor is a completly nonsence, in my opinion, is debian that owns a favor, since it was Ubunto that got all those ppl to sit and try it… Ubunto gave a chance that Debian never did…
Today ubuntu, tomorrow kubuntu, guntu the next day and so on…good future of Linux…yeah right
People are like cows…they really need someone to guide them in right direction or they go and do their own shyt…which is what mostly happening around 100s of Linux distros…
I stopped using Ubuntu because of it’s reliance on another distribution. This article summons my reason: Ubuntu is beginning to hurt Debian with overwhelming popularity. Yet Ubuntu uses Debian as its package base and without it, Ubuntu could not survive without major developmental changes. Its imperative that Ubuntu assist Debian everyway it can for their own sake.
Now I have a strict policy of using self-sufficient distributions that rely on themseleves for new and updated packages.
Edited 2006-09-10 17:05
You just can’t have a serious discussion regarding Ubuntu or Linspire without the mod trolls rushing in. f–king ridiculous.
Naturally, it’s a great thing if Ubuntu and Debian want to improve their relationship. It’s not going to be that simple to get good code flow due to rather different developer groups, but it’ll sure work better if they try.
On the other hand, I can’t really see what people are talking about when they say that Ubuntu owes Debian this support. What code flow-back has Debian ever gotten from any of its myriad other child distros? Xandros, Linspire, and the dozens of others on Distrowatch? Very little, I should think. Ubuntu has had great success, both in market and mind share. That is strange, because they haven’t actually done that much: Mostly they stabilized a little, polished a lot and wrapped everything up nice and shiny. The reason I use Ubuntu on my desktop is because it’s very smooth, coherent and intuitive from the very beginning, with almost zero customization needed to get a comfortable workspace. At the same time, I realize that it’s 98% Debian code I am running. I could also use Debian; in fact, I have.
But there is a difference between “could use if I had to” and “would use gladly”. Ubuntu achieves the latter by effort; Debian, for all its praisworthy development work, has never much exceeded the former. IMHO, the Debian project cares about robustness, stability, power and software availability. What it doesn’t care about, is making things nice to use for people who’re not into Linux. They certainly haven’t accomplished it in years, and if they cared, I’m certain they would have. For anybody who says Debian is soooo easy to use, try selling it to a your mother or some not-very-enthusiastic Windows user, and having them use it without constant hand holding.
>> Ubuntu has had great success, both in market and mind share. That is strange, because they haven’t actually done that much: Mostly they stabilized a little, polished a lot and wrapped everything up nice and shiny. The reason I use Ubuntu on my desktop is because it’s very smooth, coherent and intuitive from the very beginning, with almost zero customization needed to get a comfortable workspace. <<
So what do you mean, “they haven’t actually done that much”? Your comment indicates quite the opposite. If Ubuntu hadn’t done that much, I’ll launch my own distro next week.