Ubuntu’s community manager Jono Bacon talks in an interview with derStandard.at about the relationship between Ubuntu and GNOME, GNOME Shell, Unity and why the netbook market is that important to Canonical.
Ubuntu’s community manager Jono Bacon talks in an interview with derStandard.at about the relationship between Ubuntu and GNOME, GNOME Shell, Unity and why the netbook market is that important to Canonical.
Working with GNOME does not mean “Here, we developed XY. If you want it, pick it up yourself from Launchpad.”
Working with GNOME means:
1.) Put your ideas up for discussion. Are the proposed additions welcome? If no, can the differences be resolved? Likely the differences are minror.
2.) Develop you stuff in GNOME’s git repo. Surely there is some kind of “playground” directory where prototype code is developed.
3.) If your code is permanently rejected, you can still put it up on Launchpad.
Wrt Unity: Was the GNOME team even asked if there is any interest in a netbook shell? I’m not aware of that discussion having taken place.
I’d guess there is interest, but the core team is mostly occupied with GNOME-Shell (which, btw, was proposed by Red Hat pretty much like I described above).
Interestingly while Canonical is hardly known for being a KDE supporter, the two Kubuntu guys do a better job at working with upstream (though it’s not really perfect). While their cooperative work could still be better (IMO their libdbusmenu-qt should be hosted in KDE Support, not as separate project on Gitorious), but at least the D-Bus Menu work was actively put upstream to KDE by them.
Granted, their standing is easier, because they built on the ‘systray icons via D-Bus’ foundation already laid by KDE, so their work was more evolutionary there and not like Canonical’s GNOME teams approach “Hey GNOME, we replaced your systray implementation with KDE’s protocol. Yeah, we designed it to be completely incompatible with GNOME-Shell, but you can pick it up at Launchpad nonetheless if you want to and port it to GNOME-Shell yourself so we don’t have to do the work in a year or so.”
Actually, I think Canonical is doing the right thing, there is a lot of burocracy in the way upstream manage things, and even some projects have a personal dictator with the mantality of “If you bend over enought I may agree” (cough KDE), so, what Canonical is doing is “We are making this changes, here they are if you want to merge them lets talk about it, but I won’t stop because of you”. And to me that’s a very good strategy. Good for Canonical.
Edited 2010-08-09 17:00 UTC
Its a very good strategy if you are writing a large portion of the code yourselves. When you’re just leeching off what the community does, it makes you look like a jerk. The Canonical guys have every right to do there own thing. But then again, the Gnome guys have the right to just ignore everything coming out of Canonical. Its a two way street. And the Gnome guys are driving the bigger vehicle.
Canonical ain’t doing nothing inmoral, unethic or out of the law, the only people who I see complaining about are a) a bunch of childs full of envy who hate Ubuntu because is not their distro of choice or because is not using the desktop they like as the primary choise, b) they are just jerks. So that doesn’t worries me at all.
And what about upstream projects that would rather spend their time developing more code rather than hunting down bits and pieces from repos around the web?
What do you think would happen if all contribute that way? I give you the answer: Everything would break down, because patch XY no longer applies after patch YZ is applied etc.
Old SuSE (pre Novell) and many other 1990s distributors learned that lesson already. They patched the heck out of pretty much any software, just to see it break down, leaving bug fixes unbackportable etc.
Alas, if you don’t pay careful attention to upstream projects and ensure that they get the feedback they need regarding how you’re shipping their software in your distribution the software suffers and it’s a downward spiral – for it and yur distribution.
Patching the living crap out of a piece of software and backporting all and sundry is a strategy that other distrbutors learned not to do if at all possible a long time ago, and it should be painfully obvious by now that Canonical doesn’t have the expertise to backport shit since they aren’t doing any of the work so they’re unlikely to actually understand what they’re patching.
Quite why we should all be full of envy for Ubuntu or why people should be jerks for trying to point out why the quality of successive Ubuntu versions is going down to Canonical I have no idea. All Canonical are doing is giving the chain a final yank before desktop Linux in any form gets flushed down the toilet and someone starts again when Canonical goes bust.
Edited 2010-08-10 16:13 UTC
Ubuntu only had batch of 2 crappy releases (Jaunty & Karmic). Lucid is good.
I found Gutsy to be the worst in terms of screw-up. Remember the flash fuck-up. I don’t say that Ubuntu devs are responsible, but looking from the end user perspective.
Those people are full of envy not because the way Ubuntu is doing things, they are full of envy because is very popular, attack the way is doing is just a poor excuse to give it bad press, but I don’t think they will succed.
Edited 2010-08-11 15:15 UTC
Your flaming against KDE is not surprising, given how much you hate it. Considering that Canonical’s cooperation with KDE is much more smooth than with GNOME (and I already wrote that), your argument is invalid right from the start.
As I already wrote, additions/modifications have to be made in cooperation with upstream projects if you want to be recognized as an organization that works with upstream projects. Red Hat, Novell, Mandriva, etc. all work that way. Novell, for example, is developing a netbook GUI for Evolution right in GNOME’s mainline git repo. There is no logical reason why Canonical can’t.
If talks fail, there is still the option to develop the code in an independent repository. Hunting down code fragments from various repos is not something that’s beneficial to anybody unless Canonical is obfuscating the situation on purpose to make it harder for competing distributions to pick those modifications up.
That’s your opinion, not mine, I won’t go deeper into it would just be start another flame that nobody needs. You don’t like how Canonical is doing it then more power to you, I think is doing fine and great and many more like it too. And Im glad Canonical is doing it the way is doing it.
Canonical are not very popular in other areas of the Linux world.
This is because of their laggardly co-operation with other distros in sharing bugs etc.
Before you flame me, please hear me out. I forsee the time not all that far waway when Canonical fork’s the whole damm lot and goes their merry way without the other distros.
It is not all that hard to do. All they need to do is to stop all contribution back to Git etc.
Then they can go their own sweet way and develop their own GUI that will scale from Netbook upwards. They will be free of having to confirm to the Gnome committee etc. I sense a real frustration in the Ubuntu camp with Gnome. IMHO, it is only their bigger dislike of KDE that stops them from switching like SUSE is.
Sure their mods will be available free to all other distros but they will have to searching for them rather then being told about them.
Personally, I prefer Xubuntu rather than the vanilla flavoured version on my Netbook. But my main frustration with Canonical is with the quality of their releases. The last few have IMHO been getting worse each time. 10.04 is a real dogs breakfast when compared to say 8.04. But that is another story.
Alas, you have no evidence to back up what you say or your ‘opinion’. I’m absolutely delighted for you that you think things are fine. You’re not saying anything.
Canonical is already pretty unpopular beyond the desktop world for exactly the strategy you think is ‘fine’, so trying to frame this as some sort of KDE flamewar thing is a poor attempt at palming it off. We’re not that stupid.
Not only are they pissing off upstream projects by applying patches and doing things that result in bugs that they have to look into, to add insult to injury Canonical contribute little if any code so don’t understand the patches they apply in the first place. Would that make me want to pay them for support in place of someone like Red Hat? Let me think………. Errrr, no. That’s why they don’t have much of a future.
Really? Tell me who are they pissing, because I’ve barely seen anyone complaining but a couple of trolls and an ex-redhat employee who took their complaines back. But of course, extrapolating it is just another low strategy from the trolls.
Edited 2010-08-11 15:23 UTC
When did applying patches downstream become bad? Everyone is allowed to take the upstream code, change it to fit their needs/vision and release it. Last I checked from Planet Earth that FOSS movement was born with the vision of changing, patching and releasing code to fit one’s need.
I think someone was right when they said this is like the “new school” vs the “old school” because not submitting patches upstream was ALWAYS bad. I’m not sure where some people are getting the idea that it wasn’t. Just because the license allows you to do whatever you want with the code doesn’t mean it’s good FOSS etiquette. The whole point of FOSS is to SHARE. Sure you can take whatever Ubuntu does and use it elsewhere but they don’t make it easy for anyone else when it’s not committed upstream. How are you supposed to take advantage of these changes when upstream breaks compatibilty with the changes because they were more of a hack than a real solution? There are exeptions but what this isn’t it.
Edited 2010-08-14 13:39 UTC
Things aside, read the guy’s comment on which I commented. He was acting as if applying patches was wrong. Modifying code was wrong.
Nowhere in my comment did I say that Ubuntu/Canonical should *NOT* submit patches upstream. I was just replying to his comment statement “Not only are they pissing off upstream projects by applying patches”
Please don’t come to your own cooked up conclusion about anyone from a comment.
Maybe you should read the whole thread to understand the context. He wasn’t implying that applying any patches or modifiying code in general was a bad thing. You are the one who came up with an invalid conclusion from a comment because you didn’t read and/or understand the entire thread.
I have read the whole thread. I was just replying to a comment and not to the whole thread. I did not come to an invalid conclusion. His points were pretty straight forwards and didn’t required any special pre-requisites for understanding. Or he needs to put forward his point properly.
BTW I did read and understand the whole thread. I am not obliged to answer to the OP. I just replied to one comment when I saw that the guy has written just anything which came to his mind. Or probably he didn’t write what he wanted.
KTHXBYE
Just to clarify some things: it wasn’t GNOME protocol that was replaced by a KDE protocol: it was an old specification, that was agreed upon around 10 years ago and while it was ok-ish enough back then it did really not cut it anymore.
So we designed a new one, and pretty transparently since every revision of the protocol specificaion was posted on freedesktop mailing lists.
Then Canonical ran into the same limitation for what they wanted to do for the system tray and luckily cooperation and communication went well, so they adopted our protocol (some modifications to it come from them even)
The main issue of “being incompatible with the GNOME Shell” I think is right now the current client implementation on GNOME is a panel applet, so that won’t work on it, but i think nothing forbids implementing the same thing on whatever level of GNOME shell is needed.
I think if they still want to have a systray at all they will quickly run in the same problems we had, so yes I think cooperation with upstream GNOME is still possible even if it started not too well
The inverse happened for the dbus menu specification, that is the “other half” of the systray spec. we needed something like that and they happened to work just on that, so we adopted it, after the usual back and forth feedback on the spec.
This cross pollination on *both* senses is the way to go and I would like to see it happening more.
In summary, “avoid politics, get shit done”.
Seems to work for them.
Separatism is also a form of politics.
As it’s all Free Software, that’s their right (legally as well as morally), but to “get shit done” (as you put it) Canonical needs to have the work force to sustain their separatist attitude which Canonical has not. So far Canonical’s GNOME modifications are small — hardly anything that qualifies as “get shit done”.
Even larger companies failed at separatism: Novell wanted XGL and Compiz to be its pet projects. Result: The community spearheaded by Red Hat ported Compiz to AIGLX, Compiz was forked several times, stagnated for quite some time, and only barely survived (XGL OTOH is dead).
Apple didn’t open the development of WebKit just for humanitarian reasons. Given good management open development vastly improves software.
Neither Chrome/Chromium, nor Safari could achieve their current status without collaborating with the competing company on the same code base.
You are assuming that if you were not doing your project separately, you would have other guys helping you out (instead of slowing you down).
I don’t believe this would happen.
Practical example: Canonical rewrote GNOME’s systray to work through KDE’s D-Bus-based protocol.
If that work was done upstream, all developers of systray-using GNOME applications (be it 1st party or 3rd party) had the motivation to target the biggest possible audience with the improved systray support: All GNOME users no matter which distro they use.
But why would a GNOME app developer target a specification that works only on Ubuntu and not Fedora, not Debian, not openSUSE, not Mandriva, and nowhere else?
This means either Canonical ports the apps by themselves, requiring them to maintain a patchset through all updates of that app or they hope the app authors will maintain two entirely different implementations for a stupid tray icon or neither Canonical nor the app author care about the other protocol in that specific app and leave the users with avoidable inconsistency.
Edited 2010-08-09 21:37 UTC
Forking an opensource project is both legally and morally recommended. That’s precisely what opensource is for.
Ubuntu’s repositories are open to anyone, including Gnome or Fedora. That wasn’t exactly the case with Compiz or WebKit, so your comparison is a somewhat unfounded.
Vivainio is right that sometimes, especially if you want to shape the project to match your vision, you simply have to sit down and do the work yourself. Doing this directly in Gnome repositories would result in endless flamewars and the code would likely be rejected anyway (as not matching someone else’s vision). If Ubuntu wants to remain competitive and lead the Linux desktop world in terms of usability they simply have no choice – they have to continue to do more work in house than other guys do.
As long as the continue to do their work in the open and deliver better desktop than others do, I have no objections to their tactics whatsoever.
I think you have a very distorted view of what is really going on here.
What Cannonical is doing is surely legally possible. Not recommended. Forking never is – you only do it if the original project is really screwing up like in the case of Xorg forking from Xfree86. GNOME is not screwing up.
FOSS doesn’t progress by forking – it progresses by cooperation. You see this in the linux kernel project a lot – companies or individuals who develop something great, while disconnected from the community, barely ever manage to get it into Linux. Companies and individuals smart enough to work with the kernel community from the start have no problem getting even experimental versions of their stuff in the kernel. Compare ReiserFS4 vs Btrfs, the various virtualisation solutions (Xen/KVM), schedulers (CFS vs BFS) etc etc.
And often, like with the linux kernel, once companies/individuals try to submit what they have developed on their own to upstream, it gets rejected. Why oh why would that be, are these projects (Linux, GNOME, whatever) so stupid and shortsighted and NIH? No, it is usually very simple: the proposed addition simply SUCKS. Like 95% of Canonical’s much heralded “papercuts” – only a few of them were deemed good enough for upstream, the rest was just horrible, hacky ways AROUND the problem instead of proper fixes. And thus unacceptable. Typical of code developed disconnected from upstream.
Oh and I’m not saying it is EASY to work with upstream. It takes balls, it takes perceverence, persistance, stubbornness, intelligence, social skills and more. Not everyone has what it takes. And it’s easier in some communities than in others. The kernel ain’t the easies t (by far). I’ve heard GNOME ain’t that simple either, so canonical being more successful getting their work in KDE than in GNOME (despite less resources for KDE) might have more to do with less NIH and politics on the KDE side than with the amount of effort from Canonical.
Anyway, despite it being hard, you have to work the FOSS way if you want long term viability. I think what we’re seeing here is stupidity or arrogance from Canonical. Either will get them in trouble, but that’s their problem.
FOSS licenses were explicitly designed to allow the users to reuse the code or fork the whole project for any reason without asking the original author for permission.
That’s because sharing and reusing the code is a central idea of the whole FOSS movement. You won’t find a notion of “our repository” in the GPL license – it would conflict with code reuse.
Yes, it may result in fragmentation if there are too many versions of the same code floating around but it also gives the users protection, it assures them that the original author will listen to the voice of all stakeholders. In FOSS “control” of the project is conditional and anyone is allowed to challenge the original author.
or:
FOSS doesn’t progress by following the consensus – it progresses by competition.
Both statements are equally true and it is up to the users to resolve, which of them is more applicable in a given situation.
While the basis of all Free Software licenses is the ability to fork, forks are nowhere recommended as you claim.
Forks are OK if there are fundamental, unrevolvable differences between two contributing parties. I don’t see Canonical and GNOME in that position unless Canonical’s Launchpad fixation counts.
GPL (or MIT, or BSD) license says nothing about “fundamental” or “unrevolvable” differences, quite the opposite – you are free to use the code in almost any way you like.
I get your point, though – we all love and want harmony and peace. Unfortunately, there is just as much harmony and peace in the FOSS world as it is in a bazaar (and it is so by design).
The only party that decides whether it is worth forking the project or not is the user. He has to come up with his own resources needed for maintaining the fork, watch for compatibility issues with other branches, promote the branch and so on. It isn’t a lighthearted decision.
The alternative is not to fork the code but then the user has to spend time advocating for inclusion of his patches and risk that if some critical ones don’t get through the whole design falls apart. And certain “big” projects like Gnome or KDE are rather inflexible when you want to mess up around their core components.
For a company like Canonical, that wants to differentiate itself by changing some fundamental concepts of the desktop, it is not that surprising that they chose to develop some pieces of the software in house. But as long as they stick to FOSS licenses, so that anyone can reuse or fork their code, I’m OK with it.
No license says anything about the development model, but you claim they do.
Oh I don’t blame them, except when they claim to improve the ‘linux desktop’. Because the Linux desktop as a whole does NOT benefit from their work – just like the Linux kernel doesn’t benefit from most of the work the android engineers do on it – it doesn’t get upstream. Luckily that is something they are changing, I hope Canonical can change it too, and we can all finally benefit from their work (paid by both Mark and the Canonical customers).
Surely they don’t have to, it’s their right to go off in their own little cocoon. But they should’t claim they are good FOSS players, an open community which wants to bring FOSS to a higher level etc etc – they are just doing themselves a service, not anyone else. I would even claim they do FOSS in general a disservice – everyone who pays them is paying only for the direct service they get – something which, in the long run, does not contribute to the state of the Free Desktop.
If you pay Novell or Red Hat for a server, you DO contribute to the future of the Free Desktop, because these two companies have strong policies on bringing most of their work back into upstream. For others to benefit from.
Canonical, again, does NOT put much effort there, and thus is like now-gone companies like ‘linspire’, ‘xandros’ and such. They received money, put it in their products and shareholders. Their products were used for a while, then discarded. The code bitrots somewhere – nobody ever benefits from it. Waste.
So I would claim that paying Canonical for anything is about as beneficial for FOSS as paying Apple or Oracle. They don’t hurt FOSS much, but don’t do much good either. Tell me why I’m wrong, it has happened before 😉
Note this is a very personal opinion and certainly not condoned by any employer, community, friend, monkey, chair or who/whatever else…
My “whole” 😉 Linux desktop (which I use for almost 15 years now) has benefited from Ubuntu’s work. It’s almost as trustworthy as Slackware which I used first (everything was so simple then) or RedHat (before they spun off Fedora). Everything else I touched later (Fedora, OpenSuse) was overcomplicated and buggy rubbish. Perhaps paid versions of these distributions received better testing but I don’t care as long as Ubuntu works well and is free (as in free/cheap and open).
Yes, it does.
The problem with Android is not the kernel but proprietary code (e.g. their application store) that locks the platform from being used in the open.
Which policies? Signing patent cross-licencing agreements so that they are the only self-claimed “legal” distributors of Linux? Pushing Mono to lock us to .Net? Or maybe building the whole distribution (Suse) around a proprietary configuration tool (yast) as they used to do?
Yes, the above is mostly about Novel. I agree that RedHat is a good opensource player (or at least much better than it used to be) but I simply don’t see how does it makes Canonical a bad player. It’s not a zero-sum game.
My biggest concern with RedHat is the price tag of their products – they are simply too expensive. OTOH, using their free distribution (Fedora) feels a bit like running a snapshot of their cvs/git.
Can you show me where I can download the source code that Lispire or Xandros contributed from?
Some of Apple’s or Oracle’s actions are beneficial (like Apple’s WebKit contributions or Oracle’s activity in Linux kernel world). Others are not so – they use mostly BSD licensed software specifically not to give anything in return or buy copyrights and close down the whole projects.
Where exactly do you see similarity between them and Canonical?
Now you’re just being dishonest. You know very well with ‘Linux desktop’ I meant everyone using linux desktop software – Canonical only caters to its own users, like Xandros, Linspire, Microsoft and Apple.
How? It helps Android users? So how that that help ME, another Linux user with an N900? It doesn’t? So that’s my point. The work Nokia does on the N900 DOES benefit Android users. Not the other way around – big FAIL from Google (again, they are working on that now, good for them).
None of that matters to a Slackware user. What DOES matter is the fact that Novell and Red Hat contribute to upstream, so a Red Hat customer is paying for improvements to GNOME which benefit the slackware user (and Ubuntu users like you). That is a big advantage of Free Software over proprietary software – it is a collaborative model where improvements in one place help others all around the world. If you pay for Red Hat, you indirectly help children in Brazil enjoy a good Linux desktop. Canonical however does NOT play this game. You can talk all you like but that is a fact. And yes, they don’t have to, technically, the license allows free riding. Doesn’t make it a good thing that they free ride.
Use google or ask previous Xandros customers. They shipped GPL so they had to make the sources available.
You just said it. Let me quote you with two small modifications:
Some of their work benefits FOSS. If you spend $$$ on Apple or Microsoft products, 0.0001% goes to samba, cups and other FOSS products. The rest goes into products that don’t benefit anyone but their customers. Which is their right.
Most of what Canonical does is not useful for the future, for anyone else. They use GPL and BSD stuff but don’t spend any energy in giving it back, hence it sits somewhere on a website, bitrotting or they maitain it for their own users. If they fix a bug in GNOME, the GNOME devs (paid by Red Hat, Novell and others) still have to fix it. It is wasted work.
It is comparable to what for example Sony used to do. Yes, if you buy a TV from them there is GPL code in there and you can find it somewhere. Still, every bug they fix in the Linux kernel still has to be fixed by Red Hat or Novell because Soney doesn’t tell the developers of the kernel that they fixed that bug. It is wasted resources, just like what Canonical does.
You can argue that it benefits you, as user. Well, so does the work Microsoft does – GPL or not. There is more to Free Software than the license and Canonical ‘just’ complies to the license. So they are bad players in the FOSS world, argue what you like but that’s how it is.
What do Canonical employees do all day? Screw with icon placement?
Canonical has 300 employees and yet contributes very little to both Gnome and the kernel:
http://www.itwire.com/opinion-and-analysis/open-sauce/27089-kernel-…
This is all of course completely fair under the rules of the GPL but you have to wonder why they have so many followers.
As I have pointed out before on the Ubuntu homepage the word Linux isn’t even mentioned. It isn’t even on the desktop features page:
http://www.ubuntu.com/desktop/features
Shuttleworth wants to make a pretty purple OSX junior that isn’t associated with the Linux community. His actions are much louder than the words of a community manager.
Mark has put more *personal* money on Linux than anyone else (does anyone even come close?)
You mean money towards Canonical and Ubuntu.
It has already been pointed out that all volunteer distros like Gentoo contribute more to Gnome than his 300 employee company.
Linux desktop share hasn’t changed since Ubuntu arrived so I really don’t see why people give him so much credit.
People outside the company do most of the work and Ubuntu wraps it up in some touchy-feely marketing campaign. I think it’s a load of BS and I would much rather see ANY other distro take the top spot.
If it was just marketing, it wouldn’t be my Linux distro of choice.
Canonical might very well be this evil company you all say. Still, nothing stops other projects from providing their “good” Ubuntu which is clearly the distro users want.
If Canonical really does so little, it would be easy for any project to provide 3 employees or 3 volunteers with hosting space.
Ubuntu isn’t even very good but it:
– Has saner defaults
– Is easier to manage
– Has better repositories(vast, up-to-date yet stable -for a linux distro-)
than your distro.
I am willing to admit that other distros shine in some areas. But when your solution for any problem is recompile the kernel, wait until the next stable release in 2023 or use our crazy daily updated sub-distro where the system breaks each week, you can’t expect normal people to use your system.
Ubuntu says wait 6 months for something that will mostly work which isn’t good but it is better.
Nobody would deny Ubuntu is a good distribution. What ppl deny is that Ubuntu has brought linux to a higher level. It only has brought Ubuntu to a higher level – but it has to build upon linux, so by not contributing back, they have to keep all their patches themselves, maintain it forever. That is a waste of resources. Moreover, the rest of the foss community will then go their own way, waste THEIR resources on doing again what Ubuntu did. Linux as a whole thus does not progress a bit by what Ubuntu does.
This is a mistake other distro’s have made in the past, and it is not sustainable. It is a sad state of affairs…
Sure he has, and better proof that you can’t just throw dollars at a problem and expect it to fix itself I can’t imagine. Earth to Shuttleman, stop fscking with window icons when your default IM client (Empathy) still can’t connect half the time and will give me nothing more detailed than “network error” as a reason! Fix actual issues first before you start screwing and patching in client side windicator crap and break everything all over again.
your default IM Client? When did Empathy become Mark’s project? Last I checked, Empathy is hardly patched apart from icons which might make up its way from the default Theme.
Who care if Linux is even mentioned on that page? Ubuntu makes a complete OS of which Linux (the kernel) is but one part. You might as well complain that no where on that page does it mention the X Window System or the GNU userland utilities.
It’s another example of how Ubuntu claims to be about the community but wants to take as much credit as possible.
The computer world already has a Steve Jobs and that is enough. Why so many in Linuxland defend a Steve Jobs wannabe is the real question.
I think you wanted to say something else.
As per your words, Ubuntu claims to be about community. So what is wrong for the Ubuntu Community taking credit of exactly what they did(no less, no more)
I think you tried to attack Canonical, but missed it.
Why should it be? That the Ubuntu desktop includes the Linux kernel is an implementation detail, of no interest for marketing purposes.
http://www.android.com doesn’t mention Linux eather and I don’t see enyone bitching about it, grow up kid.
From the website you posted:
Then is a very hiden link, because I couldn’t even find it, for marketing reasons they don’t put it as evident, see the pattern?
Oh is that the pattern? Why not compare the Ubuntu’s website to another commercial distro like Red Hat or Oracle?
http://www.redhat.com/rhel/
http://www.oracle.com/us/technologies/linux/index.html
Both Oracle and Red Hat are clearly proud to sell Linux-based distros. Oracle even shows how they are a proud sponsor of LinuxCod. Ubuntu doesn’t even see being Linux based as a feature.
It’s just a pretty purple OS that is “brilliant” according to Canonical
http://www.ubuntu.com/how-can-it-be-free
Because they sell servers, Linux is well know there, Ubuntu main focus is desktops. Compare apples with apples please. And even if it doesn’t mention it is not a big deal for anyone but Ubuntu haters who are looking for a minimum excuse to atack it.
Wow, you really do have a distorted view of reality.
Here are some facts:
Red Hat’s main focus is *not* servers. Red Hat’s focus is corporate environments which includes corporate desktops and corporate servers alike (which is why the RHEL website already linked has three navigation tabs: Overview, Server, Desktop). See also http://www.redhat.com/rhel/desktop/
If Red Hat wasn’t interested in desktops at all, why is Red Hat the top GNOME contributor? Red Hat even employs four KDE developers (Canonical only two).
Even the Ubuntu Server page does not mention Linux:
“Ubuntu Server mixes effortlessly with Ubuntu, Windows or Mac OS environments.”
http://www.ubuntu.com/server
Damn, no interoperability of Ubuntu Server with “Linux distributions” (apparently Ubuntu is no Linux distribution according to its website ;-)).
Edited 2010-08-10 10:45 UTC
Wow! I never knew that you need to claim that your server OS is based on Linux for everyone to know that it is based on Linux. People who care to know, already know and people who don’t care won’t ever bother to find out the base of any OS.
because users rightfully don’t care how much Canonical is contributing to the kernel and gnome.
Edited 2010-08-10 09:26 UTC
For the record The “About” page on the Ubuntu website mentions Linux repeatedly although not GNU Linux.
http://www.ubuntu.com/project/about-ubuntu
Edited 2010-08-10 18:47 UTC
I can’t believe anyone is taking Ubuntu’s side on this. Back before we had Ubuntu their type of behaviour was considered bad by the community. One of the reasons I picked Gentoo as a distro way back in the day was their statement about how Red Hat was a bad FOSS citizen for not contributing upstream and how Gentoo was going to avoid keeping changes and patches to themselves by not committing them upstream. Ubuntu’s behaviour is not conducive to a healthy FOSS environment. It only benefits them and they could care less about FOSS in general. Sure it’s perfectly “legal” to do things this way but it doesn’t make you any less of a jerk. I think the supporters out there can’t see past the only disto they ever used. Meanwhile it looks like Red Hat went the other direction and has become a much better FOSS citizen.
There’s nothing wrong with adding your own technology but it should be put through the proper channels so all can benefit. Without FOSS Ubuntu would have nothing to work with. They would do well to remember that.
Edited 2010-08-10 05:43 UTC
Whatever makes you feel superior, dude.
Don’t let your insecurity get in the way of writing a meaningful post, dude. Seriously though do you have an actual defense of the crap Ubuntu is pulling? It’s not like their behaviour is only looked down upon because it is Ubuntu. It has ALWAYS been looked down upon in the FOSS world. That was the point of my post. Just because Ubuntu is now the most popular distribution doesn’t mean they get a pass. It wasn’t cool when Red Hat was doing it either.
I really don’t mean to bash Canonical, overall. I tend to regret they don’t seem to contribute upstream properly, etc. But I guess they also lack resources (I mean, sustainable resources, from services like Ubuntu One, not just philantropy).
Speaking of which… I wish I could support Canonical through Ubuntu One… But why didn’t anyone write about the absolutely terrible Ubuntu One service ?
I mean… Does anyone actually use it and have a positive opinion about it ? I tried, I tried a lot but, most of the time : it doesn’t sync, it eats a lot of cpu, it’s full of bugs.
I really wish I could support, through subscription, a company that contributes to the Linux ecosystem as a whole. But, I’m somewhat dubiousabout Canonical although I’m sure they hire some really competent guys. What do you think ?
Support Red Hat by buying RHEL Desktop subscription. It’s $80 per year and you get RHN access.
Other thing, there is JBoss Developers Studio which get you entire Red Hat portfolio for $99 a year. I have that on my workstation at work (I do Java EE development on JBoss EAP).
An alternative to support FOSS is buying Mandriva Power Pack.
It’s funny to see the comments. The Canonical supporters are clearly of the school, anything that makes my distro better is good for me and the distro. The Canonical detractors are following the more pragmatic line of those who work with one another are the ones who produce long term sustainable code.
Could it be this is the split between the older Linux generation and the recent influx of new users? The old guard went with Linux because of the whole package, software and ideology. I wonder if ideology is a big motivator amongst the newcomers.
Anyone with a longterm view towards the FOSS ecosystem can see what Canonical does right now isn’t sustainable. Upstream writes the vast majority of their code and the code they tack on in house makes Ubuntu increasingly incompatible with upstream. Upstream will always be bigger than Canonical/Ubuntu ever can be. If upstream decides to ignore Canonical, they are up the creek.
While it might look more productive to do some stuff in house right now, Canonical is starting to diverge from the base code they need to make their OS. I wonder if this will end like Corel Linux… http://en.wikipedia.org/wiki/Corel_Linux
I might now be considered an old timer but I would say cooperating w/ upstream development is much more of an pragmatic issue rather than an ideological one, and to me that pragmatism is the beauty of open source development.
For final products, I agree that some level of non-democratic control and focus are beneficial and am glad that ubuntu at least is focusing on end user desktop experience.
It sucks that they can’t do it while properly coordinating with upstream and it’s likely to burn both ubuntu and upstream projects to some level. But, hey, worse things have happened and passed.
Being bought by Xandros?
Well, maybe not by Xandros. The same pattern could arise, though.
Hack up your distro with incompatible code. Fail due to being incompatible. Get bought by a party who doesn’t understand that veering too much from the mainstream only gets you further up the creek. New owners hack it up some more. Fade further into obscurity.
sometimes doing the right thing may not be the ‘right’ thing, like packing FOSS code to make it more user friendly to new adopters, digging your own pocket and giving it away for free.
see, i told you, they are killing each other
Killing each other? No, Canonical isn’t following the unwritten rules of the Linux developer community. They are called upon it. Canonical can do two things now. They can adapt and be accepted in that community or they can keep going it their own way. It’s just that the latter will eventually mean that the larger developer community will ignore Canonical.
Once the ignoring happens, Canonical will find out that upstream can make and break their packaging experience. If upstream goes on without consideration for Canonical and changes upstream break Canonical’s developments, Canonical will either have to keep up with fixing the broken stuff or they will have to build on older code that is increasingly different from the new developments in upstream.
Mr. Shuttleworth, wherever you are, if you happen to read these, maybe you should consider stop funding ubuntu to let it dies peacefully, so that linux desktop will once again become piles of code the majority of computer users never know like it was before, 1% for over a decade. instead, you can use your money partying with women and no FOSS community will blame you for that.
Edited 2010-08-13 22:39 UTC