Here‘s an interview with Debian’s Branden Robinson. From the interview: “I think Debian’s success stems from its dual commitments to userempowerment through free licensing of the works we distribute, and to high quality through careful design and integration decisions. Not everyone is a software freedom fighter, nor is everyone a software perfectionist – but we can accommodate both of those passions pretty well. Debian is an exciting and rewarding place to be if you want to help shape the future of the Free Software landscape.”
“Interview” is a bit of a cheat, since this is clearly an exchange of written questions and answers. Strikes one as a bit of a PR job really as no awkward questions were asked, such as “Do you think Debian risks ‘irrelevance’ if the next Stable release isn’t made available by the end of 2006?”, or “What’s the state of play with the DCCA?”, or “What’s the feeling among Debian developers about Ubuntu?”
It’s hard to get any feel from the interview whether this captain is steering a fine course in clear water or is heading for a rocky shoreline.
debian is irrelevent.
I suppose Ubuntu, Knoppix, Xandros and all the other Debian-based distributions are irrelevant as well? Without the *enourmous* amount of work that goes into the development of Debian these other distributions would not exist. Saying Debian is irrelevant is probably about the most naive Linux-related comment I’ve ever heard. Then again, perhaps by your reasoning, Linux is irrelevant since Windows maintains 90+% of the market share on the desktop. Please consider what you are saying before posting.
To re-affirm your point; didn’t Mark Shuttleworth state that there wouldn’t be an Ubuntu without Debian?
Most Debian based distro’s take a snapshot of unstable or testing and create their products. So if Debian is irrelevant, then all Debian based distro’s are irrelevant.
FYI: If that statement was true; over 100 distro’s are irrelevant; out of approx 400 on distrowatch. I guess that someone should send out an email to all the users.
PS: I am both a *bsd and Debian user.
Cheers
“debian is irrelevent.”
Ehhh, maybe it’s just me, but Debian is so easy to setup, especially with the new installer and the net install disk. Downloading and installing a minimal environment, plus any additional apps you might need, is just too easy. There are only two distros I would use on the desktop right now – Gentoo & Debian. For servers, Debian is definitely #1. It”s definitely not irrelevant for some of us.
Not as long as it’s the number one distribution on DistroWatch.
http://distrowatch.com/dwres.php?resource=independence
Yes, I believe that the debian derived distros are irrelevent also.
No, I belive GNU/Linux is relevent, I don’t use any other platform.
It is the most naive commnet you have heard? Well, considering I am forced to develop on it daily, support it daily (it being debian, its a legacy desicion I have to live with) as well as RHEL, SuSE *BSD’s, I can tell you from my commerical experience debian is irrelevent.
Perhaps it is you that is naive to think that ubuntu/xandros/linspire make debian relevent or to even think debian is relevent in the first place.
As the person who considered your statement naive (and still do), I’m curious to know what in the GNU/Linux world you do consider relevant if not Debian? You have said nothing to backup your statement other than try to flaunt your expertise with a very vague statement.
BTW, I do not think Ubuntu/Xandros/Linspire legitimatize Debian, but rather their combined popularity certainly make the platform quite *relavent*. Debian has and continues to be one of the most influential and well developed distributions in the GNU/Linux world. To call that irrelavent hardly seems fair let alone substantiated. Perhaps you can offer some more depth to your statement that can prove otherwise??
As someone who has had to support and develop on various distributions in the past and present I can say I found Debian to be the easiest. Mandrake was the most difficult to work with followed by SuSE, Fedora Core, and Redhat.
FreeBSD falls in there before SuSE and after Mandrake, but only because it’s philosophies are different than most linux distros and that took some getting used to.
Ah! I UNDERSTAND! I get it now…
“I believe GNU/Linux is relevant…I don’t use any other platform”
So Debian, you say, is irrelevant to YOU because you don’t use it? Is that the logic? Oh, and you do use at least one (perhaps several?) other platform – *BSD
You devlop on Debian daily? Really? And RHEL, SUSE, and multiple BSD’s? Wow. You’re likely one of the most multitalented trolls I’ve ever encountered!
Your commercial experience tells you debian’s irrelevant? Really? Because last I heard, Debian wasn’t overly concerned with being commercial. I don’t see many options to buy support etc. Do you? Never seemed to be an issue.
Did it occur to you, perhaps, that Debian’s goal is to create and promote Free Software? A philosophy, not simply a business?
Further, have you ever thought that maybe (and this may be a stretch) that the very nature of open source software leaves it mostly immune to market forces. Open Source can be very democratic – the developers and users are in charge. Period. The corporations may sweeten the deal, yes – but they can hurt too.
If Red Hat, SUSE, Canonical, Linspire, etc. were all to die, Linux would still remain relevant and exist. Why? Because of groups/distros like Debian. Part of freedom is freedom from being a slave to the Almighty Dollar/Euro/Peso, from capitalist forces, from competition. It’s about expression, about work. Debian can be seen as inspiration to many other distros. Imagine Ubuntu without Synaptic/Apt-Get. Yeah. Not fun for me to think about. How about no dpkg?
Free Software is an art. Can art be a product? Yes. Can it be commercialized? Yup. Sold? Mmmhmm, to a degree. But in the end, if the galleries shut down and the artist is no longer popular with the rich folk, will he suddenly become irrelevant? Not as long as at least 1 human being enjoys/gets something from his art. Even if that one person is him.
I’m done feeding the trolls. Maybe you should look up the definition of relevance and realize that you should explain what you mean. Relevant to what? To whom?
#1. Try posting your name and not anon-e-mouse! Take some responsibility in your commenting..
I happen to disagree with your comment about debian being irrelevent.. Still nothing as good as APT yet.. Make system admin a breeze.. I know that anything that’s in a .deb pkg is going to have a 90% chance of running perfectly.. Usually what I do now is use klik to test software and if it works I APT get it.. Somewhat irrelevent if you ask me eh? But debian is irrelevent..
C’mon guys, learn to stop biting the trolls
This one can’t even spell irrelevant.
Good, solid distro with excellent layout and delivery.
Let the numbers tell, it’s very well supported in the community.
I whish they do a bit better hardware support. But never the less if “Stable” is set on a supported H/W they make a kick arse server to run.
As a desktop, if one likes living on the edge and has the latest H/W, it could get a bit challenging at times.
Keep up the Good Job And to all Debian developers.
1) Best package management, easiest upgrade. Intstall it just once.
2) 100mb download means I don’t have to download, or install, or uninstall, a bunch of cr@p I don’t want.
3) Use super-stable proven technology, or bleeding edge with all the bells and whistles, or somewhere in between, debian doesn’t care.
4) No strong arming to use this or that windows manager, or desktop environment. Use whatever you want. Don’t even load X-Window and run from the CLI if you want.
As I said, it works for me. I tried dozens of distro, but always come back to debian on the desktop.
JMHO.
Praise to Debian!To all the Developers involved Past Present and Future. All Ubuntu/Xandros/Linspired/etc wouldn’t exist without the MotherShip. Open Source software is barely understood but is big reason behind even Such as OSX. Don’t let the Marketer’s fool you. Be part of the Source.
I used Debian/sid for over an year.
IMHO using Debian for a desktop usage is a waste of bandwidth.
First of all Gnome usually take many months before going to sid, unstable !
Also if you launch the apt-get upgrade or whatever there are tons of patch to download each day.
You know I remember to see gsteamer, don’t remember the exact name, patch 1, 2, 3 for many consecutive days.
If you do an upgrade each week, well many many MBs to download.
I know it’s unstable but with debian there’s no way to use a modern, relatively see my gnome’s previous comment, desktop if you don’t use sid.
By using OpenSuSE I don’t have those many patches to download each day and it’s stable…and works out of the box.
For the server side, cannot post anything.
Wow.. Then you must really go through some pkgs! I run an ubuntu/debian server setup and I’ve had it run for about 5 months so far and have only had to d/l a total of 60 megs of pkgs at the MOST… Much less then RHEL in which it seemed I had to do that a week… SuSE and mandriva significantly less but equal to ubuntu.. But as long as people write software.. There will be mistakes and things will have to be patched… If you think this is bad.. Try getting info about security breaches from apple… No patches for a month then BaNG! 20 megs of updates in one day! 😉
Hmmn, I guess the rather uninspiring nature of the interview hasn’t produced any interesting thoughts here, just a lot of trolling. Acutally, if I recall, “irrelant” or “irrelevance” was used by Debian’s founder Ian Murdoch who said that if the next Debian Stable didn’t come out in a darn sight less than the three years it took to produce Sarge (the current Debian Stable), then that is what Debian would risk. It’s a fair comment, though not in the sense it which some have used it on this thread.
Maybe one day Debian might look at their development tree, so that what is now Testing and Unstable were rolled into one, branching into two things, a version that’s primarily for desktops and a version that’s cast-iron for servers.
Hmmn, I guess the rather uninspiring nature of the interview hasn’t produced any interesting thoughts here, just a lot of trolling.
Branden Robinson’s answers are smart and well worth reading but the questions are clueless. The interviewer doesn’t even know how to spell “Debian”, he (or is it she?) thinks Debian has only one maintainer, and after hearing that Robinson’s role as a Debian developer has been mainly in maintaining XFree86, he starts asking questions about security in Debian. Right. But that’s hardly a good reason for you to start posting your trolling comments.
Acutally, if I recall, “irrelant” or “irrelevance” was used by Debian’s founder Ian Murdoch who said that if the next Debian Stable didn’t come out in a darn sight less than the three years it took to produce Sarge (the current Debian Stable), then that is what Debian would risk. It’s a fair comment, though not in the sense it which some have used it on this thread.
Mr. Murdock’s opinions about Debian’s relevance are more or less irrelevant because he’s now the leader of a (Debian based) commercial distro called Progeny and his interests concerning Debian are commercial interests. Debian is “relevant” to Mr. Murdock just as long as he can use it to increase Progeny’s revenue, while Debian’s main interest (besides making the best operating system that the possibly can) lies in advancing the freedom of software in accordance with the GNU philosophy.
You must be either ignorant or a troll (possibly both) because you apparently haven’t read any of the news about how Debian developers (the real ones, not Mr. Murdock) have several times publicly addressed the criticisms of Debian’s slow release cycle by announcing that they are now working on switching to more frequent and timely releases. The next stable Debian release is currently planned for December 2006.
http://lists.debian.org/debian-devel-announce/2005/10/msg00004.html
Maybe one day Debian might look at their development tree, so that what is now Testing and Unstable were rolled into one, branching into two things, a version that’s primarily for desktops and a version that’s cast-iron for servers.
Debian cannot be everything to everyone, like Branden Robinson suggested in the interview. (Maybe you should read it, too.) There are already several successful specialized distros that base their popularity on taking the results of Debian’s work as their starting point and then adding some special treats on top of that. There’s no way that Debian could possibly compete with such strategy and I don’t see why they should. The current arrangement serves best both free software and its users. Debian should concentrate on doing what they’re good at — building a very flexible and reliable high-quality GNU/Linux distribution which the derivative distros can use it as their base and add some special features that will attract desktop users. The developers of distros like Ubuntu, Mepis, Linspire, Knoppix, or Damn Small Linux are well aware that their distros are actually Debian systems, although some of their more clueless users aren’t always necessarily aware of this fact.
Edited 2005-12-05 11:23
You must be either ignorant or a troll (possibly both) because you apparently haven’t read any of the news about how Debian developers (the real ones, not Mr. Murdock) have several times publicly addressed the criticisms of Debian’s slow release cycle by announcing that they are now working on switching to more frequent and timely releases. The next stable Debian release is currently planned for December 2006.
Funnily enough I am typing this on Debian Sid since Debian is the distro I use. And, yes, I think I might just have heard of Debian’s newly envisioned release cycle, as well as of Ian Murdock, Progeny et al. Somehow, I doubt Ian Murdock, for one, is likely to lose much sleep over your “advice”. Have you founded any successful Linux distros lately?
Just my 2 cents, but I found Branden Robinson’s answers disappointing, perhaps because the questions were dull and predictable. A good interview needs to tell the reader more than they already know, which this one doesn’t.
Me, I’m more interested in keeping an open mind and asking questions. Such as whether the new release plans are working out or not. This isn’t a religion, it’s just a computer operating system.
And I’m much more interested in the reasoned and thoughtful reply elsewhere about the undesirability or otherwise of updating computers en masse on, say, a six-montly cycle. Agreed, a very powerful point, but I still suspect that Debian’s system of three branches, which has served it so well, is going to become increasingly hard to sustain before the decade is out. That is because increasingly ruthless competition in the Linux world will deliberately try to colonize Debian’s turf.
> Maybe one day Debian might look at their development
> tree, so that what is now Testing and Unstable were
> rolled into one, branching into two things, a version
> that’s primarily for desktops and a version that’s
> cast-iron for servers.
Contrary to popular belief, that wouldn’t work.
If you have to manage 1000 job-critical desktops, the last thing that you want to do is QA your SOP every 6 months and upgrade every 6 months. You’d like to do it every 2-5 years. That’s why RHEL is so popular and why Debian stable is also perfect for this type of desktop.
If you have an important web server that uses the latest technology, the last thing you want to do is hunting down the appropriate backports directory or compiling from source if no such DEB exists or relying or SID repository (which is the place where backwards incompatible changes are tested and firmed up). For these servers, a derivative distribution like Ubuntu server edition is perfect.
Debian isn’t supposed to be all things to all people. That’s its strength because it is a great foundation for other more tailored distros to be built and providing an infrastructure for all these tailored distros to collaborate. It’s been so good at this that 5 of the top ten distros on Distrowatch are Debian-based and Debian has more derivatives than any other distro.
Debian is a lesson to the free software community. A lesson about how not to organize developers.
Debian has now been reduced to a big “repository”, a source of software for other distros. Maybe it’s better that Debian face the fact that they have been, are, and will be utterly unable to keep up with their lofty goals of delivering a stable ditribution. If they keep working on sid, and people like Ubuntu re-package and keep security updates, we’re all better off.
The problem stems from Debian much-praised packaging format. It relies very much on human intervention. And Debian volunteers are unable to keep the flow. They’re stretched out, and this has grown proportionally with the number of packages they supported (2000 -> 5000 -> 12000). The more they supported, the more they slowed down.
This has nothing to do with them being volunteers. You just look at the BSDs. Not only to they package software, but they develop kernel and userland software. Debian does not do that, and yet it collapsed under its own weight. This happened because the whole development cycle in Debian is flawed from a software engineering standpoint. First, they make no separation of “base packages.” Everything comes in a huge lump. People who worked to solve this were largely ignored (e.g., Progeny) Second, the whole “democratic” process. Not everyone should have the same rights. It’s not how a meritocracy should work (look at Linux development and BSD distro development). Also, Debian has repeatedly been caught in religious wars. They practically reject anything that is not GPL. OSI approved licenses are ignored.
When look to your neighbours (BSDs), you quickly recognize the trick is how BSDs distribute third-party software and the fact they have some centralization. But that is that, and it cannot be changed. Debian’s dpkg is here to stay (experimentes by Conectiva have modified dpkg by increasing granularity, it resulted in slowing down the updating and making the process harder.)
I am not dissing Debian developers. I think they do their job…somewhat…I am facing the facts. I’ve used Debian form many years, but now I am an Ubuntu user and will never look back.
It seems that the best working arrangement is to have Debian provide a pool of software and then have Ubuntu come and perfect, secure, update and repackage that, while keeping up a 6 months delivery cycle. Maybe that’s how it should be. Empirical evidence supoprts this. Everything else is wishful thinking.
> I am not dissing Debian developers. I think they do
> their job…somewhat…I am facing the facts. I’ve used
> Debian form many years, but now I am an Ubuntu user and
> will never look back.
You’re *really* missing the point. Listen to Mark Shuttleworth some time talk about Debian. He recognizes that Ubuntu won’t exist if it weren’t for Debian and the Debian does most of the heavy lifting that makes each Ubuntu release possible.
You mention the BSDs as a shining example of how to do things. You’ll have to recognize that the free BSDs had a huge head start over Linux and was better in almost every way until a few years ago, yet didn’t advance anywhere near as far. The key thing is, one size *does not* fit all and Debian recognizes this. But they also recognize that they can be the framework for several tailored distributions like Dam Small Linux, Knoppix, Ubuntu, MEPIX, Progeny, and friends.
If the free BSDs followed the Debian metadistribution approach instead of trying to being all things to all people, it wouldn’t have forked into OpenBSD, FreeBSD, NetBSD and BSD might have conquered the world. In that BSD world, OpenBSD would be analogous to “Debian Stable”. FreeBSD, NetBSD, and other possible distributions like “Dam Small BSD” or “BSD Knoppix” or … would be based off of “BSD Testing” (which would eventually make it’s way into OpenBSD). The experimental repositories of FreeBSD and NetBSD would be “BSD Unstable”. Ubuntu-like distributions could be based off of “BSD Unstable”.
This has to be the most lucid comment about debian and I do agree with it… A pkg trim maybe needed to lighten the load on the APT infrastructre..
You can take this interview and time warp to 4 years ago and it will fit. It’s like nothing’s happened with Debian, like a huge Tsunami never happened…Like Debian’s the perfect GNU/Linux distro, updated and much in use today, when the truth is Debian has lost its place as a frontline distro.