Slashdot Mirror
Debian, Past Present & Future
solferino writes "Christoph Lameter, a major guru in the debian project, has put up a very well written talk that he gave earlier this week that addresses debian's past, present and future.
He includes a good background history of the project, some interesting sets of figures and projections (30,000 packages by the end of 2004!), a good discussion of the pros/cons of source based distros and his ideas about a new package manager he is developing (uPM). In all a very good read, whether you are just now considering dipping a toe into the debian well-spring or have been drinking from the source for a long time already."
I imagine this will be similar to the catastrophic Y2K bug. :)
sig.
Oh come on, by then we'll have Toy Story 11 with a whole bunch of new funky character-names ...
Life sucks.
640 are plenty for all, I thought...
Unix makes easy tasks hard and hard tasks possible. Windows makes easy tasks easy and hard tasks $29.95.
I've been using Debian for a few years now, after using Red Hat for about 1.5 years. I've really gotten to like it; server updates are easy, and running unstable on my desktop allows me to install many recent things with very little trouble (and no, unstable almost never breaks).
However, I'm not sure that listing absolutely everything should ever be a goal. Having a lot of packages is very good, because it's nice to easily have all of your choices laid out, but it can make it difficult when you're trying to choose software. I can only imagine the horror if they tried to list every CMS or MP3 jukebox (we get _buckets_ of those types of projects submitted to freshmeat, and most are very similar).
On the other hand, people shouldn't necessarily be restricted from putting new packages in Debian just because there are a lot of similar projects, because everyone has different needs. It's a difficult problem, and I'm not sure how/if the project currently deals with it (though most everything I've seen in there seems to be of reasonable quality).
WMBC freeform/independent online radio.
Debian developers are known to have strong convictions and it is easy to get into some old argument when the buttons of one group or another are pressed.
/. posters were debian developers as well.
heh, I didn't know all
I've run Debian for years, and I've always felt that other distros were better. After all, those other distros get much more press, they've got glitzy widgets and eye candy, and it's hard to resist that.
So about every 6 months I'll hear about a new version or distro and will give them a shot. I'll install them and make an honest effort to use it, rationalizing my choice just like the distro's marketing people want me to. But I always wind up throwing my hands up in disgust and thinking, "How can people use this crap from day to day?!"
Now, with many distros polluting the ideas of free software and open source -- feeding you a GPL license and then their own proprietary license which prevents you from copying CDs and giving them to all your friends or from installing on multiple computers -- there's more reason than ever to use Debian.
Debian's geek appeal is legendary. But now, with Debian's Desktop and Education sub-groups, the old idea of being proud of a geeky install is disappearing. Debian's beta installer is on par with every other distro's -- a fact that thousands of Debian users are eagerly awaiting.
Everyone's heard of apt-get and Debian's package management system. Yes, it's as slick as you've heard. But fewer people realize the huge scope of software available in Debian. I run all my desktop machines with Debian's "unstable" (think "unstable" as in changing; Debian's "unstable" release might have bugs, but there are certainly no more bugs in unstable than in the commercial release distros!). With that I have a huge selection of software -- over 10,000 packages. All of those packages are done by registered developers whose first job is to do it right.
When I read in Linux Journal or somewhere online about a nifty program XYZ123, I just try to install it -- 9 times out of 10 one of Debian's hundreds of developers has already packaged up XYZ123 for Debian. There's a huge advantage of having a distro that is controlled by geeks who like computers and who do this for the fun of it. Debian's developers are into GNU/Linux, and it shows.
On DebianHELP we call Debian "militantly free software". Yes, that's what it is. That militant attitude permeates Debian and this is Debian's strong point. I like the fact that Debian people worry about little details in the license agreements. I like the fact that Debian segregates non-free software into its own little slum. I like that Debian has a "social contract" and clear guidelines about what it's interested in and what it's not.
Many times I've often said to myself, "Gee, why are those guys worried about that stupid thing..." (e.g. the old KDE-QT license battles). But time and time again I'm proven wrong as the correct view turns out to be the morally miltant view. Besides turning out a first-rate distro with loads of software, Debian's role as GNU/Linux's moral compass is something we can't afford to lose.
Keep in mind that you can install from source when you really need it. For example, I was still running potato on my servers a while back when AOL broke Everybuddy. Newer versions were out, but Debian only updates packages for security reasons (which is occasionally annoying). Not a problem:
apt-get -b source everybuddy
This will grab the source and Debian modifications, apply the mods, and build a package. You can omit the -b option if you want to customise it.
Some will say that if you do this, you lose the stability provided by Debian's long release cycle, but I disagree. The rest of the system is not less stable because you installed an IM client (which shouldn't be able to hurt anything else, unless there are severe bugs in it). This is not any different than compiling it from source yourself and installing it into
On the other hand, installing packages from testing or unstable may upgrade libraries, and that could affect your system as a whole (especially if it's libc6). You'll have to weigh the benefits of this if it ever comes up.
WMBC freeform/independent online radio.
Well, I heard this quite a few times now, but my own experience is quite a bit different. Debian stable is at first old, not stable. I have run in quite a few showstopper bugs that were already fixed in upstream and in unstable but which never made it into stable, since the QA process which makes it quite hard for new upstream to ever make it into stable.
I think the main problem here is that the freeze is globally to all packages instead of local to small package groups, in a lot of cases a package is still heavily under development when the freeze happens and then for month or years it will not get updated, even if the upstream becomes a lot more usable and stable.
Quit writing about the history of Debian and get to work on those packages!!!
I think this article points out very well the problems with Debian that cause it's extremely slow release cycle.
First is it has way too many packages it tries to support. The articles says 9000 packages for 3.0 Woody. RedHat has something like 1800(I am sure this figure it slightly off) packages in RedHat 8.0 Psyche. So Debian has 5 times has many packages to support. I do think RedHat would do better to support more packages, but not that many more. I might add say abother 200 packages to RedHat which would put them at 2000, still 4.5 times less than Debian.
Second is that Debain tries to support way too many platforms. My personal philisophy is that while it would be nice to have the exact same distribution for every platform it is just impractical. It would help their development greatly if the cut platforms. If they wanted to be logical about it they could probably look at the number of users of each platform and create some minimum number of users to support that platform.
I do think that simplifying package maintaining is a good move, but they are trying to fix the symptom instead of the problem. I applaud them for delaying to make releases extremely stable, but I think some of their maintainers have the wrong idea. One example that comes to mind is the XFree86 maintainer continuing to maintain XFree86 4.1 months after 4.2 had been released.
Another point about Debian has been it's horrible installer. I am hoping the Progency installer does take over and is also improved upon.
I use RedHat and have for years, but I am looking for a new distribution. I am doing so because of bad decesions RedHat has made in my opinion. These include rushing releases to meet a deadline instead of holding back and making it truly stable. They love to talk about their Q/A team, but their touch is very obviously lacking in RedHat 8.0, but at least they aren't as bad as Mandrake Q/A.
Then there is having Havoc Penington as an employee and on top of that having him as a Gnome Developer. He has a philisophy of simplify the user inferace to make it more usable. I agree with this idea, but he takes it way to far. His definition of simplification is dumbing everything down and removing very useful features and settings. Even worse he has convinced many Gnome developers.
I have tried Debian, Sourcer, Mandrake, Gentoo, and even Slackware before RedHat. None were better than RedHat in my opinion. Like when I used Windows, I wish for something better.
Havoc Penington, the bane of my Linux desktop.
Nope. If you'd read the article you'd've seen that uptill woody they released every year.
The reasons for Woody taking extra long were:
- BIG jump in archs...up to 11, with stuf like X-Windows not actually designed for those archs, but they fixed it!
- Security Patch maintainers where up to there neck in work and it was a bottle neck (remember the planned may release? This was why it was prosponed)
- They only started fixing the above 2 probs after the freeze so that no new stuf could get in.
But thankfully these issues are resolved now, so Sarge should release (ahem) on time. And anyway the PM (Project Manager) of Sarge wants to have cds of testing too.
Moderation: +4. Modded 70% Funny and 30% Overrated. 100% Saturated.
Yes, this is sometimes annoying. One example is Mozilla; for a very long time, we were stuck with M18 in Potato, while new releases were certainly an improvement.
I understand that there are more variables in Debian with all of its supported architectures, and it wouldn't be as easy to simply release updates of later versions as distributions such as Red Hat do; you can't be sure of the impact it will have everywhere, and backporting security fixes is safer.
Perhaps a "mostly harmless" package repository could be created. No, "testing" doesn't count, because the packages in there will often be built against new libraries, and you probably don't want to go there. But this could contain binaries for packages such as Mozilla, which gets updated a lot (1.1 really is much better than 1.0) and would be unwieldy to build from source). These binaries would be built on a potato system. Those who wanted this sort of thing could simply add another line to their apt sources file, and accept the small risk.
It's possible for someone to do this on their own; Adrian Bunk maintained a repository of several updated packages so that 2.4 series kernels could be used on Potato. But I think it would be nice to have this as an official part of Debian. It doesn't sound so great to say, "Oh, yeah, you can do that; just get the packages from $THIS_GUY".
WMBC freeform/independent online radio.
If you got that reference, I'm really sorry.
Anyway, please read this:
http://lists.debian.org/debian-devel/2002/debia
Maybe this will clear up a few things. Debian is supporting these architectures because no other Linux distribution does. As the message states, XFree doesn't even support as many architectures as Debian; the Debian project is how users of those architectures get XFree86 at all.
Maybe you feel that they are not important, but I think that the people using them would disagree. Obviously, there are enough people who use each platform to do the work of porting packages to it. What makes you think that they would turn around and do some other, "more important" work instead if support for their architecture was dropped?
[And isn't this why most hardware manufacturers don't release Linux drivers? Because "most people" use Windows?]
Debian exists as it is for many reasons, and there is nothing else like it. It is not going to change into your idea of the perfect distribution. However, there are several distributions which are addressing some of the "problems" _you_ (and others) have with Debian. Most of these amount to pretty graphical installers and a few other things, and are only for x86. Since that seems to be what you want, why don't you try one of them? IOW, don't complain that Mozilla doesn't have an integrated AIM client; use Netscape instead.
WMBC freeform/independent online radio.
Umm.. I'm not sure about this exponential extrapolation thingy. By the same logic, they would be supporting something like 120 architectures by 2006 :-)
The future sure does look bright for debian. The way I see it, debian went through a huge backend change during the potato-woody release. The number of architures soared and the number of package become increadible. Essentially the developers where forced to automate the entire build process across all platforms. This was hugh, and puts debian in a really good postision going forward. From what I can tell, debian is the only distribution where scripts control so much of the backend in building, distributing, and bug tracking the system.
:P )
Now that this change has been accomplished there is alot of "low hanging fuit" for debian to pick.
The installer has been reworked and is currently in early beta testing. the whole thing is now modular to allow for easy porting to different architures, and to to allow for a very flexable install. Most debian developers want to be able to pop in a cd, have it detect most of the hardware, then automatically install baisc packages. The new installer allows for this, while retaining the power to customize or trouble shoot installations.
Also there is the new PGI installer. Version 1.0 is out now. This is as simple as redhat/mandrake installers. It has great hardware detection and set up and gnome desktop.
If you want to try thy hardware detection routines in debian (package discover) then try popping in knoppix 3.0. This is a debian system on a live cd. Pop it in to most PC and the network, x, sound, and usb are all configured.
The desktop is coming to debian. Both gnome2 and kde3 are about to move from experimental to unstable. They should be in sarge for the next stable release. gnome2 will enter this weekend, and kde3 is awaiting the completion of the transition to gcc-3.2, but its ready to roll.
There are now many subprojects that will help puch debian onto the desktop: eg the desktop , the education and the music (demundi) sub projects are all starting to take off.
Debian been my primary desktop for 2.5yrs, since I gave up on redhat. I keep trying other distro, but they are just to inflexable for my needs, and to difficult to maintain. I want something to get my work done, not to continuely tweak the operating system.
AC-DC
PhD elec. eng.
PS for those who think that lindows software warehouse or ximian redcarpet are cool, try kpackage with it confgure to use apt. This is the default in the debian's kde. This combo of kpackage and apt-get is the most powerful and user friendly package installation i''ve ever seen. ( Of couse I personally still use apt from a terminal.
I
If you look at that nifty chart provided at the beginning of the paper, you can easily see that the release time for stable is stretching out further and further as more packages and architectures are added to the fold. At 9,000 packages and 11 architectures, it took 2 YEARS to ship a stable release. If they are right, and 2004 will bring 30,000 packages (and probably another arch or two), how long is the release cycle going to stretch out to? And will people tolerate that long between stables? Look at the grumbling and outcry about the delay in getting Woody out the door.
I find it very interesting that the article would point out several times how difficult it is to maintain all of those packages and the diffs as they are updated, then point out how using a source-based distribution makes that kind of thing much easier. And yet, the author seems to suggest that source-based distros are somehow not as feasible as binary-based distros. He even goes on to call source-based distros "immature". Perhaps in the Linux world, but how long has FreeBSD been around? It's okay to borrow ideas from other groups when those ideas seem to be working. I think that the Gentoo project has done a great job in taking the idea of a "ports" system, addressing the shortcomings, and putting a workable source distribution system on the Linux platform.
In my mind, if Debian is going to continue scaling to 5-digit package listings, the project might want to look into the possible benefits of switching to a source-based distribution system. Look at what Gentoo has done, address any shortcomings, and develop a better source distribution system. Doing it the current way with 30,000 packages to maintain, we might not see Debian 4.0 until 2010. And there are probably a lot of people who can't or won't wait that long.
I can see how it, from a x86 user's perspective, would seem reasonable to drop support for most architectures. 95% of all computers are x86:s, right? But when you're one of those people who use more architectures, it's wonderful to be able to use the same distro on several systems.
I run debian on my TiBook and on my P-III at work (as does everyone there, except for management). It really is a blessing to have the same environment everywhere and that everything works (almost) the same way, whatever computer you're using. Not that I use the more exotic archs, but I definately can understand why it is desirable to support other platforms as long as people are using them.
It definately would be nice if the university (that I go to) would replace their aging Solaris 2.6 installation on their sparcs with woody. But that probably won't happen...
-- Free speech is only free if your time is worth nothing.
These are really related:
/. many times before, including the practicalities of building from C source, the relevance of Mono, standardization of the C Sharp language and the ownership of Java technologies.
The problem is the combinatorial explosion of testing effort dictated by the support of a large number of packages on a large number of platforms.
To this there's no easy answer. However, I do not believe that the 'competition' will really come from source-based distributions such as Gentoo mentioned above - ultimately the amount of testing to be done is the same, by getting you to compile things yourself, Gentoo makes it no more likely that a combination will actually work.
Ultimately the threat to Linux as a platform is the Dotnet virtual machine - a software platform comprehensive and abstract enough to reduce the n*m testing needed for Linux and Windows today back towards the 'n' of a single platform. Once Dotnet gets established, the relative cost of writing cross-(hardware)-platform applications will plummet and Linux will be unable to catch up.
The only genuinely equivalent technology available to Linux is Java. Therefore the only viable strategy for a group such as Debian, meaning a group that is serious about having broad hardware support and comprehensive package support and some assurance of quality and comptibility, is to embrace Java, encouraging the development of Java applications and supporting the Java VM as comprehensively as possible.
These issues have of course been discussed on
To date, the only real counter-arguments that have stood up are those of simple denial, that is, putting off the day where cross-(hardware)-platform compatibility has to be addressed so far into the future that it is likely that Linux will already have become an irrelevance by the time convergence takes place, or the reckless and naive assumption that the open-source community can clone, and will be allowed to clone, the the Dotnet platform in its entirety.
It will be interesting to watch how key development streams such as Debian, KDE, StarOffice etc. attempt to reconcile these conflicting demands. From the perspective of Java developers like me, it's becoming hard to resist the rather depressing conclusion that at least some of these difficulties are self-imposed.
Anyway, please read this:d ebian-d evel-200204/msg01343.html
http://lists.debian.org/debian-devel/2002/
A quote from this message sums up exactly in one sentence why debian is worthy of the support and admiration of the community:
"I refuse to treat non-i386 users like second-class citizens."
As a linux-using powerbook owner, I thank you. I can't count how many times I have been told that I am not worthy of consideration because my niche is too small. Thank you debian for extending the useful lifespan of my computer several years while everyone else snorts derisively and tells me to buy a new Mac.
Its a really handy tool ported from debian. See http://apt.freshrpms.net
Now if only Red Hat would adopt it instead of up2date...
-- Thorin sits down and starts singing about gold.
I like how he makes utterly false claims about his data based on his misleading graphs.
Look at his debian growth graph. He conveniently skips the year 2001, making it look like the growth in recent years is something other than linear. He even states "Note that the number of packages seems to be growing exponentially."
The truth is, he's crammed two years of growth into a one year slot on the graph, making it appear to be accellerating. In actuality, if you imagine that growth spread over two years (as it actually is) it looks damn linear.
I guess even volunteers without corporate agendas are subject to fradulent data analysis.
Erik
There is an article in the german Linux Magazinthat gives some explanation about new 'version mixing', 'downgrading' possibilities in the new apt software under 'woody' (sorry only german language). So you can keep your base system stable and add some additional software from 'testing' or 'unstable', or you can try the 'testing/unstable' release, and if those releases are too buggy, you can downgrade to 'woody' again.
- Yasa
The plots in chapter 2, The Past, are somewhat misleading in that they suggest a steeper growth rate lately that is actually true:
:-), while the year 2001 is missing (making the latest growth seem exponential).
The time scale (x axis) is nonlinear!
The year 1996 is listed twice (thus making 1996 a particulary long year
Debian does grow rapidly, but not *that* rapidly.
--- The light at the end of the tunnel is probably a burning truck.
The Debian package format was introduced in 1995. The RPM package format was introduced in... 1995.
In terms of age they are pretty well matched. Red Hat was a growth of several other earlier package managers (RPP, PMS, PM) which may have made the transition a bit easier.
Probably the biggest thing Red Hat had going for it is that Caldera providing some of the funding for the initial development. In other words, it was multi-distribution from the start, whereas Debian for a long time was proprietary to their distribution (yes, I realize there are now several Debian based distributions out there).
Matt