Debian GNU/Linux 4.0, codenamed Etch, had been due to arrive by December 4, 2006, but it’s been delayed because some developers have ‘deliberately’ slowed down their work. According to a blog note by Andreas Barth, Debian developer & release manager, the delay has resulted because “Some people who used to do good work reduced their involvement drastically. There was nothing I could do about, and that happened way before I started full-time on release, but on the global picture that still counts.”
Disgruntled Debian Developers Delay Etch?
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSAlert.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I suppose this means Etch will be outdated at the time of it’s release?
Has Debian ever NOT been outdated at the time of its release?
Has Debian ever NOT been outdated at the time of its release?
Yes.. Obviously, you have not been around long enough to know that.
“””
//Has Debian ever NOT been outdated at the time of its release?
Yes.. Obviously, you have not been around long enough to know that.
“””
I’ve been in the Linux scene since 1996 and Debian has been behind the times pretty much the *whole* time.
This was going to be *the* time that they were current.
Predictably, some moral issues (which only Debian devs could possibly understand) have closed in to restore what I consider to be “normality”.
I use CentOS and Fedora.
Edited 2006-12-19 03:11
While i have been using Debian off and on since about 1997, I will say that generally Debian stable is usually a few months out of date upon release, but this is only because they usually go into freeze for several months to make sure that everything works as it should.
You really honestly can’t compare Debian and Fedora Core or CentOS. The reason for this is the sheer number of packages that Debian uses.
The ONLY reason that Ubuntu can manage it’s release schedule and Debian cannot is because Ubuntu has a small (relatively) amount of packages that are in it’s main (read supported) repository. All others are in Universe or Multiverse, which is not supported, and in general break a lot of times, sometimes with no fixes in sight. A good example of this is the gnome-sword package. The version in Edgy Eft depends upon libgnomesword5, but only libgnomesword6 is in the repositories. No one has bothered to fix this, even though there is the “Christian Edition” of Ubuntu, but I believe it’s based on the Dapper Drake release.
The thing about Debian Stable is that it IS stable. I would compare Debian Stable to Red Hat, NOT Fedora Core. Fedora, every time I have used it, has not been the most stable of operating systems. It’s probably on the scale of a little less stable than Debian Unstable without experimental packages.
I have been running Etch for some time, and it is actually quite up to date. Granted it’ll slide back a little bit due to it now being in Feature Freeze, until they can squash the last few release critical bugs. It is also out of date with Gnome 2.14, but that is due to a bug within GTK 2.10. I wouldn’t honestly mind if they delayed Etch long enough to fit in Gnome 2.18.
Currently I run Ubuntu on my Desktop and Laptop, but I run Debian on my Server and on my HTPC. Debian is perfect for the latter two, simply because you can completely customize everything from a very minimalistic start. Ubuntu isn’t really good for that, even the ‘server’ install puts too much on for those who like to keep it lean.
Also there are quite a few underlying packages that are generally still newer under Debian than there is in Ubuntu.
The thing about Debian Stable is that it IS stable. I would compare Debian Stable to Red Hat, NOT Fedora Core. Fedora, every time I have used it, has not been the most stable of operating systems. It’s probably on the scale of a little less stable than Debian Unstable without experimental packages.
Fedora Core “a little less stable than Debian Unstable without experimental packages”?
(Btw, who runs Debian Unstable without experimental packages?)
Sorry, but that’s, imho, an exaggeration at the expense of Fedora Core. Compare it to Deb. Testing, that would be fair.
Edited 2006-12-19 16:16
Most things in unstable are actually stable upstream versions.
Experimental is somewhat of a dumping ground – an entirely seperate distro comparable to Rawhide, and most users have no idea how to even access it, let alone actually use it.
This can be evidenced by the fact you apparently don’t even know about it.
Experimental is somewhat of a dumping ground – an entirely seperate distro comparable to Rawhide, and most users have no idea how to even access it, let alone actually use it.
This can be evidenced by the fact you apparently don’t even know about it.
True, I didn’t know “experimental” was a name for something particular in Debian land. I figured “Unstable” would reasonably be less stable than “Testing”, which I found to be comparable to Fedora Core in stability (I had full X-crashes with Testing and Nautilus crashes with FC5, and found the former slightly more annoying).
However, if Debian’s distro nomenclature is indeed misleading (nofi) in this respect, I withdraw my above post and will first try Unstable, or refrain from talking about Debian.
I’ve ran Fedora Core for only a short time and it is definitely more comparable to Debian Unstable.
Apparently you don’t know how Debian works. A new package is introduced into Unstable, then after two weeks there of testing to make sure there is no broken dependencies or outright bugs, it gets put into testing. Fedora Core is basically the testing ground for Red Hat. They generally put packages into it just because it is new.
The third party repositories for Fedora Core are far more important than Experimental for Debian. In fact a lot of times I’ve seen packages in Unstable being newer than the Experimental ones. The only Experimental packages I usually use are the nvidia drivers, nothing else.
I suppose this means that Debian will continue its long-tradition of outstanding cross-platform support and quality releases that deliver unmatched stability, not just the latest version number of an untested piece of software.
Debian has, in my opinion, the strongest quality assurance of any distribution, but hey it may not have the absolute latest package. So what?
Absolutely. Debian is rock solid. Most people simply do not need the latest and greatest, sadly, society has taught us from a very young age that we must have the latest and greatest.
For those that don’t like Debian’s ideals, or the fact that it’s not supreme cutting edge, go use Ubuntu or Mepis and stop bitching. No one’s making you use Debian.
Just remember, it if wasn’t for Debian, Ubuntu and many others would cease to exist.
Dave
I suppose this means that Debian will continue its long-tradition of outstanding cross-platform support and quality releases that deliver unmatched stability, not just the latest version number of an untested piece of software.
Debian has, in my opinion, the strongest quality assurance of any distribution, but hey it may not have the absolute latest package. So what?
Not only that but for those that want the latest and greatest of everything can always use a healthy mix of Etch and Sid which is no more unstable than most “stable” releases of others distros.
To me this seems to be kind of one of those “No shit” moments. If you single out a small group of developers among the group and basically send a message saying their time is more important than anyone else’s, therefore they should get paid; this tends to piss off the rest of the group who are being told their time and effort is second class.
To me this seems to be kind of one of those “No shit” moments. If you single out a small group of developers among the group and basically send a message saying their time is more important than anyone else’s, therefore they should get paid; this tends to piss off the rest of the group who are being told their time and effort is second class.
The Debian folks were very careful not to send this message. Getting a new Debian Stable out the door involves a great deal more than simply squashing bugs in packages A, B and X. It involves a great deal of organizational complexity, attention to detail and careful prioritizing. In a word, management. That, so far as I can see, is what the Dunc-Tank folks were/are being paid to help with.
If a few thoroughly immature, prima-donna devs wilfuly refuse to see this and genuinely think that squashing bugs in their pet packages is all there is to a new Debian Stable, then Debian is better off without them, imho. In the short term this might cause a delay but in the long-term it will improve the stock of talent at Debian generally. Self-indulgent stupidity with “to hell with our users” tacked on has been the bane of Debian for too long now.
If a few thoroughly immature, prima-donna devs wilfuly refuse to see this and genuinely think that squashing bugs in their pet packages is all there is to a new Debian Stable, then Debian is better off without them, imho.
It would seem not. Doesn’t look like Etch is going much of anywhere without those “thoroughly immature, prima-donna devs.”
This
That, however, did not turn out to be the end of the matter. Many developers, led by Joerg Jaspert, a well-known Debian maintainer and programmer, issued a position statement on October 26. In this statement, which was published on the Debian developers’ announcement list, the developers spelled out why they objected to the Dunc-Tank initiative.
for some reason doesn’t strike me as immature or any of the other ridiculous things you came up with (obviously without you having actually read the article).
Edited 2006-12-18 22:40
If a few thoroughly immature, prima-donna devs wilfuly refuse to see this and genuinely think that squashing bugs in their pet packages is all there is to a new Debian Stable, then Debian is better off without them, imho.
—————————–
It would seem not.
Perhaps, but only if you look at the short-term. In the long term, however, it’s inarguably better to amputate a gangrenous limb, even if the amputation does cause short-term trauma.
Doesn’t look like Etch is going much of anywhere without those “thoroughly immature, prima-donna devs.”
No really, what *is* your opinion on the economics of Chinese tea retail?
Doesn’t look like Etch is going much of anywhere without those “thoroughly immature, prima-donna devs.”
Dunc-Tank was clearly voted on and clearly accepted by a majority of Debian voters who were and continue to be given the skinny on what Dunc-Tank is and what it’s doing. An attemtpt to recall the Project Leader over the matter was decisively rejected.
In most sane, normally organized outfits that would be the end of the matter. I guess only in the strange world of Debian does it then have to be pointed out that damaging the project after it’s been so clearly voted on is a bit reprehensible. If some devs feel that strongly against Dunc-Tank, the correct action in the circumstances is to resign.
What do I say all this? Because I like Debian very much and don’t at all like the project’s apparent propensity for self-destruction.
Edited 2006-12-18 23:19
What do I say all this? Because I like Debian very much and don’t at all like the project’s apparent propensity for self-destruction.
Absolutely agree. The majority voted in favour, the last thing Debian needs for its rather unpredictable release schedule is a few devs acting like petulant children.
There needs to be a single payment pool that companies can donate too, and then an equation can be used to compute pay for ALL developers from that. The equation could take into account that developers working more hours / having worked for more years / write more lines of code / etc get a bigger slice of the pie.
Edited 2006-12-18 22:16
A meritocratic payment pool with objective measurement is fine in theory.
In practice I guess people will start to behave differently in order to fudge the stats in their favour.
The whole thing is tricky and maybe some folk were reaching their moment to move on and this is the straw that breaks the limit.
I do not agree. Splitting resources up according to such an algorithm would probably delute them to such a degree that no one would really benefit much from it. The sum of $50+-X or so pay-checks per developer could probably be spent much more efficient by paying a few temporary (near) full-time salaries for a month or two to help release new stable versions, just as what is done now. The important thing is to have the *Debian community* elect or re-elect devs that would get this short-term financial backing, which I think creates an scholarship platform that in the long run would favour technical/leadership excellence and recognise and award people according to their degree of contribution. This is more in line with the “academic” and democratic spirit I think is part of the Debian project.
How does the structure of Debian’s organization differ from the one around the Linux kernel development? The kernel has both paid coders as well as volunteers, and I haven’t heard of friction there slowing down releases. How is Debian different? Is it simply that it lacks a strong SABDFL?
Linux doesn’t have this problem because virtually all the developers are paid.
Please recommend an alternative unbiased OS news web page.
Here you go chappy:
http://www.google.com
They cater to everyone there!
Slackware is the only stable, time tested and fully supported alternative there is!
No “Dunc Tank” debacle with Pat V. The Man has it all under control…
Thanks and happy holiday Patrick. Thanks for a great distribution!!
Not really. Slackware is ok tho I have a few objections that makes it impossible for me to use it:
– installation routine is very poor
– proper package management is misssing
– depends on _one_ person (single point of failure)
– no 64-Bit support
– no GNOME support
– partly as much outdated as Debian
And no, I don’t want to have second or third hand support. I want it to be part of the system.
no 64-Bit support
http://www.slamd64.com/
no GNOME support
http://www.droplinegnome.net/
http://gsb.freerock.org/
partly as much outdated as Debian
Besides having a rather old-ish monolithic X.org (6.9) and having a 2.4.x option (2.6.17 if you rather use a 2.6 kernel), what else is “outdated”?
– Gilboa
FYI, I use Fedora for workstation, Debian for testing and Slackware for low-end/embedded machines.
/quote/
– installation routine is very poor
– proper package management is misssing
– depends on _one_ person (single point of failure)
– no 64-Bit support
– no GNOME support
– partly as much outdated as Debian/QUOTE/
I’ll address these fallicies one at a time:
“- installation routine is very poor”
Says who? The install is ncurses based and is very easy and simple to use. By “poor” do you mean that it doesn’t autodetect everything and set it up for you? Slackware gives you the power to configure your system your way without getting in the way.
“- proper package management is misssing”
You mean it doesn’t resolve dependencies and hold your hand like Debian does? Slackware gives you the control without the bloat and hand holding. The package DB doesn’t get corrupted like Debian’s can and does.
“- depends on _one_ person (single point of failure)”
I’ll take the streamlined development of Slackware over the politics and bickering that is prevelent in the Debian community. Slackware did not get the be the oldest distro in continual development by producing a sloppy distro. It is what it is and where it is due to the craftmanship that goes into it.
“- no 64-Bit support”
There are unofficial ports to x86-64 and when 64 bit reaches critical mass Slackware will be there…
“- no GNOME support”
As Pat V has stated in so many words: This is open source…you want Gnome? Build it yourself or get it from third party packagers–isn’t hard to do at all…
“- partly as much outdated as Debian”
You’re joking, right? Though Slackware uses sane defaults and tends to remain conservative it does include up to date packages. Plus, any competent Slackware user can easily compile their own updates and make their own packages for the system. After all, Slackware comes with the tools and libraries to do this. In Debian I guess you have to wait for the outdated package to be updated with a slightly less outdated package…or was that just a bug fix? (Those who live in glass houses…)
“…I don’t want to have second or third hand support. I want it to be part of the system.”
Security updates and bug fixes are routinely handled and supplied in a timely fashion. As for “official support” there is the support forums and email support. Debian is non-commercial so there isn’t any “official support” for it. Besides, there is a great and robust community around Slackware. Lastly, as Pat V said in a recent 90-minute interview, Slackware users tend to be self-sufficient in the support department. So a 24/7 support is unecessary (but you get that in the forums and irc groups).
Slackware Rocks!
Edited 2006-12-19 19:46
Not that I’ve actually used Slackware (it’s one of the few that I haven’t actually), but some things about Debian.
In all the years (about 9) that I’ve used Debian, only ONCE have I ever had a package database become corrupted, and it was an easy fix. On RPM based distributions I’ve had it happen multiple times and it was always to the point where all the different ways on the net would not fix it.
Also you can compile things just as easily on Debian as you can on Slackware, it’s just easier most of the time to install a package.
As far as the installer goes, Woody was the last Debian distro to have something much like what you describe the Slackware installer as. Sarge was still ncurses based, but had great hardware detection, Etch will have both GTK based and ncurses.
Debian Rocks as well, just different strokes for different folks. Slackware to me seems like just one step different than Gentoo. But even Gentoo resolves dependencies. Pretty much any Linux or FreeBSD does now days. If Slackware doesn’t, then that’s just part of that system, and if you like it, then by all means stick to it.
Debian packages are easy to understand the versions on. You have something like 1.2.3-4. So that would be package version 1.2.3 and debian revision 4. Not exactly all that difficult, is it?
THere is no such thing, and in this particular case, why is this biased?
@porcel – This argument makes the assumption the upstream packagers do no testing on their own. Yes there are integration issues but Debian is on the extreme end when it comes to release policy. This is to the point where it is outright silly, and insignificant bugs hold up much needed upgrades (which themselves fix bugs, security, performance).
—-
Ubuntu has managed to find a perfect balance of testing, currency, and release timing. Not to mention bickering of paying leaders to expedite development is nonexistent. Its not about their work being more important and its not like they are profiting. It is about them taking time out of their lives (which requires subsistence) to focus on an objective that requires the work of a full timer, and the simple lack of funds that more people cannot subside off of FREE software.
For childlike politics like these, Debian has become largely irrelevant and given rise to the successful Ubuntu project, which focuses on delivering technology, not politics.
ubuntu builds off debian’s unstable branch. debian is hardly irrelevent
Please, let’s not turn this into another Ubuntu vs. Debian thread. We’ve had plenty of these here at OSAlert already during this year. These two projects are highly different with different goals, and the way they are intertwined grants recognition to both of them. If you can not see that you need to read up a lot on these projects, and perhaps most so on the history and accomplishments of Debian. The strictness of Debian has helped the FOSS community in many ways, for instance bringing about the Qt GPL licence if I am not mistaken.
This article is based on speculation, not on facts. There’s absolutely no proof that Etch would have been released on time if there wasn’t this Dunc-Tank business. No-one was paid to supervise the Sarge release and look how that went. Etch has so far turned out considerably better than Sarge.
Etch is now in full freeze prior to release and the developers only need to complete the final version of the installer and to squash the remaining RC bugs (or downgrade their severity, or simply drop the buggy packages from Etch). Whether some developers are disgruntled or not shouldn’t have much effect on that result.
i must say i agree with the group, it should be an all pay or no pay thing and they were all basically happy with how things were before this dunc-tank came along… and i totally agree with this being a ‘well-duh’ moment. of course if you pay a few peeple for doing work they are all going to want paid, hoestly i think the debian dev’s should be paid but dont see how to make this possible. there for i just humbly thank them for the work they do on the best distro ever, every chance i get.
It’s a free and open source project, built on volunteer supplied labor. If some companies want to sponsor a group of developer’s work, then let them, but that doesn’t mean that every deserves to get paid. What that means is that a group of developers found someone to pay them for their work. Nothing is stopping the rest of the volunteer developer base of Debian from organizing and looking for funding.
If the moneys were coming from the Debian Project directly and going to pocket groups, then I could see a problem. If they were paying the release leader and other folks like that, then I don’t see as big an issue. If every developer is on an even playing field with regards to the Debian Project itself, then there shouldn’t be a big problem. The folks that are leads are either elected or appointed by elected officials, so in my mind, money exchanging hands from the project itself seems more fair.
When money is coming from the project to some developers and not others, then I think there is a problem. If the money is coming from third parties to certain developers, good on the certain developers for raising the cash.
This article quotes a blog note by Andreas Barth. It’s interesting to notice that there’s now a small update added to that blog entry, pointing out that Barth doesn’t support the conclusions of the article.
http://blogs.turmzimmer.net/2006/12/18#20061211
Other groups contribute money to Debian developers. Should they be boycotted too? If Dunc-Tank wants to pay for someone to help develop etch, then that’s their business. Many of the developers contributing to Debian do work on a volunteer basis and knowing full well that they were not going to receive a paycheck. If some of these developers disagree then they may voice their opinion to the Dunc-Tank group. Holding up etch development seems to go too far and only hurts the Debian project overall. It serves no purpose but to destroy what others have worked very hard to build.
If you want *edge then you use testing/unstable. If you want a stable OS then you use stable. The only way to know software is stable is to beat on it a while so nothing truly stable is released last week. So yes, stable will be behind the *edge somewhat.
Some tasks require a full time commitment. Whether you think they should or not, they do. All DT did was pay several people to go from part time status to full time so that the release could be finished. Lets face it, you cant expect these guys to quit their day jobs to do debian full time without pay. They got bills to pay too.
I don’t understand the objection to being paid.. what if someone was working on Debian proper for a while, who everyone else assumed was on purely voluntary basis, turns out to be privately paid specifically by someone else for his work? “OMFG you cheater! You’re supposed to be working for freeee!!!”
And what logical difference would it make if that “someone else” turns out to be involved in the Debian group? However, since there is no officially policy of paying developers and since DuncTank itself is technically an outside group, the funding of developers really seems to be a matter that is ultimately private to the individual.
Of course I’ll admit that it can lead to perceived favoritsm and bruised egos.. and while it’s certainly natural to feel that way (including myself for a bit) I think it’s hypocritical to judge that way. Rather people should resolve to being glad for others’ good fortunes especially since it’s involved in contributing to a shared interest and that they could now afford to put even more time towards towards improving that shared interest.
I very much agree with you. And being able to make a living from helping a project like Debian and its user community for some time would be a fantastic experience. Remember that in the end, Debian is all about its users, which may or may not happen to be Debian developers as well.
A problem could potentially arise if there are unexpected conflicts in interests from the community compared to the sponsor though. It is important that priorities should always be set by and for the community.
I am downloading Debian 3.1r4 (DVD’s iso images)right now, should abandon that download and wait for Debian 4 release or would it be wise for me to continue with the download. I get around 256 kbps (~ 30 KB/s) at home and it would take me easily more than 3 days to download the iso images. Thanks.
keep going.
>> I am downloading Debian 3.1r4 (DVD’s iso images)
>> now, should abandon that download and await Debian4
>> release or would it be wise for me to continue with
>> the download. I get around 256 kbps (~ 30 KB/s) at
>> home and it would take me easily more than 3 days to
>> download the iso images. Thanks.
My approach would be:
1) Download the CDROM image rather than the DVD image;
— 95% of your needs will be met from the CDROM and
— the package manager can easily pull the remainder.
2) Use BitTorrent to pull the iso. Get on with life
— and the isos will appear.
3) Expect a large quantity of updates when you have
— installed.
Bon Voyage.
If you want to use Debian at home, you don’t want Debian stable. You don’t want all the ISOs either.
Get the etch RC1 installer from here: http://www.debian.org/devel/debian-installer/
Either the netinst CD (150MB, install the rest as you go) or the first 1 or 2 CDs from the full set. Upgrading to the next stable release should be smooth and automatic.
If you decide to follow the testing branch, expect frequent updates and an up to date computer for the rest of its lifetime.
If this sound a bit too much, don’t get discouraged. Debian is fun! One of the best out there.
Tim Holwerdi
Hi, My name is Tim Holwerdi.
I am gonna tell you my last dream…
I am an Aszzhole in search of Notoriety…
I work in a Website that offers news of IT and Open Source.
I pretend that I do it for the sake of love for IT, but the fact is that, I am expecting good revenues for the
future…
If not, why should I loose my time looking for IT news in other IT Web Sites that offer what I am not able to
offer… for the sake of these IT weirdos geeks and Open source-free computing fanboys…? c’mon…
I think I know more than the rest, of course… and I am always right!
Yes, I know more than anyone of you about Computers, and about anything else you can imagine! even If many people prove me the contrary, I am still right…
Me and my Mac go together everywhere, I even sleep with it, which is somehow problematic, cause as you can imagine, is not easy to have sexual relations tru an USB port, or a FireWire one… but I am in love anyway!…
Anything that is not Mac or commercial, is just wacko rubbish!
And, of course, is not going to offer me anything, because all these Open Source weirdos have no future, and are not gonna advertise in my site, or pay me money… I dont even talk about the FSF retarded hippies!
At best the big companies that now move to Linux, and pretend to be Open Source, worth a little bit, and may be a source of revenues in the future if the have some sucess…
Cheers…
P.S. Apple Rocks… Linux sucks… (MS is very good also, cause they have plenty of money, and are the pattern of our great western Businnes Economic and social system…)
That’s good advice dimosd.
I always use the Netinst CD and only yesterday tried out the optional GUI installer to see how it faired. I had to say I was quite impressed. Apart from the mouse point changing to a text input cursor but not back again I had no problems.
Although I was left a little puzzled as to why Gnome hadn’t been apt-get’d during the installation. The first time I fired up Apitude there were a ton of more packages waiting for me to download and install!
I find the testing branch isn’t too bandwidth intensive from an update point of view, however, unstable is another story entirely!
…this story is actually correct, I have to say the developers are childish.
DT doesn’t have anything to say about the development process it self, doesn’t make any decisions about packages to be a part of Debian, nor does DT have any influence in any way.
It’s a developer initiative, paying two developers in order to secure schedules. The result is the rest won’t work because they _feel_ it could lead to a two-class system. That claim is incorrect, since DT doesn’t have anything to say about what to be included in Debian. It merely pays two developers. And it is no more than that.
So if you were in a pool of volunteers and everyones work benefited everyone else and THAT was the payment, that suddenly a few of those volunteers setting something up to pay two of those *volunteers* $6000 each would not bother you in the slightest?
I would obviously be somewhat envious, who wouldn’t? But personally people can donate to whomever they want, as long as the receivers of the donation aren’t granted more rights than anybody else.
The latter would piss me off.
Exactly! And a volunteer can put in less effort if they feel they aren’t appreciated. Certainly a volunteer is not expected to perform a certain amount of work, especially the same commitment that a *volunteer* who has $6000 in his pocket puts in.
So if some of those volunteers are a bit put-off and have slowed down some then I understand it. People put in time and effort into something they enjoy. If something has happened to make that person enjoy it less and they put in less time and effort then I don’t think anyone could hold that against them. At least I don’t…
Nooo no. You can’t twist my words that way
I still think they are childish, since the devs who receives money aren’t granted more rights than anybody else.
Being envious doesn’t make it okay to act this way.
All that “feeling” stuff is hilarious. There is no reason for them to feel less appreciated. Not getting the money isn’t a reason, unless it is specifically said that those who got the money did so because they were much better.
Nobody has come with such a statement, therefore there is no basis for “feeling” less appreciated. It’s childish, and as grown-ups they should know better.
Now, if the paid developers had more rights that would be a major problem. Right now we just have a bunch of laze whining childish developers, who cannot accept a decision made by a majority. Booo-hooo.
Money doesn’t divide people – rights divide people.
But certainly a volunteer does not have to justify putting in less volunteer time do they?
They have chosen to put in less time because they are not happy about something.
I can’t hold that against them, they are volunteers after all and any time they are willing to give is appreciated. If something has come up and they feel like volunteering less then I am just happy and thankful for whatever time they are willing to donate.
It isn’t like I am paying them after all…
Of course a volunteer doesn’t have to justify anything, really. But when they justify it with these reasons, they _are_ childish. But they are of course entitled to be so, but it’s still childish to complain about something being donated to somebody else than themself. When that’s the reason for them doing less work it _is_ childish, and one is entitled to bash them for that.
If they perform less work because of the money then they aren’t doing this with their hearts and therefore do not deserve any praise.
You might have a good point…
I dont think they are doing less work because of the money. I think they are unhappy that dunc-tank creates a situation and it does reflect on debian and so they are upset about the whole thing.
So if dunc-tank had raised a million dollars and gave it all to one person – no difference? Ten million? What if the money was given to the DPL himself?
So if dunc-tank had raised a million dollars and gave it all to one person – no difference? Ten million? What if the money was given to the DPL himself?
Hmm I understand that can look problematic.. I think that determining whether giving millions to 1 or few selected really is problematic needs to also take into consideration the interests of the sponsor. It’s as korpenkraxar mentioned, if the sponsor(s) are giving millions in order to steer Debian into their own direction then yes, that would definitely be bad thing. (not unlike politcal lobbying, lol)
But if a few received millions as a reward from their sponsor(s) for doing work that benifits the sponsor but is aligned with Debian.. then I don’t think there’s an issue. Think about all the highly paid developers working full time on linux in Redhat, Suse, IBM, HP, etc. Fictitious example: you might have many volunteers working on the firewire framework and firewire devices, while you might have a handful of developers working on iSCSI whose combined gross pay could very well be in the millions, over say 2-3 yrs and more so when you factor in the implicit benifits like healthcare, 401k, paid vacation, travel, conferences, classes, etc
Edited 2006-12-20 10:21
Money doesn’t divide people – rights divide people.
So if a woman and a man perform the same job, then the man being paid twice as much is not divisive or would not cause some *feelings* of any sort?
Edited 2006-12-19 19:18
Divisive? No, not if the payment is a result of donations or is otherwise voluntary. If the payment is an obligation like in a company the situation could be different, depending on the situation and contracts between employer and employees.
I already stated that there could be feelings involved, like being envious. But one thing is being envious, another thing is acting like a child.
Just because one have feelings, doesn’t mean one acts on all feelings, right? It’s a matter of selfdiscipline.
Perhaps what the Debian project really needs a quality screenwriter – perhaps David Mamet, to punch up the dialog and characters a little bit. Make all of this business read a little more like Glengarry Glen Ross.
We need more passion, rage, and betrayal, and less grouchiness.
That’s my suggestion.
it’s either:
A. They’re zealots for the stereotypical “Debian Way” of not releasing jack till it’s far behind the rest
B. Those people truly are busier this year and have legitimate reasons to have slowed down (hey it’s reasonable, happens to us all)
C. They are working “extra hard” on this release and certain developers decided to go against the schedule because “it just takes so long” *takes baby steps on purpose*
D. That group lives on the Gaza Strip and shit happened.
lol take your pick, they’re all equally possible, except maybe D which is the only one that can truly be identified.
OR
The fact that an article was written about it brought it WAY too much (unnecessary) attention by putting it into the spotlight as if something were wrong, when really other distros do this all the time anyway….whereas this writer simply put it out of proportion, (see: FC6 planned release vs actual release) — objectively, the article could be completely off target, yet at the same time could be absolutely correct.
So, whatever?
Everyone got involved with OSS development for the love of it. I don’t see how the fact that someone else gets paid to do the same thing decreases that love.
A lot of people make a living developing for Linux. This argument seems to suggest that all the volunteers should stop until they get Linux jobs as well.
While I understand the envy/jealousy I disagree with the response.
some thoughts I hear on the workfloor:
to me it appears that either the naming scheme of debian is not really how it’s supposed to, or the bitching about it is right. or the choices are not well thought. or maybe too much thought about (look at the browser stuff for instance)
say: “debian unstable” — well it’s not unstable […snip explanation] — why not calling it unstable then. it’s useable, right ?
say: “debian stable is always behind” […snip explanation] — if a particular set starts with 2.4.x while therest of the world uses 2.6.x, it seems old, yes. and then the answer is “you can use 2.6.x, it’s there”. maybe such things leave people thinking it’s old. so maybe the decisions aren’t right and accept the feeling people have it’s old.
people also tout the stability — several other systems, newer (and in the eyes of debian an unstable environment) are just as stable…
the whole idea behind debian and it’s naming may need some rethinking.
=====================================================
I do agree on several of these points. for some customers it was in fact the reason to use SLES instead. not thta it hurts debina, but just something to think about…
The original scheme was, I think, that there are two distributions inside Debian: the released distribution and the development distribution.
Packages in the former distribution are never upgraded to newer versions, only security fixes are added or backported from the development distribution. It is called the “stable” distribution precisely because packages are never upgraded.
But packages in the development distribution are upgraded all the time, as new versions become available. Because the package versions change, it’s called “unstable.” Of course, upgrading software to newer versions always introduces potential dependency breakages, potential problems with packaging, and also potential new problems that are inherent in the packaged software itself. So this additional unstability adds a new meaning to “unstable” — the constant upgrades can break your system.
The “testing” distribution was later added to Debian. I think that the idea behind “testing” was mainly to make it easier for Debian to prepare new stable releases, because the sheer size of Debian (many supported CPU architectures, lots of packages) made it an exceptionally difficult task to prepare new releases by unpaid volunteer developers who also have real life to live beside this GNU/Linux hobby.
After packages have been tested in “unstable” for about ten days, they are moved to “testing” if no big problems have been found in these new packages. When Debian prepares to make a new “stable” release, packages in “testing” are frozen and put under rigorous bug-squashing.
From the users’ point of view, the “testing” distribution makes it somewhat easier to track a development distribution and to get newer versions of applications while avoiding some unstability of the “unstable” distribution. It’s possible to pin “testing” as the default distribution to track (see “man apt_preferences”) while also making the packages from “unstable” distribution available. This way you can track “testing” and tell APT to fetch any missing dependencies from “unstable,” and you can also upgrade some select packages to newer “unstable” versions while keeping most of your system safely in “testing.”
Another option would be to run a “stable” system and to enable upgrades from a special “backports” repository. This “backports” repository is not officially supported but it is still maintained by official Debian developers. Packages in the “backports” repository are taken from “testing” or “unstable” and then re-compiled against the current “stable” system but their possible security problems are not actively surveyed or fixed, like in “stable.”
http://backports.org/dokuwiki/doku.php
Then there’s the “experimental” repository but that’s not really a full distribution like “stable,” “testing,” and “unstable.” “Experimental” is just a temporary storage place for some new packages that for some reason or another have not yet been considered suitable to be added to “unstable.” There is usually a good reason why the packages in “experimental” have not yet been added to “unstable” — some packages in “experimental” are known to break dependencies or they may contain some other known problems.
well you excellently explained why the naming scheme doesn’t cut it.
“it’s called stable because there are no upgraded packages”.
people don’t get such things. you know how the naming is made, I know, but my neighbor thinks “stable” is something that doesn’t crash. ok, so he installs and finds all is old, very old, compared to other distributions that are stable too but containt current software.
now it’s up to your imagination why the onther naming schemes don’t fit too well either.
yo start with unstable. what’s the first word youthink of ? “my computer is unstable” etc.
that’s why i said that it should be rethinked and not given a name that doesn’t tell what it exactly is.
stable is no more stable as RH or SLES, openSUSE etc. However the latter have packages …. you see?
Yes, it might make sense to rename “stable” as “server,” “testing” as “desktop,” and “unstable” as “development.”
Debian could also prepare snapshots of “testing” say, once in every six months, and spend some time fixing any broken dependencies in those snapshots, and then release them under the name “Debian desktop release” or something like that. They could be a bit more lax with the bug-fixing of those “desktop releases,” and they could still concentrate their main efforts in preparing the less frequent and higher quality “server releases.”
In fact, I’ve seen some Debian Developers suggesting special desktop-oriented Debian releases on the mailing lists. But such ideas take a lot of hard work before they can become reality, and I’m not sure if there exists enough interest among the DDs to push the idea forward. Things are already pretty good as they are in the Debian land and there’s the old UNIX wisdom: “If it ain’t broken, don’t try to fix it.”
They already do. It’s called Ubuntu
Ubuntu is an ancient African word, meaning “I can’t configure Debian.”
No, really, Ubuntu is quite nice and it adds some stuff that Debian doesn’t have (like a brown theme and non-free drivers for laptop support) but Ubuntu is still just a subset of Debian — it supports less architectures and fewer packages. I think that an official desktop-oriented Debian distribution would be successful in attracting new users and developers.
“””
Ubuntu is an ancient African word, meaning “I can’t configure Debian.”
“””
A common misconception. A more accurate translation being: “I don’t have the freaking time to do for the computer what other people’s computers have been doing for them since 1986.”
Check this link. http://www.dunc-tank.org/background.html
Maybe it can explain a few things.