“A group of leading developers calling themselves Dunc-Tank is preparing to pay selected Debian developers to complete specific projects. But although Dunc-Tank’s first goal is the practical one of ensuring that the next version of Debian is ready for its scheduled release, its announcement has also publicized a previously private debate about what happens when free software developers suddenly receive pay for what they are already doing for personal reasons.”
The article summary reads “Dev’s” singular but it should be “Devs'” plural since it is the motivations of more than one single developer that are under scrutiny. (The original article title correctly uses the plural form).
Well, I was forced to shorten the title as it did not fit. The above is still correct, as you can also speak of a single developer; the title is still correct.
However, I agree with you the plural form looks better. Let me change it.
Shouldn’t you get paid for editing other people’s articles?
At one point early in the history of the GNU project, Richard Stallman was homeless; he was sleeping in a borrowed MIT office that the AI Lab was letting him use. Apparently there are some who think that the Debian leadership should make similar sacrifices, or else squeeze in a bit of Debian work between their work and family commitments, no matter what kind of negative effect that has on the project’s ability to ship a release.
Self-destructive behavior? No, no, no. It’s what we call democracy and freedom of speech. In all democratic decision-making it’s important that there exists an opposition that points out the potentially negative outcomes of new experiments.
Apparently there are some who think that the Debian leadership should make similar sacrifices, or else squeeze in a bit of Debian work between their work and family commitments, no matter what kind of negative effect that has on the project’s ability to ship a release.
No, there are people who think that Debian project has failed, or is not sustainable, if it needs to burn a couple of people to get a release out every time. Ideally, it should be possible to get a release out without burning people to squeeze time between their day job and family, or without paying them to help. If Debian project is to continue as a volunteer project, a better structure is needed.
It’s a bit hard to see what the fuss is about. Dunc-Tank are only proposing to fund two devs and only if needed for a short while to help get Etch out on time. If the idea works, well yes it might be extended. But we are nowhere near that yet.
Debian is a formidably complex project (one I love) and if this is what it takes, why not? The guys on the Dunc-Tank board are serious players, after all. When it comes down to it, do you trust them? Yes or no.
I can see why some folks might object to Dunc-Tank: it feeds money into a volunteer project and means that some contributions might appear to be more equal than others. OTOH, there is a lot more pressure on Debian these days, I would guess. Plenty of other distros are hungry for what Debian have – CentOS, Ubuntu, SuSE, etc.
Just my 2 cents, but the days are over when Debian could perhaps afford to dictate terms to its users and live apart from the world on a “release when ready” basis that has absolutely no time frame at all. Look at the soap opera that produced last time around. You can talk principles all you want, but they also need to be matched with some practicality. These days it is much easier for your users to up and sticks and go if they think your development process is unreliable in some way.
I’m just a Debian user (Testing). Pay or no pay, I’ve no doubt at all that the next release of Stable will be delayed if the Debian leadership feel the quality is not there. To do otherwise would be suicidal. But if the Dunc-Tank move helps them to meet their targets, I cannot see it as a problem. There are limits to how much you can ask anyone to do, let alone for nothing.
Edited 2006-09-21 22:40
It’s a bit hard to see what the fuss is about.
Same here. If it were called paying them a bounty, would anybody notice or care?
I mean, I think it’s wonderful that a lot of OSS coding is volunteer work done as of a labor of love. But there are times that, for the sake of expediency, its necessary to hire somebody to write code 8 hours a day, 5 days a week, or to dangle a bounty to keep people really motivated.
Edited 2006-09-22 00:10
I agree, particularly with the point of expediency. The ambition of the project — or more aptly — the ambition of the Free Software Movement is to increase the body and quality of code that is available on a freedom basis. To me that has always meant that the ethical question involved relates to the actual freedom of the software itself. Kudos to everyone who donates their time and efforts gratis towards that effort; never-the-less, the question of someone recieving pay seems like an orthogonal issue to me compared to the actual goal of furthering the status of the Free Software Movement. Advancing free software requires work and where there is lack of gratis resources to do the work, there is room for incentives to produce the desired outcome.
After-all the desired outcome is to better everyone in society — I do not see why that should imply that the only suitable way of doing so would bar all forms funding for required labours. As long as the software remains free and developers are free to choose to either donate their time or be paid for it at their discretion, I’m all for this.
Works for me. I would like to see it expanded more actually….
I am amazed that many people consider this as a non-problem. You can’t imagine what will happen to the volunteer-based project when only some of its members are paid? I am speechless.
Have a look at this sarcasm by Josselin Mouette, a very active Debian developer for example:
http://np237.livejournal.com/9564.html
Does this really create any additional “problem” between those that are paid and those that are unpaid? Let’s face it: those that are going to work more than a mere part-time on something still need to pay bills, and “Thanks for your work!” and other such things doesn’t pay the bills. You can’t realistically expect people to pour their lives into a free project beyond what they can afford to do, and to get things done more readily requires long unbroken chunks of time, because many such things have a long ramp-up and ramp-down time before effective and efficient amounts of work are done. All those that want to do something for platitudes only will still do it, and those that can’t afford to do that, but are capable of doing it full-time if their survival needs (and preferably at least something a little more than that) are fulfilled, then where is the real problem? Is it the issue of money? Well, FOSS isn’t supposed to be all about doing things for no money, is it? It also isn’t about who gets paid and who doesn’t: the overall goal is software that is free and open for people to have full access to the code and standards used, which this still accomplishes quite nicely. What it comes down to is this: it is hellishly unfair to expect people to give their resources in any form of time, money, etc. beyond their ability to pay, and nobody has the right to insist that anyone do so, either for pay or for free. Everyone has the right to choose to participate as much as they want, either for free or pay, until they sign some sort of contract that binds them beyond what they did purely voluntarily before.
If someone wants to be such a “It must be free without charge!” purist, they should shutup and get writing the software themselves, and lead by example!
You can’t realistically expect people to pour their lives into a free project beyond what they can afford to do, and to get things done more readily requires long unbroken chunks of time, because many such things have a long ramp-up and ramp-down time before effective and efficient amounts of work are done.
No one expects volunteers to pour their lives into Debian. Debian is a volunteer project, so if Debian can’t find long unbroken chunks of time from someone, well, that’s unfortunate. Paying someone may be “effective” and “efficient” — but then it’s no more a volunteer project, is it?
Effectiveness and efficiency is not the ultimate goal, is it?
Where is the real problem?
If you can’t see that, well, it’s hopeless to argue with you. Also, you are exhibiting a typical strawman strategy (but I don’t assume any malice). For example,
What it comes down to is this: it is hellishly unfair to expect people to give their resources in any form of time, money, etc. beyond their ability to pay, and nobody has the right to insist that anyone do so, either for pay or for free.
Who said otherwise?
Edited 2006-09-22 05:08
You can’t imagine what will happen to the volunteer-based project when only some of its members are paid? I am speechless.
You don’t have to imagine; just take a look at the Debian archives, and you’ll see that happened to a lot of software distributed there. Linux, several GNU projects, Xorg, GNOME, KDE; all these projects have a healthy mix of volunteers and people employed to work full-time. None of them got into a flamewar about it.
Linux, several GNU projects, Xorg, GNOME, KDE; all these projects have a healthy mix of volunteers and people employed to work full-time. None of them got into a flamewar about it.
GNOME certainly got a flamewar about its bounty if I recall correctly. In case of Xorg and other many projects, it’s not the project itself that funds the developers. But in this case, can you see Dunc-Tank really “outside” of Debian? That is the issue.
GNOME certainly got a flamewar about its bounty if I recall correctly.
GNOME had full-time developers from the start, from Red Hat. Other projects I didn’t mention before are Mozilla, OpenOffice.org, etc.
But in this case, can you see Dunc-Tank really “outside” of Debian? That is the issue.
In my view it’s clearly obvious that it’s outside of Debian. Does the money come from Debian? No, it doesn’t, so there you go.
I hope Microsoft is listening to this one. Maybe they’ll learn from the Debian devs that paying people to work full time on free software is the way to kill it, and then they’ll start paying some devs to “eliminate the competition”.
You can’t imagine what will happen to the volunteer-based project when only some of its members are paid? I am speechless.
I would ask if I even want to get paid. No. Seriously.
Money comes with all sorts of strings attached.
A volunteer can quit at any time. A volunteer can go home at night and say, screw it, I’m going to play guitar hero/kingdom hearts/counter strike instead.
A volunteer chasing a bounty has to weigh taking time off vs how badly they need/want the bounty.
But somebody who’s getting a paycheck has a contractural obligation and if they don’t produce in a timely manner there are repercussions.
Another random reaction from a Debian developer which may help other readers:
http://alfie.ist.org/blog/2006/09/21
I think this is a bad idea.
If this was going to be done it should of been users setting it up and managing it. Something truly seperate. I am not sure the funding should be directed specifically either. Maybe round robin loot dispersion or something. Maybe bounties.
Maybe it is time for a “new” debian and a project that is “old”debian?
It’s called Ubuntu, dumbass.
Yea, and I guess you would also say it is freespire. YEAAAA OKAAAYYYY!