Slashdot Mirror
For The Love Of Open Source
Jim Madison writes: "Is the open source movement about the joy of hacking? The latest edition of FirstMonday has an interesting academic study that says "No!", it is only natural in our traditional political economy that software be developed with public funding in the safety of academia when the markets are immature. Have moved into a post-scarcity gift culture or is the report correct that open source uses and needs the subsidy of public investment to grow within traditional industrial capitalism?"
I've asked a lot of people and it seems to me that most people who evangelize and use OSS are using it because OSS projects are usually (but not always) free software (as in free beer not free-dom). Why did Loki fall into trouble? Because the million linux kids out there use linux because its free of cost and wouldn't spend a dime to keep a great company alive. As such, I think OSS is failing...
It seems to me there are two factors in the creation of open source projects:
Many open source programmers (Linus, the guy who started PHP, and others) say they set out simply to "scratch an itch." This is the desire/need that underlies so much of what's been done...a small number of individuals who have a burning idea, and who start making it happen for their own reasons.
But not all programmers are free to spend endless time and money on their pet ideas. If you have a very tolerant and generous employer or a lot of free time (and no spouse/girlfriend/boyfriend), I guess that helps. But it also helps if you are working in an environment -- university, gov. agency, etc. -- where the prevailing values support your work.
I.e., in a for-profit company, you are unlikely to get official recognition/resources for your open source work. But in an academic or government setting, where profit is less important than the usefulness of the software, you may well be able to pursue your personal "itch" with the backing of the institution.
Just my $0.02...
you can't label it as being done purely for joy or purely for economic reasons.
some people do it for the love of the art.
some people do it to make a political statement about our economic system.
some people do it as pure research to benefit the body of knowledge in the software development field.
to try to say that all open source software is done for reason X is a little shortsighted.
It's precisely that type of linear thinking that makes other people say 'open source is communism' or 'open source can be taken seriously because it's done as a hobby'.
as with anything in life, the motivations for any one movement are so complex that pinning them down is something of an impossible task.
Interestingly, he picked GNOME and Linux as the projects to analyze -- had he used KDE instead of GNOME the numbers would have come out much more strongly in favor of his hypothesis.
I dunno, though -- I do it for fun, not because I expect any financial gain from it.
What I'm listening to now on Pandora...
It seems to me that the hypothesis of the gift culture and the findings of this study are entirely compatible. This research does not show, as it author seems to be suggesting, that "scratching an itch" is not the primary motivation for free software development. Rather, it shows that this motivation does not trump traditional economics.
... if only they had a bit more time, or wouldn't be giving up tremendous potential income if they did...? How many times have you thought that yourself?
How many people do you know who would work on projects X, Y, and Z
Since many will post without actually reading the study, everyone should note that the raw data came only from the kernel and GNOME. I doubt that kernel+GNOME developers make up the majority of open source developers. And I wouldn't consider it an accurate sample set of developer's either. Kernel hackers are a special breed, to say the least. And GNOME developers certainly don't completely encompass the average application developer, such as command-line, internet, or just plain x-window.
I'm an open source application developer (in my personal time), and find this study does not at all include my perspective. Obviously I'm not the majority, but I think it's missing a lot.
Developers: We can use your help.
This discussion comes up on Slashdot every couple of months or so.
The article's assumptions might be true for some people, but there are still many, many people who still develop out of love for doing something useful. In fact, there are probably more today because the Internet has allowed people to contribute and connect that never could before. The network effect has been made possible by the Internet.
I develop open source software because, yes, I do love it and I want to do something useful. One of the primary reasons I got a PhD and live an academic life is so I can do this and still support my family. I have must students contribute to my projects as part of their assignments, as well. Many of them have gotten the open source bug and are contributing now as well.
I develop open source because:
1. I don't want to support it. Let people find usefulness in it. Let them contribute as well. But I don't want to spend 80 percent of my waking hours solving naive questions.
2. I don't want the risk that comes with marketing software for money. I don't want to risk my livelihood by starting a business to support my software, either. Really marketing something takes 3-5 good years of your life to do it right, and it involves risk.
3. Yes, I like to help people and I benefit from what they contribute. I'm not anal about having to have GNU software only, but I do support what they do. I feel like I am giving back to the common pool when I develop open source apps.
4. I am not a competitive person. In fact, I absolutely dislike it. I prefer to develop a useful app to the best of my abilities. I find joy in the fact that if someone else solves the same problem I do, we can e-mail each other and combine our efforts and be friendly to each other rather than compete and try to drive one another out of business. Everyone benefits when we work together.
Disclaimer: I am not socialist. I love free and open economies. IMHO, competitive business economies are the best thing we've come up with yet. They keep people honest. But in those economies, there is plenty of room for community-welfare ideas as well.
I've always seen parallels between people who contribute to open source projects and people who build their own airplanes, boats and cars. They get pleasure from creating something useful or beautiful. The pleasure is enhanced by sharing it with others, receiving positive feedback and belonging to a group.
Some manage to convert their hobby into a business, occasionally a thriving business. Sometimes they make an important contribution to the field. Mostly, they just enjoy creating and sharing.
This is not a topic for economic analysis. This is not a topic for any kind of analysis. It is something that is rewarding to its participants and that's enough for them.
Why do you think Linus is still plugging away at the Linux kernel? Not because he's absolutely needed, but because he likes to do it.
In other words, he finds it desirable. I would include that under "need/desire."
Is the open source movement about the joy of hacking? The latest edition of FirstMonday has an interesting academic study that says "No!"
.com company [probably] but there is economic incentive in that people with winmodems wouldn't have to purchase regular modems if they could get their winmodems to work.
If this is the question the statisticians had in mind when composing the study, then they started from a logically flawed position. This study supposes an either-or dichotomy, which is not atypical of this type of study. A good deal of open-source work, according to those who work it, is done for altruistic reasons, even if there are [potential] commercial benefits. For instance, getting Winmodems to work in Linux isn't going to spawn another
Moral of story, open source development done for the reasons of economic benefit or open source development done for reasons of pleasure are not necessarily mutually-exclusive entities. There is more often than not a middle road, wherein it seems the majority of developer's intentions lie.
For the most part, if I'm interested in substance and functionality over glitz, OpenSource projects have what I want -- with the added benefit that if it isn't precisely what I want, I can fix that. If there's a problem, it usually gets noticed and fixed sooner. Authors take pride in their work, as their very name and reputation is attached. And amazingly enough, all this comes at a very low price tag.
That doesn't say that OpenSource comes at no cost. The economics are slightly different. To be a consumer in this market, I have to have about the same amount of knowledge in my head about how my computer worked when I was running DOS back in the '80s. I recognize my computer is not a do-all appliance with pre-canned solutions I have to accept or not use. I can mold it to my will... and surprisingly with relatively little effort. OpenSource lets me venture into the realm of the unexplored if I so choose, or I can stay well within my comfort zone.
On the flip side, to be a contributor, I recognize I may never get rich directly from my contributions. However, I can get noticed. I can get famous. I can get appreciated. I can be worth more to my employer, whether from experience or name recognition.
It's personally rewarding, providing personal growth, a sense of community, and is fun to boot. I've yet to get this experience out off a sealed package off the shelf.
If you're the type of person who find yourself doing a View Source when you visit an interesting website, then you've got enough of a streak of curiosity to survive consuming OpenSource.
With familiarity and tmie, it's easy to contribute. Contributions don't have to be just code. They can be suggesting ideas, reporting bugs, play testing, or even proof reading.
- "The 'utility function' Linux hackers are maximizing is not classically economic, but is the intangible [product] of their own ego satisfaction and reputation among other hackers."
and then saysIn other words, if Linus Torvalds says he does it just for fun, he must be lying because fun is hard for an economist to quantify. Likewise if Eric Raymond says he does it for ego, he must be lying, based on the same reasoning. Personally, I write open-source textbooks because I hated all the choices from the big publishers -- my motivation is my own professional satisfaction and maximizing the enjoyment of the work I do as a teacher. But don't believe me. I must be lying, because professional satisfaction and enjoyment are hard for economists to measure.
If we don't want to admit that fun is an economic motivator, then why do people go to Las Vegas to gamble? They lose money on the average, but the point is that it's fun.
The author doesn't make his point very clearly, but he seems to be saying that there is more open-source development per capita outside the US because programmers in the US can make loads of money, so they want to do that instead of relaxing with a nice free software project. OK, so there are differences in the amounts of money lost by doing free software, but what do these people gain by doing free software? The author only seems to want to talk about the loss, because the gain is cultural and personal, and hard to measure. But if he believes the gain doesn't exist, then why doesn't open-source software development cease immediately?
Find free books.
Hasn't it been obvious? Open-source developers are Marxists, working towards a common good, trying to move software development along to benefit everyone.
If Carl Marx were alive today, he would probably be astounded. His ideas have lead to failed societies, and much suffering. Yet his ideas prevail among a group of geeks working in capitalists societies, collaborating all over the world.
I think that it's only possible to be a partial Marxist. I develop open-source software because other people are developing software that I use for free. They use mine for free in return. However every other aspect of my life is capitalist, and I am cool with that.
One thing that will be interesting is to see how open-source affects the software industry over the long term. States are proposing that Microsoft spill out the source code for some of their products, and also Micrsoft has the lame shared source thing going on. These are baby steps towards a big revolution, IMHO.
One of my biggest reasons for working on open source projects is that the software world is driven by network effects. If my employer had decided to go with ActiveDirectory for everything instead of developing Ganymede (if ActiveDirectory had even existed when we needed a solution), we would have tied ourselves to the wheel of Microsoft fees and upgrades, in perpetuity, forever. I personally didn't want to see that happen, for ego reasons and for the sake of my l33t UNIX job skills. My employer didn't want to see that happen, because it might have given far too much power to Microsoft over our operations. No sense being too dependent on any one vendor when you can do something about it.
It's the exact same reason why AOL is supporting the Mozilla project.. if AOL had to depend on Microsoft's good will to provide Internet services to its customers, it might at any time have its customers taken away from it, assuming a compliant DOJ.
So, yes, there's economic rationality there, there's also cultural issues, there's also ego, and pride of work, all of it.
- jon
Ganymede, a GPL'ed metadirectory for UNIX
> On the whole, people want to work and innovate because they want $
/can/ choose to support it, in a 3rd party manner, if they want; and given a sufficient level of knowedge, and a wider enough user base, they even make a living from it, resulting in an increased sense of return on investment by the owner/user of said software, thus ensuring that they will spend on support in the future. This is as opposed to spending on the 'invention', only to discover that they don't know how to use it, and no one can help them. The chance of them repeat spending more on that branch of technology must be lower than in the successful support scenario. I suppose that the only thing that scares the innovation pundits is that this money goes back to the people supporting the status quo rather than the companies furiously trying to file their latest patent, even tho both paths contribute to the overall health of the economy. And here's where the ugly truth is .. I think countries see the innovation in their economies not meant to keep the domestic economy strong, but rather to stay ahead of /other/ economies. I think this is the true point of globalism as it relates to innovation and IP. As nations, we havn't gotten over our geocentric biases yet (I have a hard time picturing an IBM exec feeling good about bringing technologies to a foreign land, as opposed to feeling good about opening up yet another market to IBM), so how on earth is globalism meant to benifit everybody if it's only being used as a means of protecting IP and ensuring that one country must start contributing to anothers economy until the patent expires (and at the rate we're going, it seems like patents will never expire in the not-so-distant future).
.. in other words .. no economic system is an island, and I believe the blind belief of innovation as our salvation is responsible for that. There seems to be no place in our world for cultures who simply wish to live happily without innovating, and, unfrotunately, those countries are being punished for it daily due to their lack of desire to grow economically on the world stage. I believe many of the values OS echews is a desire to grow and innovate at a natural level, as people see fit to contribute; not as a means of 'staying ahead of the competition', which, of course, is primarily why the vast majority of software being sold is closed.
I'm not sure where along the way we decided that people want to innovate. Some people do, sure. Inventors have existed in every culture, every age, to varying degrees. However, I'm of the opinion that this sweeping assumption, coupled with current copyright and patent laws, causes our countries to invent useless technologies, or at least technologies which cause as many short term problems as those they purpot to solve, while ensuring that you cannot make a decent living independant of the IP-owning corperate body, contributing domain knowldge, slight improvements, or supporting it.
Obviously, it's a vast oversimplification, but my point is that there is no money, currently, in smart technophiles going into support. I feel that if enough importance was placed on it, in terms of salary, working environment, and responsibility and accountibility, I'd wager that people wouldn't feel half as disillusioned by technology than they feel today; and would likely be in a better position to see where/how they could contribute to the economy.
This is one of the things, I think, that is attractive to OS developers. Other people
> at least our system keeps them in check to some level
I'd agree with this if you're referring to keeping them in check on the soil of the country it is based in. However, look no furthur than Nike, Starbucks, The Gap, etc. They're very existance has spelled the death of numerous economies, industries, markets, children, etc in other markets where there is no one to defend them. All because those cultures and economies were based not on innovation, but the maintenance of quality of life (for which there is far less a dependance on the growth of wealth). I agree that Capitalism has been the best system for those who live in it, but don't forget for all the 'freedom flag' waving the US does, it's very obvious that they are in no hurry to bring regulated democratic capitalism to countries which are suffering due to its existance unless they represent a potential future economic or political (read: communism, dictatorship) threat, sometimes even going so far as to silently lobby for governments (check out the NED) that do little but provide social and economic laws that permit US based companies to make large profits. From slaves to sweat shops abroad, capitalism has always depended on populations, cultures, and people who are not granted the same checks and balances as those who live in it's homeland.
Whew
"Old man yells at systemd"
The author presents essentially two pieces of evidence:
:-) growth in economic opportunities for programmers.
the historical migration of free software development from US to worldwide, and
the fact that being a programmer in the US has become a good gig lately
then jumps to the conclusion that this means that US programmers weren't altruistic, merely opportunistic (worked for universities before, corps now).
But he doesn't examine other areas:
the US/Europe ratio may be declining over time, but the US hacker/US population ratio has likely been increasing -- overall free software activity in the US has certainly not been overrun by the lure of proprietary software's lucre.
the appearance of free software predated widespread online use in US -- maybe the story will be the same elsewhere. That is, is the situation in Hungary today similar to US in 1984 -- only the hackers are online, so the hacker/online ratio is very high?
But I think the main flaw in his argument is inflating ESR's gift-economy rationale (which I suppose he does so purposely to puff up the importance of his conclusion). Even ESR isn't so much saying that free software hacking is completely without regard to economic conditions, but that it's an unexpected response to these conditions (hence post-industrial).
I'd claim, and I think ESR might agree, that free software is an efficient means of production (shared resources), niche penetration (scratching itches), and market penetration (network effect) made possible by BOTH the economic (free time + university grants + young single contributors) AND cultural (want props + want to contribute + crave technical knowledge) situations of hackers.
In other words, the fact that hackers could do some coding for free without starving, and that they were wont to do so, ran into the happy accident that doing so could produce some really good shit.
This would explain the experience of the 1990's -- unbelievable growth in free software and simultaneous insane (literally
The author's argument might lead one to believe that open source would wither and die if the corporate world paid programmers well enough. The simultaneity of the dot-com boom and the Linux boom deny that.
I'll grant to the author that the European countries present an economic situation more favorable to free software. In fact, I'd amplify the fact by saying that European government support for free software has largely economic motivations -- they don't want to lose to MS/Sun/Oracle/IBM any longer.
But this fact may support a post-industrial thesis as well -- workers in northern European countries enjoy more free time and have a better safety net than US workers -- so they have less to lose from partaking in a little free coding.
This is the crucial distinction: a post-industrial explanation for free software contributions doesn't put them outside of the economic situation -- it relies on the coders having the opportunity to engage in non-economic activity.
That still leaves intact two "revolutionary" conclusions from the history of free software -- that significant production can occur outside of the wealth motive (if the survival motive has been taken care of and the infrastructure exists), and that that production can (in the case of software) be more efficient in creating use value than a wealth-driven model.
"You can't get something for nothing." - my grandfather, on the stock market and Reaganomics.
Using the number of people connected to the internet as a measure of the population that each country is able to contribute to OSS projects is just plain nonsense.
The USA has a much higher proportion of 'programing illiterate' on the internet than any other country in the world.
Of course it would then appear that the USA isn't keeping up with the rest of Europe.
The underlying statistics he uses are meaningless and as we all know you can create any conclusion you want from false data.
- AndrewN
And the reason they do so is because it so neatly fits in with the from each according to their ability, to each according to their need credo which characterises ideal, if not real (by which I mean actually existing or existed, rather than genuine) communism.
Better to be despised for too anxious apprehensions, than ruined by too confident a security. --Edmund Burke
While it's refreshing to learn that Open Source has matured enough to garner some interest in the academic Economics community, the author's admitted lack of familiarity with the subject of his thesis--that exotic and enigmatic creature, the Open Source Developer, comes through. Less pretentious jargon in the vein of "the gift-economy" (is this some economist appropriation of Claude Levi-Strauss that I missed in the sixties?) and more pragmatic observation would have been welcome.
:)
1)The importance of the English language. Completely underestimated in this study. Why else would France, a very Socialist economy with a 35-hour work week have such a low level of Open Source activity? Compared to Sweden and Germany, the English language proficiency in France is extremely low.
2)Personal fulfilment. Most dedicated Open Source developers share a joy in creation. Ego is involved, but not so much as related to recognition from others, but a desire to prove something to themselves.
3)Desire for a community of peers. Once you create something you want to be able to share it with people capable of appreciating it. If your expertise is pretty obscure or high level, you're going to have to go to an online virtual community to find people with the same interest or proficiency.
4)The Corporate IT world is a cubicle wasteland. From a personal point of view, this is hardly a validating world for engineers. Even at so-called "technology" companies, it's rare to see engineers promoted to any positions of true importance. It's all about marketing sales and MBAs. Engineers tend to see something in scientific terms, things either work or they don't. This makes most of them poor players in corporate politics. Even if they are any good, they are likely to be pretty turned off by the process. How many technology decisions are dictated by top-down partnerships, which make absolutely zero sense from a technological point of view? Far too many.
If I were to profile the Open Source developer, the person most susceptible to the phenomenon would be an individual with something to prove, someone not experiencing any (or much) fulfilment in their day job, where they feel isolated, not in contact with their true professional peers, someone who is given insufficient control over and ownership of their work.
Abstract economic theory is well and good, but it very often fails to credibly explain human behavior.
As for de-bunking myths about Open Source, a far more interesting story was Slashdot's post a week ago on Marc Fleury, leader of the JBoss project. http://www.jboss.org/vision.jsp. JBoss delivers Enterprise class software. You'll have to get past the J2EE jargon and the ego, but Marc makes some interesting points: he and his developers are mostly not students, they have worked in the corporate IT world, they do care making money, and thus have a vested interest in delivering solutions that meet the needs of the corporate IT community.
While I feel basic psychology as opposed to economics has far more to do with the source of Open Source contribution, economics has everything to do with Open Source success--but that's another thread
Although Marx was pretty pathetic if judged in modern scientific terms, he did have at least one good insight, which is that historical development determines what kind of economy you can have. You can quibble with his classification of feudal/capitalist/socialist, but it's certainly true that your average peasant of millenia past could never have conceived of going to the library and reading Consumer Reports to decide what kind of refrigerator to buy. Capitalism isn't a natural phenomenon -- it had to be invented, and someday it may become obsolete. Maybe the free information movement is a sign of something like this, although I don't think I'll really see it in historical perspective in my own lifetime.
Another important thing to realize is that even after the advent of capitalism, there have always been pockets of noncapitalism. No, hippies didn't invent the concept of a commune in the 60's -- it goes back at least to 1920's-era anarcho-syndicalism. Then you have the shakers, the amish, the amanas, etc. So even if the talk of "the new economy" is overblown, that doesn't mean the whole world and everyone in it is behaving like some kind of textbook idealization of capitalist economics.
BTW, on the off chance that the parent post was really meant seriously...sorry, nope. Marx conceived of a dictatorship of the proletariat, not a let-it-all-hang-out, find-your-own-bliss nonconformist hippie-hacker paradise. He also theorized that capitalism would be destroyed by its own internal contradictions: as the poor got poorer and the rich got richer, increasing class antagonism would result in revolution, led by the factory workers, who are the most politically advanced part of the proletariat. Even if you want to argue that his prediction came true in some cases, I hardly think kernel hackers fit the profile.
Find free books.