On the funding of open source
By webmink on Sep 08, 2008
Over the past week the question of who pays for open source and whether it faces a tragedy of the commons in the future has kept coming up. There seems an instinctive acceptance that open source is a charitable venture that should seek collective donations like any other non-profit. I don't think the tragedy of the commons applies to a true open source community, and I think the assumption it's all done "for free" is a mistake.
Three threads came together for me this evening from diverse sources:
- Chris Anderson, the author of The Long Tail, is now blogging his way to his new book about the impact of "free". I've found the posting stimulating and insightful and I'm finding the blog as valuable as his earlier "long tail" series. His posting last week about the three basic economic models involving getting something without paying was very clear, laying out all the approaches I could think of for leveraging "gratis". But at the end he made me wonder if he's completely clear about the "Free" in open source meaning "liberty" and not "price". The trigger was when he suggested the third model - "freemium" - was the model behind
open source's "support included" commercial versions of Linux.While the models he describes can certainly apply to products delivered to consumers, I am not so sure they apply to the commons at the heart of peer production of which Yochai Benkler wrote.
- As it happens, I am currently reading Clay Shirky's (excellent) book Here Comes Everybody, which explores commons-based peer production in terms far more acessible than Benkler. I have reached the chapter called "Personal Motivation Meets Collaborative Production", which largely uses Wikipedia as an example1. Clay's book so far has been an ode to the power-law curve, and this chapter pointed out how the attribute of collaborative communities that, characteristically for power-law contexts, 80% of the work is done by 20% of the people (with one or two doing 80% of the work done by that 20%)2. Looking at Anderson's third diagram it's clear he's talking about the same sort of phenomenon.
- The third thread came from several conversations in different contexts about cash donations to open source communities. I have been being a a nuisance on some open source boards I participate in over accepting large donations from non-participants in return for giving them recognition. I've been asserting that it's not right to allow outsiders to gain the benefit of being seen as part of a community just because they made a large financial contribution. I still stand by that assertion, despite getting considerable push-back from several people whose opinions I respect.
Paying Your Own Way
These three threads all come together in the observation that open source is what happens when several different people choose to work together on the same code base rather than working separately. Each of them is there for their own reasons; each covers their own costs and contributes the code they choose to. There is no pooling of funds to pay for work to be done because everyone is solely responsible for their own costs.
As a consequence, there is no fiscal power that any contributor holds over others, so no-one has the right to tell the others what to do. An open source community is an example of a group of people choosing to synchronise their mutual interests, each at their own expense, for the benefit of all involved including themselves3. While there may be a non-profit organisation for administrative reasons, an open source community is not a non-profit or a for-profit.
Now, if the motivation of one or two of the participants is to then offer the software as part of a "freemium" plan, that doesn't mean the whole project is there to serve their activity. That's just the motivation of one or two participants at work. They are not giving away their work without payment; they are giving away the contents of the commons at the same price at which they acquired it.
As long as their activity doesn't "take over" and disrupt the interests of others, no-one minds too much. Your motivations for participating are rarely my business. There's a "long tail" out there too, made of a large number of others who have their own motivations to be there and who are covering their own costs as a part of executing on those motivations. And the commons isn't spoiled in any way by being more widely used.
If that's the case, allowing a donor to give only money to be able to use the community's "brand" flies in the face of the basic dynamic of the community. Whuffie is not for sale. I understand that some communities have created an adjunct non-profit, and that body can use the money for socially useful things like hosting or employing a facilitator. But the money that's needed ought to be coming from the community members themselves, not from an outsider trying to wrap themselves in the community's "flag". The community is only about contributing as far as that is about collaborating.
Mesh of Motives
Any open source project that actually has a co-developer community is thus not an example of Anderson's third model because profit doesn't come into it. Some of the contributors might be, but the community as a whole is actually a mesh of different participants, all with their own motivational models and all paying their own way to achieve them outside the context of the community. If those motivational models involve business, I am sure they will be about payment at the point of value. But the community itself is about the liberty to align interests, not about the presence or absence of profit. Once again "free" deceives english speakers...
- All the way through the chapter I've been wondering the effect of the strong Wikipedia culture. Shirky somewhat idealises contribution and the fact that there's an inner circle of "Wikipedians" taking actions like deleting non-notable entries rather than developing them isn't reflected.
- When I read about this at the end of last year it was a small revelation: whenever you see an 80/20 rule at work, there's an underlying power-law curve involved. I wonder how I had managed to miss this for so long.
- Before I get accused again of saying open source doesn't involve altruism, let me be clear that I don't think that. I do, however, reject the definition of altruism as being "an act that benefits other but not me."