Law in the Internet Society


In this paper I examine Eben's argument that anarchism produces inherently superior functional goods when the marginal cost of production of each new unit equals zero. Specifically, I employ the arguments of David Stark and Gina Neff in their article "Permanently Beta" and Eli Noam's arguments in "The Economics of User Generated Content and Peer-to-Peer: The Commons as the Enabler of Commerce," and suggest that there are conditions at the micro level which provide additional insight into the set of conditions under which Eben's argument works and that anarchism might not be the right way to describe the mode of production.

Efficacy and efficiency of anarchic production

Eben's argument rests on the efficacy and efficiency of the free software movement in a net enabled society. The evidence for such productivity has occurred following the adoption of the General Product License (GPL) and other variations of open source licenses across the movement. (Evidence is ever more plentiful - Samba, Mediawiki, Apache, Firefox - the list goes on.) Few are now willing to defend the "closed" proprietary model as advantageous (see Shawn Shell for an article) Microsoft does so yet even they have opened a open source lab which seemingly seeks to benefit from external contributions of resources though doesn't license them in a "free" manner. Moreover, it is generally accepted that the success of free and open source software proves that it is of a comparable quality and reliability as that of proprietary software. These functional advantages are underpinned by the fact that if the code doesn't quite work as needed the technologists have the ability to fix it themselves, and no less importantly, technology executives avoid the game playing inherent in selection and subsequent purchasing process for software licenses. Furthermore, free software users can be confident that they won't be left managing proprietary tools for which support is either suddenly no longer available or 30% more expensive as a result of an arbitrary commercial decision. Without question, the facts on the ground suggest that free software production works and works well.

Open source production under a "free software" model

Considering Eben's claims for anarchic production more closely, though, there is a need to recognize the criticality of the GPL to open source for several reasons. First, even Eben suggests the GPL is the greatest achievement of Richard Stallman (Moglen, Anarchism Triumphant). This claim is notable since the development of the GNU operating system and its subsequent marriage with the Linux project was no inconsequential success in itself. Second the GPL and its derivatives, are what differentiates free software from merely open source software and though a number of projects don't use the GPL license it is entirely reasonable to conclude that without this legal artifact the groundswell of participation in open source may not have occurred as it has.

That said, it is curious that a relatively modest legal document enforced through the norms and mechanisms of the state is thought critical to facilitating what Eben describes as anarchic production since you might define such production as "lacking order, regularity, or definiteness" if you draw upon the Merriam-Webster definition of the term anarchic. Moreover, since anarchy, as defined by The Concise Oxford Dictionary of Politics, is "the view that society can and should be organized without a coercive state," it suggests that despite its unconventionality calling this this mode of production "anarchic production" is not quite appropriate. The mode of production that works as a result of the GPL operates as a result of the existence of state-enforced mechanisms. At the micro level production occurs through loose organizational structures, which David Stark might call heterarchical, which contain some level of order and modularity. Additionally, it coexists with profit seeking firms seemingly fruitfully for both parties, which suggests to me that despite its success is far from independently anarchic.

Micro structures and second order effects

Undoubtedly the internet society, a state of societal operation that is underpinned by an electronic network, is a necessary condition for the viability of peer produced, zero marginal cost digital goods, yet the success of such a mode of production requires a richer definition. (For me the seeming anarchy in the mode of production inherent in the development of free code is perhaps more a result of a lack of understanding of the properties of such assemblages of people than anything else.) However, explaining that this mode of production can occur doesn't explain why it does occur.

Stark and Neff in their article "Permanently Beta" identify the properties of the digital as privileging a mode of production that is forever unfinished, yet also one that is populated by "a hodgepodge of formal and informal organization[s]" alongside "practicing communities." Noam in his article specifically identifies a narrow context, at the beginning of an innovation cycle where he argues community production of content has a competitive advantage over other forms using strictly micro-economic arguments. What Noam doesn't accept is that the innovation life-cycle may, if one accepts Stark and Neff's arguments, never include a phase that privileges either the competitive marker, or even the oligopolistic or monopolistic firm. There may never come a point where enclosing what has been produced inside a private entity occurs. (Perhaps some products in the future will always be of the people, by the people, for the people?)

Explaining why open source production does occur and is currently so successful requires a recognition that artifacts such the GPL have been central to its success, that it arisen in a context in which it has aided (and been aided by) profit-seeking firms aiming to use its products in order to make a profit on other non-zero marginal cost goods. Identifying all these artifacts, and the particular properties of the network, including the actors within the networks which succeed, is a task only just begun.


It is for these reasons that I argue the word anarchy is inappropriate in describing social production of digital goods and that there is more to the success of free software than an argument that it occurs merely because the net permits collaboration. "Free" software is succeeding not only because the internet society exists, or as Yochai Benkler might write, peer production, is possible, but also because of the existence of a set of licenses, an economic environment which aids it, and the uncertainty reduction benefits users acquire independent of the mode of production (but not the licensing regime) itself. These artifacts and second order uncertainty reducing effects are as important to its success as its efficacy and efficiency. More analysis is required.

-- TomGlaisyer - 25 Oct 2008

If anyone has any thoughts on this please feel free to add a comment

A possible negative result of open-source software development:

-- MarcelEggler - 29 Oct 2008

While I won't dispute the importance of the GPL in the current free software environment, I would characterize free/open source software licenses as a hack that engineers the existing copyright regime into something more akin to what you would have under anarchistic conditions.

The GPL would not be as necessary without copyright and patent laws. In an environment where legal protection didn't exist in the first place, commercial software developers would've sought to extract value from something other than the code from the very beginning.

Widespread acceptance that software code could be copyrighted or patented didn't happen until the 70s/80s. Prior to that point, software development followed patterns that are quite similar to the development patterns now seen in the FOSS community. And commercial software developers like IBM subsidized their software development costs through other means -- selling the hardware and the support.

The GPL (and its enforcability through the mechanisms of the state) does create different baseline conditions from pure anarchism; it forces parties to distribute source code if they distribute object code. But if copyleft was a necessary feature of an environment under which anarchistic modes of production create better functional goods, then how do you explain the considerable success of BSD-licensed FOSS projects?

-- AndreiVoinigescu - 22 Dec 2008

Andrei, Thanks for your comment. I would agree that the GPL is a hack on the prevailing use of copyright law. I will suggest though that without it coders could attempt to extract property rents via the coding equivalent of creating a trade secret by refusing to distribute source code. Hence my argument that the GPL (and its place within a stable legal system) is critical. Regarding your point on BSD I would rather argue its relative lack of success in comparison to the GPL as demonstrating the value of the GPL rather anything else. I wouldn't go as far as to say there wouldn't be production without the GPL just that pure anarchism would not have produced quite as much code.

-- TomGlaisyer - 23 Dec 2008



Webs Webs

r24 - 21 Feb 2011 - 18:49:08 - TomGlaisyer
This site is powered by the TWiki collaboration platform.
All material on this collaboration platform is the property of the contributing authors.
All material marked as authored by Eben Moglen is available under the license terms CC-BY-SA version 4.
Syndicate this site RSSATOM