Decision Rights in Open Source

We continue our treatment of a number of research essays on the practicalities of the governance of open source communities, here we focus on decision rights.

For related treatment, see also our wiki entries on Particpation Architecture and the Governance Rights Typology.

The following is an excerpt from draft version of the article: Of Hackers and Hairdressers: Modularity and the Organizational Economics of Open-source Collaboration. By Richard N. Langlois and Giampaolo Garzarelli Journal Industry & Innovation, Volume 15 Issue 2 2008.

The chosen excerpt focuses on decision rights and challenges Yochai Benkler’s tripartite distinction between markets, hierarchies and social production.

Consider the following Table:

PRODUCTSInside contracting /or


Classic Market
EFFORTSClassic Firm*Voluntary Production*

Commentary by Langlois and Garzarelli:

We can distinguish four possibilities:

Along one dimension is the issue of design: is assignment to task (and maybe even the division of labor itself) generated through the a centralized process or does it arise from the self-identification of collaborators with tasks? Along the other dimension is the problem of information and agency: are we talking about products cleanly measured and priced or are we taking about exchanges of effort that involve costs of measurement and agency?

In the upper left-hand box, the division of labor is centrally designed, but the products of that labor are easily measured and priced. This is the world of inside or outside contracting.

In the lower left-hand box, the division of labor remains centrally designed, but the cost of measuring and pricing transactions makes it cheaper to purchase the effort of collaborators directly. This is the classic firm.

In the upper right-hand box, participants self- select their contributions; but measurement and pricing costs are not prohibitive, and those contributions take the form of products offered on spec. This is the classic market. Finally, in the lower right-hand box, participants self-select their contributions (also effectively “on spec”); but those contributions come directly in the form of effort rather than of effort embodied in a product. This is the model of voluntary or open-source production.

Discussion of the Typology:

This two-dimensional schema has advantages, we argue, over the tripartite distinction Benkler (2002) offers among markets, hierarchies, and what he calls peer production. Benkler argues that a perceptible trend toward the increased importance of human capital in production is leading toward peer production and away from both markets and hierarchies. It may well be that, with economic growth and an expanding extent of the market, there is a general trend rightward in Figure 2, what Langlois (2003) calls the phenomenon of the Vanishing Hand. But an increased importance of human capital and greater spontaneity of production is consistent with markets as well as with decentralized collaboration through direct effort. Even apart from the likes of books, musical scores, or screenplays, there are a plethora of “consulting” services – from legal representation to brain surgery – that are priced on markets. The interesting issue, in the language of Baldwin and Clark (2003), is: when can cooperation be effectively measured and priced (and thus turned into “transactions”) and when not?”

“The most extreme form of a voluntary arrangement would occur when the self-selection of the collaborators itself actually creates the division of labor. This is far from unimaginable: it is exactly what happens in “the market” in the largest sense – including the market for software in the large. It also arguably happens in the context of academic open science, where the pattern of knowledge emerges from the self-selected research choices of the participants. If we cast our gaze down to a less lofty level, however, there almost always seems to be some pre-existing structure of possible tasks from which the participants choose.

At the level of any particular software project, the self-selection of workers to tasks takes place within the context of an established architecture or (at the very least) an established technological trajectory.”

For more, see how Christian Siefkes proposals on effort-sharing are distinguished from this concept of effort-trading.

Christian send us a note of caution concerning the concept of effort trading:

I think it’s a misnomer for the practices which I describe more aptly as “effort sharing”. Evidently, nothing is traded within open-source/peer-production projects, hence the term “effort trading”
doesn’t make sense.

I guess that’s another instance of the misanalysis we’ve recently seen on the Oekonux list: if you have the preconceived notion that all human interactions are “trading”, that naturally you will apply this term to all human interactions, whether appropriate or not. It’s like Patrick Anderson’s
sad (und usually wrong) assumption that lovers are “trading” with each other. I would strongly advise against using the term “effort trading”, since it gives people an utterly wrong picture of what is actually going on in free software and other peer production projects.

1 Comment Decision Rights in Open Source

  1. AvatarAnittah Patrick

    I like this framework and agree that it seems to be an improvement on the Benkler framework. Thanks for the excerpt and look forward to reviewing the draft in detail.

Leave A Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.