Bit by bit – a review of “Two Bits”

I finally found the time to read Christopher Kelty’s book Two Bits – The cultural Significance of Free Software. Kelty is one of the few other anthropologists studying Free Software in general, and his work has been a huge inspiration in my thesis work on Ubuntu, so naturally, my expectations were high.

As Kelty argues, we’ve been drowning in explanations of why Free Software has come about, while starving for explanations of how it works. Thus, Kelty’s focus is on the actual practices of Free Software and the cultural significance of these practices in relation to other aspects of our lives.

Kelty’s main argument is that Free Software communities are a recursive public. He defines a recursive public as a public “whose existence (which consists solely in address through discourse) is possible only through discursive and technical reference to the means of creating this public.”

It is recursive in that it contains not only a discourse about technology, but that this discourse is made possible through and with the technology discussed. And that this technology consists of many recursively dependent layers of technical infrastructure: The entire free software stack, operating systems, Internet protocols. As Kelty concludes:

The depth of recursion is determined by the openness necessary for the project itself.

This is a brilliant observation, and I agree that the notion of a recursive public goes far to explain how the everyday practices and dogmatic concern for software freedom is so closely intertwined in this public.

The book is divided into three parts, each part using a different methodological perspective to examine the cultural significance of Free Software.

The first part is based on Kelty’s ethnographic fieldwork among geeks and their shared interest in the Internet. I found this to be the weakest part of the book. His ethnography does not cover the actual practices of Free Software hackers, but rather on the common traits among Internet geeks, which certainly supports his argument (that they’re all part of a shared recursive public), but doesn’t give a lot of depth to understanding their motives.

The second part is based on archive research of the many available sources within the various open source communities. In my opinion, this is the best part of the book with both deep and thorough analyses of the actual practices within free software communities, as well as vivid telling of the pivotal stories of “figuring out” the practices of Free Software.

The final part is based on Kelty’s own participation (anthropologist as collaborator) in two modulations of the practices of Free Software in other fields, the Duke University Connexions project, and the Creative Commons. These are stories of his own work “figuring out” how to adapt Free Software practices in other realms. These practices are still in the process of being developed, experimented with, and re-shaped – like all Free Software practices. And this part gives a good idea of what it feels like to be in the middle of such a process, though it offers few answers.

Being a completely biased reviewer, I’ll stop pretending to do a proper review now, and instead focus on how Kelty’s analysis fits with my own study on the Ubuntu Linux community. Kelty argues that there are five core practices, which define the recursive public of Free Software. Kelty traces the histories of “figuring out” these practices very well, and I’ll go through each in turn:

Fomenting Movements
This is the most fuzzy on Kelty’s list of five core practices. I understand it as placing the software developed within a greater narrative that offers a sense of purpose and direction within the community – “fomenting a movement” as it were. Kelty has this delicious notion of
“usable pasts” – the narratives that hackers build to make sense of these acts of “figuring out” after the fact.

In my research, I found it very difficult to separate these usable pasts from the actual history within the Free Software movement, and my thesis chapter on the cultural history of Ubuntu bears witness to that. So I am very happy to see that Chris Kelty has gone through the momentous task of examining these stories in detail. I find that this detective work in the archives is among the most important findings in the book.

Sharing Source Code
A basic premise of collaboration is shared and open access to the work done – the source code itself. The crux of the matter being giving access to the software that actually works. Kelty tells the story of Netscape’s failure following its going open source with a telling quote from project lead Jamie Zawinski:

We never distributed the source code to a working web browser, more importantly, to the web browser that people were actually using.

People could contribute, but they couldn’t see the immediate result of their contribution in the browswer that they used. The closer the shared source code is tied to the everyday computing practices of the developers, the better. As Ken Thompson describes in his reflections on UNIX development at AT&T:

The first thing to realize is that the outside world ran on releases of UNIX (V4, V5, V6, V7) but we did not. Our view was a continuum. V5 was simply what we had at some point in time and was probably put out of date simply by the activity required to put it in shape to export.

They were continually developing the system for their own use, trying out new programs on the system as they went along. Back then, they distributed their work through diff tapes. Now, the Internet allows for that continuum to be shared by all developers involved with the diffs being easily downloaded and installed from online repositories.

As I point out in my thesis, this is exactly the case with the development of the Ubuntu system, which can be described as a sort of stigmergy where each change to the system is also a way of communicating activity and interest to the other developers.

Conceptualizing Open Systems
Another basic premise of Free Software is having open standards for implementation, such as TCP/IP, ODF, and the world wide web standards developed by the W3C – all of which allows for reimplementation and reconfiguring as needed. This is a central aspect of building a recursive public, and one I encountered in the Ubuntu community through the discussions and inherent scepticism regarding the proprietary Launchpad infrastructure developed by Canonical, the company financing the core parts of the development of both the Ubuntu system and community.

Writing Licenses
Kelty argues that the way in which a given software license is written and framed shapes the contributions, collaboration and the structure of distribution of that software, and is thus a core practice of Free Software. Kelty illustrates this by telling the intriguing story of the initial “figuring out” of the GPL, and how Richard Stallman slowly codified his attitude towards sharing source code. This “figuring out” is not some platonic reflection of ethics. Rather, it is the codifying of everyday practice:

The hacker ethic does not descend from the heights of philosophy like the categorical imperative – hackers have no Kant, nor do they want one. Rather, as Manuel Delanda has suggested, the philosophy of Free Software is the fact of Free Software itself, its practices and its things. If there is a hacker ethic, it is Free Software itself, it is the recursive public itself, which is much more than list of norms.

Again, almost too smartly, the hackers’ work of “figuring out” their practices refers back to the core of their practices – the software itself. But the main point that the licenses shape the collaboration is very salient, still. As I witnessed in the Ubuntu community, when hackers chose a license for their own projects, it invariably reflected their own practices and preferred form of collaboration.

Coordinating Collaborations
The final core practice within Free Software is collaboration – the tying together of the open code directly with the software that people are actually using. Kelty writes:

Coordination in Free Software privileges adaptability over planning. This involves more than simply allowing any kind of modification; the structure of Free Software coordination actually gives precedence to a generalized openness to change, rather than to the following of shared plans, goals, or ideals dictated or controlled by a hierarchy of individuals.

I love this notion of “adaptability over planning”. It describes quite precisely something that I’ve been trying to describe in my work on Ubuntu. I used Levi-Strauss’ rather worn duality between the engineer and the bricoleur to describe part of this, but I find Kelty’s terms to better describe the practice of collaboration on a higher level:

Linux and Apache should be understood as the results of this kind of coordination: experiments with adaptability that have worked, to the surprise of many who have insisted that complexity requires planning and hierarchy. Goals and planning are the province of governance – the practice of goal-setting, orientation, and definition of control – but adaptability is the province of critique, and this is why Free Software is a recursive public: It stands outside power and offers a powerful criticism in the form of working alternatives.

As Kelty points out, the initial goal of these experiments wasn’t to offer up powerful criticism. Rather, the initial goal is just to learn and adapt software to their own needs:

What drove his [Torvalds’] progress was a commitment to fun and a largely in articulate notion of what interested him and others, defined at the outset almost entirely against Minix.

What Linus Torvalds and his fellow hacker sought to do was not to produce “a powerful criticism” – those almost always come after the fact in the form of usable pasts to rally around – rather, their goal was to build something that would work for their needs, and allowed them to have fun doing so.

I find that this corresponds very well to the conclusion of my thesis: that the driving goal of the Ubuntu hackers continues to be to build “a system that works for me” – a system that matches their personal practices with the computer. A system that is continually and cumulatively improved through the shared effort of the Ubuntu hackers, each adapting the default system to his or her own needs, extending and developing it as needed along the way. As Kelty writes in his conclusion:

The ability to see development of software as a spectrum implies more than just continuous work on a product; it means seeing the product itself as something fluid, built out of previous ideas and products and transforming, differentiating into new ones. Debugging, in this perspective is not separate from design. Both are part of a spectrum of changes and improvements whose goals and direction are governed by the users and the developers themselves, and the patterns of coordination they adopt. It is in the space between debugging and design that Free Software finds its niche.
(…)
Free software is an experimental system, a practice that changes with the results of new experiments. The privileging of adaptability makes it a peculiar kind of experiment, however, one not directed by goals, plans, or hierarchical control, but more like what John Dewey suggested throughout his work: the experimental praxis of science extended to the social organization of governance in the service of improving the conditions of freedom.

In this way, Free Software is a continuing praxis of “figuring out” – giving up an understanding of finality in order to continually adapt and redesign the system. It is this practice of figuring out that is the core of cultural significance of Free Software, as we continue to figure out how to apply these learnings to other aspects of life. Kelty does well to describe his own efforts “figuring out” in relation to non-software projects inspired by Free Software practices in the final part of the book. Though these reflections do not come across as entirely figured out yet.

All in all, it is a brilliant book. But given its Creative Commons license, it poses an interesting challenge to me: Remixing – or modulating, as Kelty calls it – the book with my own work (and that of others – like Biella) to create a new hybrid, less tied up in the academic prestige game.

(Maybe then I can change the title, because that continues to annoy me: Why is it called Two Bits? Apart from the obvious reference to computing in general, it doesn’t seem to have any other relevance particular to Free Software?)