Notes for Yuri Takhteyev Coding Places: Software Practices in a South American City

Key concepts: disembedding and reembedding, practice, virtual ethnography, worlds of practice.

Related theorists: Anthony Giddens, Theodore Schatzki.


A Note on Translation, Quoting, and Pseudonyms

0 The Wrong Place

Look at software development from the wrong place to learn about place in the knowledge economy; relate to work of Janz.

(1) I will try to show in this book that we have much to gain from looking at software development in this somewhat unlikely place, and more generally, from looking at high-tech work in “wrong” places. By doing so, we can learn a lot about place and its persisting importance in today's “knowledge economy.”

Practice and Place

Global worlds of practice key constitutive elements of globalization in addition to geographic context.

(2) The book looks at people who inhabit simultaneously two different contexts. One of those contexts is defined geographically. . . . The other context is an instance of what I call worlds of practice—systems of activities comprised of people, ideas, and material objects, linked simultaneously by shared meanings and joint projects. . . . I argue in this book that global worlds of practice are the key constitutive elements of globalization.

With the Internet it seems like software development can be done anywhere; nonetheless, centralized in San Francisco.

(3) Given the abundance of uploaded knowledge, engaging in software production seemingly requires little more than a computer, stable electricity, and Internet access—all of which are available in places like Rio de Janeiro even to the relatively poor.
(3) Such stories, however, must not distract us from another notable feature of the world of software: its stark and persistent centralization. . . . Together, market capitalization of IT companies headquartered around San Francisco comprises over a third of the world's total.

Examine behavior and motivations of software developers in periphery places to better understand position of Silicon Vally.

(4) To understand the truly exceptional position of centers such as Silicon Valley, perhaps it helps to spend some time contemplating the periphery. What do software developers do in such places? Why do they do it?

Reverse notion of fluidity of technical knowledge and study how it moves in space.

(4) Instead of assuming that technical knowledge is naturally fluid and trying to understand what barriers keep software development so concentrated, I take the concentration as a given and seek to understand how the practice of software development moves in space at all, investigating the work that is needed to establish this practice in new places.

Practice as concept between knowledge and daily life.

(5) The concept of “practiceprovides us with a useful analytic layer between the more abstract, propositional notions of knowledge and the messy details of daily life.

How do practices move?

(5) Focusing on activities, and especially on systems of activities, makes it easy to see why the practice of software development would cluster in a handful of places, since it helps us recognize the many different pieces that would need to be put together to re-create the practice in a new place.

Key themes: Giddens process of disembedding and reembedding, cumulative and parallel nature of reproduction process, diasporic situation of peripheral practitioners, complex relation between individual and collective efforts reproducing foreign practices, interaction between cultural and economic layers, and paying attention to actors reflexive understanding of the world.

(6) My discussion of practice in place focuses on several themes. The first is the process of disembedding and reembedding (Giddens 1991) involved in its reproduction across space: people engaged in a practice that is based somewhere else often have to reassemble the practice around imported elements, substituting for missing pieces what happens to be available. . . . The second theme is the cumulative and parallel nature of the reproduction process. I look at the local practice of Brazilian software developers as a partial reproduction of the American software practice. . . . Third is the theme of a “diasporic” situation of the peripheral practitioners, who engage simultaneously in two cultures: the local mainstream culture and the globalizing world of the practice. . . . Closely related ot this is the complex relation between individual and collective efforts of reproducing foreign practice. . . . The fourth theme is the interaction between the cultural and economic layers of the practice, and the need to look at the two simultaneously, considering the situations when one of those layers is present and the other is missing. Finally, I stress the importance of paying attention to actors' reflexive understanding of the world, the possible futures they can imagine individually and collectively, and the factors that influence this imagination (Giddens 1979; Appadurai 1996).
(7-8) While I do not see this centralization as a puzzle per se, I do believe that there are many explanations that are wrong and self-serving and that such explanations may themselves contribute to the persistence of centralization.
(7) I also intend to show how such peripheral work contributes to the continued dominance of remote centers. . . . By fixing their gaze solidly on foreign technology and investing efforts into making it work locally, peripheral developers often deny to local projects the attention that such projects may need.

Invitation to policy makers to follow author on visit without making policy recommendations.

(8) I do not make specific policy recommendations, but I invite policy makers to follow me on a visit to a world that they govern (in part) but do not always understand, to see the challenges faced by people who inhabit this world and to consider how helping them face those challenges may contribute to the larger developmental agenda.

(8) While I believe the patterns I explore in this book could have been shown using many other cities, the choice of the specific place can make a difference. . . . I believe that my choice of place gives us a good balance: a city present on the world map, yet not quite one of the “global cities”; in a developing country that seems to be gaining momentum, yet doing so at a pace that allows for some reflection; and with a history of IT policy that goes back a few decades—putting some of the most important events in this history far enough back to allow for critical analysis.

Free / Open Source Software

Emphasis on floss as reliant on global, computer-mediated interaction, yet reflexive of global influence of American hacking culture, involving complex negotiations of culture, language and geography.

(9) The book focuses disproportionately on a specific form of software practice known as “open source” or “free” software development.
(9-10) Such communities are also remarkably dispersed and rely predominantly on computer-mediated interaction, with members often having little idea where on the planet other participants happen to be. At the same time, however, the geographic concentration of those communities rivals that of the software industry, with rare projects that originate in “wrong places” often quickly moving their centers to the West Coast of the United States. The global culture of such communities is based largely on the “hacking” culture that originally developed in American universities. . . . As I will try to show, participation in open source projects involves a complex negotiation of culture, language, and geography, and is often
harder than engaging in other forms of software practice, since it requires more fluency in foreign culture and demands more of the resources that may be hard to find in places like Rio de Janeiro.
(10) On a more abstract level, open source development also simply represents a
new way of developing software, and thus highlights the challenge of keeping up with the evolving practice based far away—what we could call “synchronization work.” Looking at how Rio developers respond to this challenge may therefore help us understand how people engaged in other worlds of practice respond to changes that take place in those worlds.


Lua is a programming language developed in Rio that is gaining global popularity and reflects contradictions of little local use and primacy of English in its user community.

(10) Several chapters of the book look closely at a particular open source project that would be unusual by most measures: Lua, a programming language developed in Rio de Janeiro that has recently gained substantial global popularity around the world—in particular, among software companies based in California.
(10) Lua is the only entrant into this exclusive club [of programming languages] from a developing country.
(10-11) Lua's position in Brazil, however, presents us with an even larger puzzle. Almost no local communities make use of Lua in their products. Lua's large and active community interacts primarily in English. . . . Lua's global success has so far done little to rescue Rio de Janiero from its position as a “wrong place” for developing software.

The Project

Deep ethnographic project sprung from dissertation research.

(11) This book is based on an ethnographic project—an attempt to understand the experience of a group of people through an extended engagement with them. In my case, this meant a combination of over one hundred interviews, extended presence in places where software work was being done, and at times active engagement in the members' projects.
(12) By September 2004 I had decided to focus my dissertation research on software developers in Brazil and their access to software knowledge from the foreign centers of software practice.
(13) As my Portuguese fluency improved, I conducted more interviews in Portuguese, eventually using English only with the developers who spoke fluent English and preferred to talk to me in it.

Discovery of cultural biases against talking about technology; seen as reflecting conflicts between local and global communities.

(14) Talking about work, he explained, was simply not considered cool in Rio—young men are expected to talk about soccer and women, not computers. He assured me that my other interviewees did talk about technology with friends, and that I just had to know how to ask. As I soon came to realize, small differences in wording and intonations did indeed affect greatly the interviewees' readiness to talk about talking about technology.
(15) I came to see those contradictions as reflecting the underlying conflicts between their commitments to the local place and to the “global” (but often also quite foreign) technological practice. I also started recognizing in those tensions the different images of the world that the developers had.

Fieldwork focused on Lua and Kepler then typical Java applications for local clients.

(15) I then decided to dedicate half of the second phase of my fieldwork to Lua and Kepler, reserving the other half for a study of a more typical case—some company building custom web applications for local clients, using Java.

Difficulty of seeing all the work going into a software project.

(15-16) I soon confirmed my suspicions that mere physical observation does not go very far when studying software work: one mostly gets to see people staring at their screens, typing, and occasionally swearing. . . . Without literally looking at the developers' monitors over their shoulders, both at work and at home, and keeping track of their solitary work, private emails, and instant messenger conversations, cell phone calls, and face-to-face chats, one can hardly see all the work that goes into the creation of a software project.

Active participation of researcher began by writing wiki in Kepler to help developers and collect interview data, presenting an explicit method of studying software work as active participant observation; contrast to in situ yet detached method of Rosenberg.

(16) After we went through a number of options for wiki software, I made a fateful decision to write my own wiki in Kepler, which was after all a platform for developing web applications such as wikis.
(16) I found in such active participation an answer to many of the problems of studying software work that troubled me at first. While no method can reconstruct the project in its entirety, active
participant observation provided me with a partial solution: a situated and integrated picture that weaved together some private emails and instant messenger conversations, some late night conversations over pizza, and quite a few hours alone in front of the monitor making sense of debug traces.

Compares virtual ethnography to Nardi study of World of Warcraft; both virtual approaches involve deep and continuing participation by the researcher.

(16) (footnote 12) In this way, my work combined elements of traditional (though multisited) in situ ethnography with what could be seen as a case of “virtual” ethnography, drawing on many online interactions. In this sense, my fieldwork had nontrivial similarities, for example, to Nardi's (2010) study of World of Warcraft.
(17) Faced with this choice, I decided to get involved seriously. This led to a struggle to maintain balance between my life as an ethnographer and my life as a Kepler developer, but in the end I felt it was worth it.

Virtual projects create opportunity and perhaps obligation for ethnographer to maintain commitment to project.

(17) Virtual projects done over the Internet create an opportunity—and in the view of some members an obligation—for the ethnographer to maintain commitment to the project through continued remote participation.

Obliges me to ask Takhteyev what programs were used to process those hundreds of thousands of word sequences as part of interview process to resurrect and incorporate prior procedures.

(17) I returned to the United States in August 2007, bringing with me 150,000 words of field notes, not counting notes and recordings for over a hundred interviews.

The Chapters that Follow

1 Global Worlds of Practice

Globalization does not eliminate space; global worlds of practice cut across local places, especially technical work.

(21) The first premise is that globalization is a real phenomenon, and quite likely one of the most important dimensions of the set of transformations taking place in today's world. The second is that we cannot understand globalization just as a matter of space ceasing to matter. . . . Globalization means a growing importance of global contexts that cut across local places. The system of relationships that comprise global software development represents one such context. . . . For understanding the globalization of technical work and knowledge, a particular kind of global context is crucial: I call it global worlds of practice.

Schatzki nexus of doings and sayings.

(21) I use the term worlds of practice to refer to systems of activities comprised of people, ideas, and material objects, linked (and defined) simultaneously by shared meanings and joint actions. Each of such systems represents, to quote Schatzki (1996), a “temporally unfolding and spatially dispersed nexus of doings and sayings” (89).

The Liberalization

(110) The end of the market reserve is sometimes seen as a tragic collapse of an enlightened national policy under the pressure of neoliberal globalization (e.g., Schoonmaker 2002). It is important to remember, however, that the market reserve was itself an alliance in pursuit of globalization and its end signified, above all, a desire on the part of many members of this alliance to seek globalization by other means.

End of market reserve in Brazil forced many from hardware to software.

Could cultural transformation from hardware to software represent a dumbing down of the collective human side contributing to human and machine symbiosis form of intelligence?

(111) As Jorge saw it, developing software was an easier task than many of the ones he had faced as an electronics engineer. In a similar way, many former computer companies have transformed themselves into software factories.

Software development transformed by Internet access; effect in Brazil compares by some deformation to US periphery of local concentration of power and creative source of software, Silicon Valley and later centers in the United States.

(111) In 1992, Rio and Brazil became connected to the Internet, a new computer network that was rapidly growing in popularity around the world. Access to the Internet enabled real-time access to the World Wide Web, transforming the practice of software development.

Takhteyev, Yuri. Coding Places: Software Practices in a South American City. Cambridge, MA: The MIT Press, 2012. Print.