These were the kinds of projects Smith wanted members to realize—projects that have an emotional component and seemed to need social support. If it’s something members could do on their own, “don’t bother coming,” Smith declared.

This rubric represents the legacy of two distinct gendered meanings built into the single word ‘craft’ by artisan communities (Adamson, 2010; Lippard, 2010 [1978]) and progressive era domestic DIY activity in the United States (Gelber, 1997). The first con-cerns feminized connotations positioning the quotidian as the place where power rela-tions can be voiced and contested. Feminist writing of the 1960s and 1970s exposed the historically gendered nature of craft and its ties to domesticity. According to art historian Glenn Adamson (2010), this scholarship reframed amateurism not as an acceptance of circumstance they needed to transcend but as a mechanism by which to judge the degree of gender prejudice. Feminist art historian Lucy Lippard (2010 [1978]) has argued that the category of craft even made possible the recognition of more female artists, expand-ing the realm of fine art to include quilts, textiles, and forms of material rehabilitation. Rehabilitation, Lippard claims, as a type of “inventive” patching (e.g. remaking clothes and recovering old furniture) becomes a mending of objects and public dignity.

Participants in these groups celebrate technical exper-tise over skills that resonate with mainstream practices and ideals (like advocacy). Drawing out similar relations, Ellen Ullman (2012) uses her personal account of writing assembly language to argue that work closer to the machine helps (often male) programmers main-tain a higher status in computing cultures. In her examination of the largely male free/libre open-source software (FLOSS) community, Dawn Nafus (2012) extends this argument to hacking discourse in what she terms a “pushyocracy.” FLOSS members’ open scrutiny and “highly masculinized, aggressive online talking” shaped the perceived worth of individual contributions to expose “both the material aspects of computing and the social identities that people create for themselves through engaging with programming [...as] cultures made by and for men” (p. 671).

During our visits, members paid US$60 for monthly dues, and the composition of the group tended toward White members (although active members included nonwhites; Smith, a first-generation Chinese-American, for example).3 The HackerMoms space sat at the border of Oakland and Berkeley where storage units and liquor stores once stood. Bakeries and gift stores now lined nearby streets, catering to a growing number of upper-middle-class residents, subtlety reproducing class posi-tions and exclusions.

Creolization, however, does not seek or avoid conflict but aims to transform tech-nology or come up with new practices so that the technical systems better serves user needs. Creolization draws on bricolage practices (Lévi-Strauss, 1966), identifying the components of the technology that can be re-used, modified, or recombined to create something new. Creolization finds users more deeply involved in transforming technol-ogy than baroquization.

Upon closer examination of the appropriation phase, we distinguish three modes of tech-nology appropriation: baroquization, creolization, and cannibalism, by analogy with cul-tural appropriation. Baroquization is the filling-in of technological spaces that providers intentionally leave blank for users to personalize devices and applications; creolizationis bricolage, the recombination of the technology’s components to create something new; and cannibalism is creative destruction, an innovative act that requires breaking down the existing to invent something new.These three forms of appropriation correspond to increasingly confrontational stances users take vis-à-vis technology providers, within a context of asymmetric power. Baroquization is docile, where users follow an appropriation script laid out for them by the provider. Creolization is playful and unpredictable, where users re-mix the providers’ script with their own, in ways that may clash with providers’ interests. At the extreme, cannibalism is deliberately hostile to the providers’ interests. These various appropria-tion modes correspond to progressively deeper user involvement with technology, requir-ing increasingly sophisticated technical skills. All three modes represent creative ways for users to assert greater control over technology, mold it to fit their lives, and make it their own.

First, the range of possibilities explored depends on the latitude for experimentation afforded by the technology. Second, the intensity with which innovations are explored reflects the

Users do new things in new ways. Users explore new possibilities, including some beyond what motivated initial adoption. For instance, users personalize devices and applications to integrate them within their practices. Some users will re-arrange devices in ways that reflect their personalities

Our three-stage cycle of technology evolution begins when users adopt a technology and employ it to support their social or business activities. In a second stage, users appropri-ate technology: they experiment, try possibilities, and shape technology to fit their needs. In the third stage, providers take back control and repossess technology to re-configure it, reacting to the changes introduced through user appropriation. A new technology results, entering a new cycle as users adopt, appropriate anew, and prompt further repos-session, and so on (Bar, 1990).

First, we view technology evolution as a three-stage cyclical process of adoption, appropriation, and repossession. Users drive adoption. Users and providers alternatively drive appropriation and repossession, as users lead appropriation, while providers react when reclaiming the resulting innova-tions. Second, we identify three appropriation modes—baroquize, creolize, and canni-balize—that represent increasing degrees of power contestation by users. And third, we identify three repossession modes—co-opt, combine, and block—that represent increas-ingly antagonistic reactions by providers and mirror users’ appropriation strategies.

Born in the Plantation, the Hacienda, the Latifundio, and the Mine, creolization is now “scattered in those sheet plates and concrete mazes where our common becoming is adventuring itself, in favellas and mega-cities” (Glissant, 1995: 87). Alive and well, cre-olization can be found where Latin Americans live, in the spaces where they are exposed to new technologies. Born of avoidance (like Internet Protocol (IP) packets that find a route around obstacles) and mixing (like mashups and re-mix), creolization fits the realm of ICTs.

The baroque succeeded because it expressed something all Latin American people (Indians, Africans, mestizos, and even sons and daughters of Spaniards born on the con-tinent) had in common: the rejection of the domineering and distant center. Carpentier (1995) explains that to understand “Why is Latin America the chosen territory of the baroque?” we must look at the people and processes that allowed them to finally own the continent: “Because all symbiosis, all mestisaje engenders the baroque. The American baroque develops [...] with the self-awareness of the American man [...] the awareness of being Other, of being new, of being symbiotic, of being a criollo” (p

Gil, who later became Brazil’s Minister of Culture, described tropicalismo as “No longer a mere submission to the forces of economic imperialism, but a cannibalistic response of swallowing what they gave us, processing it, and making it something new and different” (Dibbell, 1989: 78). Within this historical perspective, Gil’s ministry support for Open Source and Free Software takes on its full meaning.

Throughout history, Latin American populations have had extensive experience with the appropriation of objects, people, and ideas from abroad, most often in unfavorably asymmetric situations. This tradition continues to pro-duce a culture of its own, born from multiple resistance and appropriation strategies.

Surman and Reilly (2003) focus on appropriation of networked technologies in a strategically, politically, and creatively innovative manner oriented toward social change. In this context of advocacy, effective technology appropriation includes strategic Internet use for collaboration, publishing, mobilization, and observation. Here, the delineation between the use and appropriation occurs when technology is adapted to reflect goals and culture. Camacho (2001) describes appropriation by civil society organizations at the pinnacle of a technology use ladder. In the middle of the ladder, organizations focus on adoption of conventional technology. Toward the bottom, organizations and individuals with constrained access or slow adoption rates lag behind and seek access to technology. At the pinnacle, however, pioneers and activists appropriate technology to promote causes, for instance, creating flash mobs through mass text messaging to instantaneously organize large groups of people for social protest

Appropriation considers both learn-ing-by-using and learning-by-doing as central to the development of new processes (Rosenberg, 1982). Learning-by-doing in particular assumes that knowledge emerges through bricolage—tinkering with and recombining technology elements, thus enhancing one’s understanding (Lévi-Strauss, 1966). For Tuomi (2002), learning through appropria-tion is a user-centered process whereby users meld culture with material resources to innovate.

Users may modify a device, download, or program new applications, invent unintended uses, or develop new practices that leverage its possibilities. This experimentation constitutes a re-negotiation of the power relationships embedded in the technology. This creative re-negotiation is the core of appropriation, the process through which users take something external and make it their own.

Second, we posit that users adopt technology because it makes a difference to their lives. In the case of mobile telephony, the technology opens up new socio-economic opportunities by increasing connectedness to others. Usage, in turn, reveals the politics embedded within a technology’s configuration, disclosing who controls what

For a technology to evolve in accordance with users’ needs, aiding social and economic development, the focus must move beyond mere adoption. When users appropriate technology and make it their own, new uses and innovations emerge. The appropriation process is a contest for control over a technological system’s configuration, as users, designers, and manufacturers battle over who can use that technology, at what cost, under what conditions, for what purpose, and with what consequences. This confrontation, we argue, constitutes a powerful innovation mechanism.

Inconsistencies in ideological perspective are not uncommon in activities which attempt to balance individualism and communalism. These very frictions might belie possibilities for a greater imagination or shared experience. However, we argue that it is only in the disputes and frictions between pluralities of publics that democratization emerges. Dissensus across making and hacking communities allows people to experiment, eventu-ally finding communities and processes in which they feel comfortable and can identify. These very migrations, connected by fluid narratives and practices, drive the capacity of communities to develop and innovate.

Civic hackers might be most appropriately described as utopian realists (Giddens, 1990: 154), a term Giddens employed to capture how assuaging negative consequences in a risk society required retaining Marx’s concern of connecting social change to insti-tutional possibilities while leaving behind his formulation of history as determining and reliance on the proletariat as change agents. He positioned utopian realists as sensitive to social change, capable of creating positive models of society, and connecting with life politics.

civic hackers seek to ease societal suffering by bringing the hidden workings of abstract systems to light and improve their functioning. Part of the academic discomfort with recognizing civic hacking might stem from their activities cutting across political categories that have traditionally been passionately defended: unitary and adversary, citizen and consumer, horizontalist and institutionalized, and prefigurative and strategic.

The most popular apps to date have been highly instrumental ways to request services to fix city infrastructure, such as SeeClickFix, a platform that lets residents take pictures of issues that need repair, that are delivered to the appropriate city department as an actionable item. We might think of this as a base-level civic act similar to picking litter off the ground or paint-ing over graffiti. Other activities are thicker modes of participation by generating data or metadata. The primary effort of the 2014 CodeAcross effort was to map exist-ing sources of open data. The leader of the event, D.W. Ferrell, described “our role as citizens is to complement” efforts by the government and organizations such as CfA. Contributing to data repositories served purposes for multiple stakeholders:

dis-putes over community as a particular category threatens to distract from a general focus on solidarity by activating “social bases of discursive publics that engage peo-ple across lines of basic difference in collective identities” (p. 374). A mutable, popu-larized hacker identity may have this potential, capable of processing and interpreting abstract systems of regulation.

Data activism and advocacy ranges from civic engagement (Putnam, 2001) to more oppositional activism (Jordan, 2001). In this sense, it is a spe-cific association of technologically mediated participation with particular political goals (Lievrouw, 2011) resulting in a wide range of tactics. Although open government data is still evolving and is constrained by predictions for economic growth and self-regulation, I argue it enables civic hackers to participate in civic data politics. This is particularly important because data-driven environment is often distanced from pro-viding individuals a sense of agency to change their conditions (Couldry and Powell, 2014). Data activism and advocacy can take place through organizing on related top-ics, online through mediated data repositories such as Github, and in-person events such as hackathons.

Counter to a high-minded association of “liberal projects,” projects in hackerspaces resisted a final form. The action in hackerspace is more bricolage(Lévi Strauss, 1962, p. 18)than engineering. Projects in Geekspace arose spontaneously. Their inertia drew people to the space, bringing their tools and ideas that would then supplement the ideas of others. This churn is captured by how projects were seen as a success of the new space even though very few projects were actually finished.

These more formalized gatherings were an attempt to get people working and collaborating in a space that had mainly turned into a spot for hanging out, drinking and foosball. The shift to the new space was seen as an opportunity to encourage members to use the space in a more productive way. The space needed members as much as members needed the tools. Members echoed a liberalist concern with increasing freedom of individuals to act, while retaining hobbyist cultures’ engagement with materialities. Often hackerspace members also described the need for a hackerspace as part of a shift in their city’s economy

A key part of the liberal contribution policies we’ve been building is an inversion of the typical code review process. Rather than the default mode for a change to be rejected until enough people sign off, we make the default for every change to land. This puts the onus on reviewers to note exactly what adjustments need to be made in order for it to land.

This is what a healthy project should look like. As the demands on the project from increased users rise, so do the contributors, and as contributors increase more are converted into committers. As the committer base grows, more of them rise to the level of expertise where they should be involved in higher level decision making.If these groups don’t grow in proportion to each other they can’t carry the load imposed on them by outward growth. A project’s ability to convert people from each of these groups is the only way it can stay healthy if its user base is growing.

Now developers had all the tools they needed. In the 1980s, they had to use a scattered combination of IRC, mailing lists, forums, and version control systems.By 2010, they had Git for version control, GitHub to collaborate, and Stack Overflow to ask and answer questions.