GlobTel s strategy of making WS their India ˜hub provided the broad framework on which the relationship was built. By establishing an independent telecommunications division within their organizational structure, WS sought to create a common basis to standardize practices that could be compatible with GlobTel and to ease the processes by which the relationship could proceed. The early establishment of a GlobTel lab in WS helped to provide exclusivity and a sense of identity to the staff. The early standardization of the physical and technical infrastructure helped GlobTel build a sense of confidence in WS, which contributed to a rapid development in the volume and level of work. Of the Indian partners , WS was one of the first to be given projects leading to ownership. Ownership came with its own tensions, however, in particular related to workforce stability and new technologies.
WS seemed to be willing to serve as the Indian ˜hub , and be guided by the GlobTel way of doing things. WS resistance to GlobTel s attempts to micro-manage seemed relatively milder and less explicit than that of the other partners. On the contrary, it seemed to express a sense of openness and eagerness to learn and adapt GlobTel ˜best practices and methodologies. This adoption took place at multiple levels from organization structure to project management practices, to employee reward and recognition schemes down to even the employee badges and the common language used around the building. These processes reached a kind of ˜peak when WS seconded their HR manager for two years to GlobTel where she would report on WS (and India activities) to GlobTel. This represented an extremely powerful way for GlobTel to develop stronger inscriptions of their standards in WS and India in general. The dominant GlobTel system of information, activities and structures, however, made some of the WS managers worry about the erosion of the WS identity among the workforce, especially the younger staff.
A key tension at WS concerned the validity of the pricing model. It had been considered appropriate in the initial stages when work was routine and WS was learning the ropes of GSA and telecommunications. But it became questionable when work involved ownership transfer. WS started to question payment based on time and material, and to suggest it should be based on the value and knowledge they brought into the process. This was not acceptable to GlobTel for various reasons, including IP concerns, expected increase in costs and fear of a loss of control over the process.
Two external but interconnected events significantly influenced the course of the relationship. The first was the Internet, that changed the technological focus in the industry and put a question mark on the future of DSP technology and its supporting budgets . The second was the political restructuring that took place in GlobTel s corporate headquarters. GlobTel had various labs in North America (Montreal, Ottawa, North Carolina, etc.) that were each aligned with different labs in the UK, France, Israel and India. The various North American labs typically dealt with different technologies and products and so realignment in technological focus affected the various counterpart labs and, through the network, the offshore activities. Through a series of events, a corporate decision was taken to transfer a significant amount of the DSP technology from a lab in Montreal to the UK lab aligned with them. And since WS was aligned with the UK lab, a corporate decision was taken to ˜transfer this transferred technology to WS. WSgained as a result, and their business showed a growth as compared to the other Indian partners who showed a decline in 1998. This growth, however, created another tension “ that of WS accepting work in a low-growth area (of DSP) when the rest of the industry was showing a rapid growth (50 “60 per cent per year). This caused a further tension, that of trying to retain the best talent without being able to offer them work on what they perceived as the latest technology in the industry.
It is interesting to examine the nature of the standardization efforts and how they progressed over time. In general, an increase in standardization initiatives was accompanied by increasing levels of work (from initial bug-fixing to feature development to ownership), both in quality (from peripheral to core technologies) and in volume (large-scale ownership transfer). The broad trend seemed to be that with a steady increase in the content and quality of work, there was an increasing need to standardize first the technical domain and progressively the management processes. Although the initial focus was on technical and infrastructure standards, with time the focus shifted to processes and practices, especially to systems of HR management. A more advanced and extreme form of standardization came with the attempts to create the ˜global GlobTel manager framework.
Expatriate managers drawn primarily from GlobTel s International R&D Group (GRDG) were key actors in guiding the various standardization initiatives, including infrastructure, office space and management and technical know-how. With the processes of ownership transfer gaining ground, WS started having more direct linkages with the UK lab it was aligned with, and viewed GRDG as redundant. GRDG had succeeded in establishing a robust infrastructure that so strongly inscribed the GlobTel way of working that its own role was seen to be redundant. The GRDG set-up was slowly phased out in India; people either resigned or returned to the corporate fold of Glob- Tel. Somewhat paradoxically with increased standardization GRDG, the agency that had been responsible for setting up this infrastructure, was withdrawn. Metaphorically, GRDG served the role of setting up the scaffolding of the building, and once that was done, the scaffolding became invisible.
The metaphor of scaffolding provides an insight into Latour s (1999) question about ˜what is gained , what is lost, and what remains invariant in the process of translation? Latour raises this question in his discussion of the notion of a ˜ circulating reference , and the manner in which the idea of standardization is tied up with the concept of ˜invariant :
A reference is not simply the act of pointing or a way of keeping. Rather it is our way of keeping something constant through a series of transformations. What a beautiful move, apparently sacrificing resemblance at each stage only to settle again on the same meaning, which remains intact through sets of transformations. The rupture at each stage of the ˜thing part and its ˜sign part. The details are often lost, and what remains is the horizon, the tendency. Reduction, compression, marking, continuity, reversibility , standardization, compatibility with text and numbers “ all these count infinitely more than adequatio [does this mean resemblance?] alone. No step “ except one “ resembles the one that precedes it, yet in the end when I read the field report, I am indeed holding in my hands the forest of Boa Vista. (1999: 56)
A key point that Latour seems to make is that standardization involves a process of small translations where some form of a ˜global standard is introduced at the local level and activities are subjected to a redefinition with reference to this standard. This introduction and comparison takes place through a series of translations involving a process of dialectical interaction between the local and the global, where something new is gained, something is lost and something remains the same. The invariant part of this process of translation reflects the strength of the standard. In Latour s Boa Vista case, the Munsell number serves as a reference that is quickly understandable and reproduced by all the colourists in the world on the condition that they use the same compilation. It permits a crossing of the threshold between the local and the global. In our case, what remains invariant can be conceptualized in terms of the scaffolding used to set up the structure of the relationship including support of a number of different technical, management and physical routines.
At one level nothing has remained the same, as happens with everything with the passage of time. Initially, WS did not have a strong expertise in telecommunications or in GSA, and the relationship with GlobTel helped them to develop it. Through a series of contested and uncontested translations, WS incorporated a number of GlobTel s management processes to create a ˜WS “GlobTel hybrid reflecting a culture of GSW and telecommunications expertise, albeit with a dominant GlobTel reference. What is gained is this new technological and business expertise, a steady business of DSP legacy projects and management values shaped strongly by GlobTel. What may have been lost in this process of translation are some high-quality staff who did not want to be limited by GlobTel s legacy work, an erosion or redefinition of the WS identity and perhaps some of their work practices being superseded by GlobTel s processes.
While gaining and losing are hazards that organizations have to engage with in the present context of globalization, a complex and ongoing question is what remains invariant in this process of translation. When the scaffolding is removed, the building is left behind. Even though people who live in the building may change over time, and it might be used for purposes different from those previously intended, the structure has a defining influence in the nature of these changes and redefinitions. We can see GlobTel s contribution to the creation of such an infrastructure within WS, including the physical building, the technological infrastructure, the expertise, the management structures and processes and a certain culture of doing telecommunications work within a largely North American approach. Although the specific people in the building move away, the structure has an influence on where they go, what kind of new people come in and the kind of work that goes on.
There is thus both an ongoing discontinuity and continuity, with a stronger tendency towards continuity. This tendency is something that people knowledgeable about the software industry in India would associate with WS “ a reputation for being relatively passive, being conservative with an extremely good business sense, and keeping a low profile despite extraordinary financial successes. The company has been a key influence in shaping the larger trajectory of the industry. With the decision to accept the legacy route, WS took the path to enter areas that made excellent business sense rather than jumping into risky ventures that would involve cutting-edge technologies. From that perspective, accepting the legacy route can be seen to reflect and reinforce some of the existing ˜low-profile tendencies, some of which, it could be argued, have remained relatively ˜invariant .
The case points out the extreme complexity inherent in the infrastructure that supports GSW. This complexity emphasizes Hanseth s (2001) point that new under- standings and strategies are required for complex arrangements such as information infrastructures and not seen as information systems that tend to be stand-alone rather than part of complex networks. Standards and processes of standardization, as we use the terms, are enmeshed in these complex physical, technical and managerial infrastructures . They reflect Fujimura s (1992) notion of ˜standardized packages that relate to technologies adopted by multiple social worlds that construct new and temporally stable definitions of these technologies. This notion of standardized packages is broader than Star and Griesmer s (1989) notion of ˜boundary objects that enable coordination of work by members of different social worlds having different perspectives and agendas . Standardized packages reflect a broader conceptual and technical workspace that combines several boundary objects, and serves as an interface between various social worlds to facilitate the flow of resources . This space that defines a GSA concerns the physical spaces of the two partners, the electronic spaces created by technologies like videoconferencing and the Internet, and the various standards and processes that link the work practices within these physical and electronic spaces and with other nodes of the network.
The case points to the futility of attempting to build ˜universal standards , since they are constantly redefined, negotiated, reinterpreted and applied differently. Hanseth and Braa (2001) have argued that attempts to create universal standards often lead to the opposite effects of creating complex and ˜non-standardized systems:
Each time one has defined a standard which is believed to be complete and coherent , during implementation one discovers that there are elements lacking or incompletely specified while others have to be changed to make the standard work, which makes various implementations different and incompatible “ just like arbitrary non-standard solutions. This fact is due to essential aspects of standardization. The universal aspect disappears during implementation, just as the rainbow moves away from us as we try to catch it. (2001: 264)
Knowledge or technology is shaped and also shaping the meanings and actions of the heterogeneous network of actors surrounding the design and use of standards. The question is not how the technical content of particular standards is best applied in universal settings, but how different local particularities interplay with these standards to redefine their meanings at both the universal and local levels, how global standards are embedded or not in local practices and how actors respond to these improvisations. As Timmermans and Berg (1997) argue, although universals exist they only do so as local universals , embedded in local infrastructures and practices, paradoxically as a multiplicity of universalities. It is not a matter of dismissing universal standards or celebrating local particularities, but of developing a pragmatic balance (Rolland and Monteiro 2002) that blends the universal and local in particular contexts.
This case analysis provides ideas on how such a balance can be aimed for by under- standing the mutual linkages between standards and their role in shaping the process of the GSA relationship. These interlinkages are complex since the growth of technical standards helps to upgrade the level of work and to take the relationship to a stage of growth requiring a more sophisticated standard such as that of the ˜global manager . External events, like industry developments, contribute to further destabilization of existing standards and create the need for new ones. These dynamics emphasize the quest for universal standards to be akin to ˜catching the rainbow (Hanseth and Braa 2001). Understanding them can help develop standards that can sensitively support the relationship, and yet be flexible enough to respond to local needs. A first step to this is the need to reconceptualize standards:
A primary aim of standards is to support the universalization of activities , and with it to facilitate some form of ˜mass production
Standards should be seen as a means of simplification and abstraction that defines and enables communication across actors separated by time, space and cultures
Standards are formal “informal, explicit “tacit, external “internal, and emerge incrementally through a series of politically negotiated translations
Standards apply not only to the technical, but also extend to the management and physical domains; standards across these domains are interconnected, and there may exist a hierarchy of standards with physical and technical standards serving as a precondition for the setting up of management standards
There is an ongoing interplay between the need to apply more ˜global standards and the need to provide flexibility at the local level
The process of defining and enforcing standards takes place through a series of ˜ translations and in each such move some aspect of the standard changes and other aspects remains; what remains ˜invariant is of interest and is what we conceptually identify as the standard
Translation processes are the key to shaping the process by which a GSA relationship evolves over time, which in turn define the nature of further standardization efforts; understanding this complex interplay over time is crucial.