In 1997, the renowned open source advocate Eric Raymond published an online essay entitled « The Cathedral and the Bazaar ». Later published as a book, Raymond’s demonstration described two antithetical models of organization which he derived from his years of activity in the field of open source software.

This post was initially published on the MosaiC website.

On one side, the cathedral, a structured, top-down endeavour based on the restricted diffusion of information and knowledge, on the other, the bazaar, a self-structured, flat collective of individuals in which constant sharing is the key to fluidity and improvement. In both cases, a delicate balance between efficient coordination and randomized exploration has to be struck. Raymond’s archetypes, like those of Weberian ascent, are thus to be considered as ideal-types that have little applicability to practical cases. Nonetheless, these depictions remain guides to interpret the organizations we engage with, either as customers, employees or creators.

Raymond’s depictions of the design processes prevalent in the software industry can be extended to describe modern phenomena of management that combine chaotic and structuring dynamics. In large-scale enterprises of a more or less bureaucratic nature, the driving forces that support the cathedral model are the only ones which may legitimately abided by. In companies like Bell Canada, for instance, senior managers operate under the belief that bazaar-like enterprises can only be resorted to once a certain number of parameters have been fully brought under rational control. From a communications and marketing perspective, an eventual opening towards social networking platforms such as Twitter and Facebook – which many marketing specialists today consider a necessity – can only be considered once the customer service targets under the "regular regime" have been achieved.

Although Bell executives have stated that recent assessments show that the issue is increasingly mastered, one may ask quite simply whether in a competitive environment that’s constantly transformed by technological change in devices and connectivity, there will ever be a time of ‘sufficient control’. In this case, the opening towards the bazaar may not be something that’s decided, but rather something that simply emerges. In the case of Bell, Twitter accounts under the names @BellCanadaWatch and @BellCanadaPR, which are respectively described as "a place for Bell to hear that people are dissatisfied with their service" and "the satirical and unofficial public relations feed for Bell Canada", have been created by unidentified stakeholders. In this case, cathedral and bazaar chose to mutually exclude the realms of significance in which the other evolves, and in doing so, lose an opportunity to meet halfway and create. It is also interesting to point out that even the most well-built and apparently strong cathedrals host a great deal of inherent chaos. Work on routines in structured, rule-based organizations has shown that routines change, and are active factors in the transformation of work practices. The case of hiring routines, as exemplified in the writings of Martha Feldman and Brian Pentland, notably show how, from one repetition to the next, the sub-processes involved in hiring a new person may vary significantly while the macro-process retains general characteristics that make it fall under the label "hiring routine".

An unplanned contribution, courtesy of Emilie Pawlak, HEC Liege (Belgium).

The bazaar, on the other hand, is a form of self-organization that is auto-coordinated in which every individual more or less sits at the same level and contributes as far as he or she desires. The bazaar is a transparent, trust-based form of coordination in which every actor takes responsibility for a part of the process without necessarily expecting an immediate return from a specific counterpart. The example of Montreal’s Societe des Arts Technologiques (SAT), as described by its founder and president Monique Savoie, is that of an internal market of projects in which a more or less stable set of individuals are regularly reassigned in order to cope with the creative turmoil generated by its project leaders. The image used by Ms. Savoie to describe this process is that of the Moroccan "souk", a barter-based system of exchanges between merchants and acquirers that can be transposed, in the case of SAT, to a "market for ideas".

Yet, the SAT cannot escape a certain, basic, structuring of its activities. In that, we realize that organizations that resemble the bazaar ideal-type, be they called Linux or Sid Lee, still need constraints in order to evolve positively in their respective domains of activity. In that, most functional and successful bazaars are partly cathedrals in the making. On the other hand, cathedrals like those we have observed, due to their size and relative inability to control everything, become the hosts of several collectives and communities that act as internal or external bazaars. These informal structures bear the potential to articulate alternative subversive discourses. Hence, when senior executives of cathedrals fear that, to open the door to bazaar-like initiatives would imperil their cathedral-building activities, they are calculating a risk based on the wrong parameters. "The bazaar would kill us", they say. Perhaps the opposite is true, rather.

Author: Francis Gosselin