I found it fascinating. In Conway’s own words, his thesis can be summarised as follows:
Any organization which designs a system (defined more broadly here than just information systems) will inevitably produce a design whose structure is a copy of the organization’s communication structure
The implications are interesting. An organisation that uses blogs and wikis and IM on a level-free non-hierarchical basis, where collaboration takes place over time and distance and silo and culture, will in time produce designs for “Conway systems†that replicate the communications structure. Is an organisation chart a Conway system? Is collaboration really that subversive?
In isolation, perhaps. But the adoption of social software and collaborative tools is counterbalanced by opposing factors such as Sarbanes-Oxley and its ilk, which reinforce hierarchies. The complexity of corporate law and tax structures also forces regional entity obeisance, again underpinning hierarchy. What Seely Brown and Hagel witnessed in China, was it despite the operating structures rather than because of them? I wonder.
Which leads me down a Chandleresque path, in terms of his suggested interplay between strategy and structure. Something to think about, how modern communications cultures will influence the organisational structures of tomorrow, despite post-facto regulation.