Harris Kyriakou | Collective Innovation


Current Research Projects



MIS Quarterly

Knowledge Reuse for Customization: Metamodels in an Open Design Community for 3D Printing

Theories of knowledge reuse posit two distinct processes: reuse for replication and reuse for innovation. We identify another distinct process, reuse for customization. Reuse for customization is a process in which designers manipulate the parameters of metamodels to produce models that fulfill their personal needs. We test hypotheses about reuse for customization in Thingiverse, a community of designers that shares files for three-dimensional printing. 3D metamodels are reused more often than the 3D models they generate. The reuse of metamodels is amplified when the metamodels are created by designers with greater community experience. Metamodels make the community’s design knowledge available for reuse for customization—or further extension of the metamodels, a kind of reuse for innovation.




Collective Innovation in Open Source Hardware

A growing community that shares digital 3D designs has created an opportunity to study, encourage and stimulate innovation. This remix community allows people not only to prototype at a minimal cost but also to work on projects they are genuinely interested in. Participants free of the limitations typically imposed by formal organizations develop products driven by their own interest.


Idea Inheritance, Originality, and Collective Innovation

This paper presents a new method for gauging innovation, and suggests ways of further understanding the role technology plays in encouraging creativity. From an organization perspective, this work provides insights into the creative process, and in particular the open innovation process, in which thousands of individuals together evolve designs, without belonging to the same corporate structure, without claiming IP rights, without exchanging money.


Networks of Innovation in 3D Printing

Innovation inside companies is difficult to see. But an emerging online community of inventors who publicly post 3D CAD drawings of their work provide a way to observe – and perhaps amplify – innovation. Our results suggest that analysis of remix network structure may provide ways of tracing innovation processes and detecting the emergence of new ideas, combination of disparate ideas.


Twitch Plays Pokemon

We are examining the interactions between 700,000 people that collectively controlled and finished a Pokemon Red game over the course of two weeks. The players voted in favor of anarchy or democracy mode more than 4 million times.







We are examining the effect of ownership on someone’s opinion related to a product and its competitors.



We are examining the differences between online and offline friendships.


Design & Market Success

We are examining the effect of design novelty to car sales.



We are looking at how we can predict future visits and ratings of businesses listed in Yelp based on prior user behavior.




Novelty and Reuse in an Open Innovation Community

Open innovation platforms store data that can be used to study the evolution of designs in the open. Network science can be applied to further our understanding of design inheritance. Objective methods of distance between designs based on their form and function can also help us understand the differences between proposed designs, and their prospective use and reuse.
We combine network analysis methods with objective methods of distance between product designs in an open innovation community to understand how specific attributes of the artifacts reused in the creation process may affect the usage of the newly created designs. The two distance measure methods capture the differences between designs in terms of their (i) shape and (ii) function. The general finding, based on analysis of a large product network of 3D printing designs, is that strategies of heterogeneous inheritance are usually better than pure strategies. Designs inheriting from a combination of novel and imitative designs do well, as do designs that are created near to some inherited designs and far from others. These findings provide insights into the current affordances of a large open innovation community and suggest ways of architecting better systems to support these communities.