Modules for the makers: new rules in the smartphone industry

This is the first of a series of posts about Project Ara by Jaime Gonzalez-Capitel, discussing the benefits and challenges of modular smartphones and the validity of their design principles.

2015 might be a turning point for smart technologies as Google will be launching Project Ara, a revolutionary product inherited from Motorola and the first modular phone to ever hit the market. Prototypes unveiled over the summer muster two differently sized endoskeletons, as well as a number of third party modules that can be removed, replaced, or exchanged while the phone is powered on in a never-ending process of customization.

Besides the two size options, both the number of slots and the size of the modules are predetermined: 1×1, 1×2, 2×2; but module uses are not, and even though we can imagine a wild competition between camera manufacturers there is no limit to the kind of sensors and functions that can be pulled into the system. With its unbeatable budget adaptability and entry-level devices in the $50-100 range, promoters claim Project Ara, “exclusively designed for six billion people”, will create a new ecology of open innovation and become the new paradigm in smart technologies.

However, unless we produce a theoretical picture of what modularity is, why it’s important, and how it determines technological endeavors, and then finally analyze Project Ara’s specific form of modularity, it will be difficult to predict its impact on the history of technology and its chances for success.

Modularity 2

Pioneers aren’t (always) reckless. Whereas Project Ara is, according to the website, ‘a development effort, not an official Google product’, the partnered initiative, “Phonebloks”, is a marketing campaign that never intended to manufacture a single chip. Does caution cast shadows on the feasibility of the product? If we are to believe the general climate among geek bloggers, who seem equally enraptured and skeptical, it does. However, even if Project Ara fails in the market, we would still need to explain whether the reason for that failure lies in the specifics of implementation, in its design principles, or if market expectations in favor of the idea are inherently flawed.

In his presentations to stakeholders, Project Ara’s leader Paul Eremenko usually references his major legacies: the B-52, with a permanent structure that allowed for the betterment and replacement of components throughout six decades; OS 360, IBM’s modular computer that famously gave way to the decentralized manufacture of highly compatible PCs; and apps (“we want hardware to be developed like software”). Even though this genealogy gives a good account of the generative mechanism by which it was possible to conceptualize a modular smartphone, in the lines of what Brian Arthur has called “combin            atorial evolution[1], neither one of these ancestors nor their sum explain what is essential to this particular development.

The module definition given by the Linux Information Project is a good starting point:

A module is a self-contained component of a system (e.g., a product) which has a well-defined interface to other components of the system. There is typically some degree of substitutability among identical and/or non-identical modules within a system or between systems.

B-52s are modular because engineers improved components and added new technologies instead of developing a new model; IBM developed OS 360 with a modular approach because it diminished dramatically the need for internal communication, as no one needed to have a complete picture of progress –but it compromised technological ownership and allowed third parties to build compatible modules. Now Eremenko and his team are trying to replicate the incredible success of the PC at the scale of the hand-held device. One of the obvious questions is whether the PC-Ara analogy is realistic.

Modularity is thus the design principle that allows users to coordinate substitutable modules. According to Baldwin and Clark[2], it can be seen as a partition into visible design rules and hidden design parameters, with the former falling into three components: an architecture that specifies modules and their functions; interfaces that act as interaction guidelines; and standards that specify conformance to design rules. As described by the authors in a more recent text[3], modular systems can evolve by means of modular operations like splitting, substitution, exclusion, augmentation, inversion, and porting, among other operators “identified in empirical investigations of design evolution”. An important measure of Project Ara’s modularity is its conformity to this cogent taxonomy.

Perhaps more importantly, modularity is ‘a financial force that can change the structure of an industry’, as it increases the possibilities of evaluating an exponentially higher number of design alternatives with aggregated economic value (Baldwin & Clark, 2006). In other words, it is also an organizational design principle, the purest form of which is the market interface embedded in product development:

we expect firms to undertake internal development of fewer components, as more product-creating firms learn how to use modular architectures to source more components through loosely coupled networks of component suppliers.[4]


In the next post I will dive deeper into Project Ara with the use of the literature discussed above and contrast whether it truly meets the criteria of modular design, and what kind of architectural, organizational and developmental modularity is used in the project. In the third and final post of the series, I will discuss the broader picture of market implications if the product is successful. How would popular modular products impact us? Is this a new idea or just the continuation of an old trend?  And finally, do we as consumers even demand a product like Project Ara?

[1] Brian Arthur, The Nature of Technology. What It Is and How it Evolves, New York, NY: Free Press, 2009.

[2] Carliss Y. Baldwin & Kim B. Clark, Design Rules: The Power of Modularity, The MIT Press, 2000.

[3] Carliss Y. Baldwin and Kim B. Clark. “Modularity in the Design of Complex Engineering Systems.” From Complex Engineered Systems: Science Meets Technology, edited by Dan Braha, Ali A. Minai, and Yaneer Bar-Yam, 175-205. Cambridge, MA: Springer, 2006

[4] Sanchez and Mahoney, Modularity, Flexibility, and Knowledge Management in Product and Organization Design, Strategic Management Journal, Vol. 17, Special Issue: Knowledge and the Firm. (Winter, 1996), pp. 63-76.

Bio: Jaime Gonzalez-Capitel is a visiting Fulbright from Spain and a Master’s Candidate in the Communication, Culture and Technology program at Georgetown University. With an interest in emerging patterns of civic engagement, his current work on the communication strategies of Think Tanks and their democratic function can be followed at

Guest Author

If you would like to submit a blog as a gnovis guest author, please contact us at