Mix4Crowds - Toward a Framework to Design Crowd Collaboration with Science Alexandre P. Uchoa, Maria Gilda P. Esteves and Jano M. de Souza PESC/COPPE, Graduate School of Engineering UFRJ, Federal University of Rio de Janeiro, RJ, Brazil {auchoa, gilda, jano}@cos.ufrj.br Abstract—Citizen Science projects are characterized by mass collaboration of ordinary citizens with scientific research. The advent of new crowdware technologies and the ease of access made possible by the Internet are reshaping the way people and scientific organizations work together. However, the particular nature of this form of collaboration requires a tailored approach that increases the chances of all involved parties meet their expectations and achieve their goals. We propose a conceptual framework to help Citizen Science enterprises in conceiving their crowdsourcing strategy and designing their collaboration systems. It departs from basic principles of traditional marketing models and incorporates the specific requirements of this increasingly popular but also complex form of web-based collaborative work. Keywords— public engagement; mass collaboration; marketing mix; crowdsourcing; citizen science; crowdware. I. INTRODUCTION The demand for knowledge imposed by problems of global impact are requiring from numerous research areas alternative models of scientific projects capable of overcoming the challenge of obtaining and processing increasing amounts of data. One model that is attracting crescent interest from the scientific community is Citizen Science, a form of crowdsourcing in which volunteers are invited to collaborate with scientists applying some human cognitive ability in large scale [1], [2], [3], [4], [5]. Thanks to the Internet, Citizen Science is expanding its reach and establishing itself as a cheap and viable alternative for scientific investigations that were previously not viable or restricted to the limited capacity of professional teams [4], [7]. But some challenges interpose. Purpose-driven collaborative crowdsourcing, especially the forms that rely on voluntary web-based cooperation, have to successfully bridge the rigidity of leadership and the goal orientation typical of projects with the flexible minded, relaxed and lively atmosphere that characterizes collaborative virtual environments [8], [9]. Professional researchers, used to work between walls and to be conditioned to strict deadlines and criteria, find themselves depending on the work of amateur volunteers that, in many cases, are just seeking for leisure [7]. This hybrid nature of Citizen Science requires from the designers of this type of crowdsourcing systems the ability to accommodate functionalities and features that simultaneously satisfy groups that may be quite different. However, there is still relatively little theoretical and empirical knowledge to support this quest [3]. Aiming to contribute towards filling this gap, we present in this paper a conceptual framework to help scientific organizations with the design of their crowdsourcing systems. Called ‘Mix for Crowds' (Mix4Crowds), the framework has its foundations on traditional marketing mix models and employs a four-dimension approach to help designers find a coherent, integrated and balanced mix of features and functionalities that address the requirements and expectations of all stakeholders, including the crowd. After introducing Citizen Science and how it relates to collective intelligence initiatives, we briefly discuss the convenience of employing marketing tools when developing systems intended to serve and attract the public. In the following parts, each of the four design dimensions of the framework is described accompanied by some suggestions of design aspects and requirements that may be considered by designers. II. CITIZEN SCIENCE – THE COLLABORATION OF THE CROWD WITH SCIENCE Citizen Science entirely mediated by computer and/or mobile devices systems is a relatively new paradigm of distributed scientific collaboration. It combines crowdsourcing, human computation and crowdware systems to explore the work force and the collective intelligence of the crowd towards solving scientific problems. The combination of mass collaboration of ordinary citizens with the scientific work of researchers demands from CSCW systems designers attention both to the task requirements of the research and the aspirations of potential and actual contributors so that the collaboration provide the desired results for all parties. Crowdware systems takes inspiration from groupware and was defined by Schneider and colleagues [13] as “a class of systems for supporting virtual and real crowds in order to provide advanced services anywhere and anytime, connecting individuals in heterogeneous environments”. Blogs, wikis, social network services, web-based crowd collaboration systems and application enabled mobile devices are all forms of crowdware. With the help of the appropriate combination of crowdware systems, volunteers can assist researchers collecting and/or analyzing massive amounts of data that cover long periods of time or large geographic areas, e.g. eBird (ebird.org); or employing some human cognitive ability in large scale [3] e.g. processing herbarium specimen sheets: Herbaria@home (herbariaunited.org); classifying images of galaxies: Galaxy Zoo (galaxyzoo.org) or discovering native Proceedings of the 2013 IEEE 17th International Conference on Computer Supported Cooperative Work in Design 978-1-4673-6085-2/13/$31.00 ©2013 IEEE 61