TY - GEN
T1 - What is Good Feedback in Big Data Projects for Cyberinfrastructure Diffusion in e-Science?
AU - Kee, Kerk F.
AU - McCain, Jamie C.
N1 - Funding Information:
This material is based upon work supported by the US National Science Foundation (NSF) under award ACI-1322305.
Funding Information:
The aforementioned challenges are especially true in the realm of e-science, where scientists in different domains, such as high energy physics, computational chemistry, and bioinformatics, increasingly rely on new computational tools to perform large-scale and data-intensive research in science. For most e-science projects, or big data projects in various scientific fields, the analytics technologies are often developed out of research projects federally funded by the National Science Foundation, National Institutes of Health, Department of Defense, Department of Energy, etc. Because the funding is usually short-term, the developers behind these technologies in e-science, or cyberinfrastructure (CI) [1], make the tools open source, and push them out to domain communities, in order to attract new users, new developers, and scientist-developers (scientists who can code tools for scientific research) to carry the tools forward for the long-term [2]. In order for the tools to survive when the funding ends, and remain sustainable beyond the inception projects, being responsive to users in order to promote further adoption and systemic diffusion is key. However, the communication challenge of user feedback is further compounded when the projects are running out of funding, and the user-base at the later stages of adoption and diffusion is much more diverse than in the early stages of design and development.
Publisher Copyright:
© 2018 IEEE.
PY - 2019/1/22
Y1 - 2019/1/22
N2 - This paper investigates the role of feedback in big data projects for cyberinfrastructure (CI) diffusion in e-science. For many of these projects, large-scale and heterogeneous datasets, multidisciplinary and dispersed experts, and advanced technologies are brought together to harness analytic insights. However, without effective CI and computational tools, the accuracy and meaningfulness of analytics results are compromised. In fact, without CI tools, raw data remain raw with hidden insights, as data analytics cannot be executed at all. In order to improve such tools for meaningful results, we argue to conceptualize the communication mechanism of 'feedback' in agile software development, with the goal of producing CI tools that are responsive to users. Based on a grounded analysis of interview data, we concluded that feedback helps developers in big data projects understand users' needs, makes tools user-friendly, prevents emergencies, and is better for developers than no feedback. Furthermore, good feedback is often structured, specific, actionable, timely, generalizable, and delivered in a tactful way. Despite the limitation of the findings being exploratory and yet to be evaluated experimentally, we argued that they still can motivate developers to be proactive seekers of feedback for their tools, productively guide developers' communication with users, and ultimately promote further adoption and diffusion of CI tools in e-science.
AB - This paper investigates the role of feedback in big data projects for cyberinfrastructure (CI) diffusion in e-science. For many of these projects, large-scale and heterogeneous datasets, multidisciplinary and dispersed experts, and advanced technologies are brought together to harness analytic insights. However, without effective CI and computational tools, the accuracy and meaningfulness of analytics results are compromised. In fact, without CI tools, raw data remain raw with hidden insights, as data analytics cannot be executed at all. In order to improve such tools for meaningful results, we argue to conceptualize the communication mechanism of 'feedback' in agile software development, with the goal of producing CI tools that are responsive to users. Based on a grounded analysis of interview data, we concluded that feedback helps developers in big data projects understand users' needs, makes tools user-friendly, prevents emergencies, and is better for developers than no feedback. Furthermore, good feedback is often structured, specific, actionable, timely, generalizable, and delivered in a tactful way. Despite the limitation of the findings being exploratory and yet to be evaluated experimentally, we argued that they still can motivate developers to be proactive seekers of feedback for their tools, productively guide developers' communication with users, and ultimately promote further adoption and diffusion of CI tools in e-science.
KW - agile software development
KW - cyberinfrastructure
KW - diffusion of innovations
KW - e-science
KW - feedback
KW - technology adoption
UR - http://www.scopus.com/inward/record.url?scp=85062626202&partnerID=8YFLogxK
U2 - 10.1109/BigData.2018.8622573
DO - 10.1109/BigData.2018.8622573
M3 - Conference contribution
AN - SCOPUS:85062626202
T3 - Proceedings - 2018 IEEE International Conference on Big Data, Big Data 2018
SP - 2804
EP - 2812
BT - Proceedings - 2018 IEEE International Conference on Big Data, Big Data 2018
A2 - Song, Yang
A2 - Liu, Bing
A2 - Lee, Kisung
A2 - Abe, Naoki
A2 - Pu, Calton
A2 - Qiao, Mu
A2 - Ahmed, Nesreen
A2 - Kossmann, Donald
A2 - Saltz, Jeffrey
A2 - Tang, Jiliang
A2 - He, Jingrui
A2 - Liu, Huan
A2 - Hu, Xiaohua
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 10 December 2018 through 13 December 2018
ER -