Powered by
8th International Workshop on Social Software Engineering (SSE 2016),
November 14, 2016,
Seattle, WA, USA
8th International Workshop on Social Software Engineering (SSE 2016)
Frontmatter
Message from the Chairs
The Workshop on Social Software Engineering (SSE) focuses on the interplay between social computing and software engineering. On one hand, social factors in software engineering activities, processes and tools are essential for improving the quality of development processes and the software produced by them. Examples include the role of situational awareness and multi-cultural factors in collaborative software development. On the other hand, social software mediates people-to-people communication, supporting human choices, actions, and interactions with each other. Social software needs to accommodate a wide range of social concepts, such as trust, governance, reputation, and privacy. Being social, the software would also need to be receptive to users’ choices and give them a voice in the design, operation and evolution decisions. The SSE workshop brings together academic and industrial perspectives to provide models, methods, tools and approaches to address these issues.
Papers
Understanding Git History: A Multi-Sense View
Kevin J. North,
Anita Sarma, and
Myra B. Cohen
(University of Nebraska-Lincoln, USA; Oregon State University, USA)
Version control systems archive data about the development history of a project, which can be used to analyze and understand different facets of a software project. The project history can be used to evaluate the development process of a team, as an aid in bug fixing, or to help new members get on track with development. However, state of the art techniques for analyzing version control data provide only partial views into this information, and lack an easy way to present all the dimensions of the data. In this paper we present GitVS, a hybrid view that incorporates visualization and sonification to represent the multiple dimensions of version control data - development time line, conflicts, etc. In a formative user study comparing the GitHub Network Graph, GitVS, and a version of GitVS without sound, we show GitVS improves over the GitHub Network Graph and that while sound makes it easier to correctly understand version history for some tasks, it is more difficult for others.
@InProceedings{SSE16p1,
author = {Kevin J. North and Anita Sarma and Myra B. Cohen},
title = {Understanding Git History: A Multi-Sense View},
booktitle = {Proc.\ SSE},
publisher = {ACM},
pages = {1--7},
doi = {},
year = {2016},
}
Info
VDML4RS: A Tool for Reputation Systems Modeling and Design
Lorenzo Bettini and
Sara Capecchi
(University of Florence, Italy; University of Turin, Italy)
Reputation has a great influence on interactions between system providers and users
(e.g. in e-commerce). Despite that, reputation systems implemented so far show
many deficiencies and suffer from reliability problems.
One of the causes is the lack of appropriate software engineering methodologies dedicated to requirements collection and design of reputation systems. On the other hand, business process modeling, service
design and reputation system modeling share many needs (represent
a variety of entities involved in the related domains), aims (improve services/business in order to meet the user's needs) and problems (they all involve
stakeholders with very different backgrounds and skills). Starting from the
intuition above, this paper proposes VDML4RS a tool for conceptual modeling of reputation systems. The aim is to provide a tool to represent reputation requirements
of the system-to-be from the very beginning in the software development allowing mutual understanding between requirements engineers and stakeholders avoiding the socio-technical mismatch in communication that can delay and harm the development.
@InProceedings{SSE16p8,
author = {Lorenzo Bettini and Sara Capecchi},
title = {VDML4RS: A Tool for Reputation Systems Modeling and Design},
booktitle = {Proc.\ SSE},
publisher = {ACM},
pages = {8--14},
doi = {},
year = {2016},
}
On the Impact of Social Network Information Diversity on End-User Programming Productivity: A Foraging-Theoretic Study
Xiaoyu Jin,
Nan Niu, and
Michael Wagner
(University of Cincinnati, USA; Cincinnati Children's Hospital Medical Center, USA)
Social network information (SNI) plays an important role in providing hints to help facilitate daily software engineering tasks, especially for end-user programmers. Social information foraging theory quantitatively predicts the effect of diversity of hints on productivity. In this paper, we explore how to best leverage this theoretical prediction to support software change tasks. Specifically, we analyze the data from an observational study involving 20 bioinformatics researchers using SNI to solve software change tasks. We further classify the SNI support by using 5 diversity categories: social network type (e.g., wiki, Q&A, etc.), contributor role (e.g., core, marginal, etc.), number of contributors, information needs concerning software architecture, and information needs organized by complexity. Our results show that the contributor role best manifests the hint diversity, and its incorporation with architectural considerations could further improve productivity. Our research offers principled guidelines for supporting better use of SNI in end-user programming.
@InProceedings{SSE16p15,
author = {Xiaoyu Jin and Nan Niu and Michael Wagner},
title = {On the Impact of Social Network Information Diversity on End-User Programming Productivity: A Foraging-Theoretic Study},
booktitle = {Proc.\ SSE},
publisher = {ACM},
pages = {15--21},
doi = {},
year = {2016},
}
proc time: 0.89