A User Evaluation of a Collaborator Recommender based on Co-Changed Files

Active collaboration is essential for the success of software projects across the development life-cycle. Unfortunately, in social coding platforms, such as GitHub, it is still challenging for developers to identify potential collaborators with whom they could engage to create new/stronger ties and...

Full description

Bibliographic Details
Main Authors: Kattiana Constantino, Raquel Prates, Eduardo Figueiredo
Format: Article
Language:English
Published: Brazilian Computer Society 2024-03-01
Series:Journal on Interactive Systems
Subjects:
Online Access:https://sol.sbc.org.br/journals/index.php/jis/article/view/3728
_version_ 1797271587645292544
author Kattiana Constantino
Raquel Prates
Eduardo Figueiredo
author_facet Kattiana Constantino
Raquel Prates
Eduardo Figueiredo
author_sort Kattiana Constantino
collection DOAJ
description Active collaboration is essential for the success of software projects across the development life-cycle. Unfortunately, in social coding platforms, such as GitHub, it is still challenging for developers to identify potential collaborators with whom they could engage to create new/stronger ties and enhance the quality of contributions. To this end, we implemented developer recommendation strategies and prototype tool to help project contributors improve their collaborations. Thus, in this work, we described a controlled experimental study concerned usability and user satisfaction to investigate the developers’ perceptions of using CoopFinder, a prototype tool to support two strategies for recommending collaborations. These developer recommendation strategies aim to connect developers of a specific project based on their similar interests. The study involved 35 participants, 18 of which were GitHub users, and 17 were non–GitHub users. We asked participants to perform the experiment tasks to find collaborators with similar interests using a prototype recommendation tool and GitHub. We reported a quantitative and qualitative evaluation of strategies and tool using the state of the practice as a baseline. As a result, we observed that recommender based on co–changed files can provide suitable collaborator recommendations to developers of a specific project. About 66% of the participants confirmed they would use or recommend this tool.
first_indexed 2024-03-07T14:06:45Z
format Article
id doaj.art-4badfac484f041968ea1027bc2f1b389
institution Directory Open Access Journal
issn 2763-7719
language English
last_indexed 2024-03-07T14:06:45Z
publishDate 2024-03-01
publisher Brazilian Computer Society
record_format Article
series Journal on Interactive Systems
spelling doaj.art-4badfac484f041968ea1027bc2f1b3892024-03-06T18:54:41ZengBrazilian Computer SocietyJournal on Interactive Systems2763-77192024-03-0115110.5753/jis.2024.3728A User Evaluation of a Collaborator Recommender based on Co-Changed FilesKattiana Constantino0Raquel Prates1Eduardo Figueiredo2Federal University of Minas GeraisFederal University of Minas GeraisFederal University of Minas Gerais Active collaboration is essential for the success of software projects across the development life-cycle. Unfortunately, in social coding platforms, such as GitHub, it is still challenging for developers to identify potential collaborators with whom they could engage to create new/stronger ties and enhance the quality of contributions. To this end, we implemented developer recommendation strategies and prototype tool to help project contributors improve their collaborations. Thus, in this work, we described a controlled experimental study concerned usability and user satisfaction to investigate the developers’ perceptions of using CoopFinder, a prototype tool to support two strategies for recommending collaborations. These developer recommendation strategies aim to connect developers of a specific project based on their similar interests. The study involved 35 participants, 18 of which were GitHub users, and 17 were non–GitHub users. We asked participants to perform the experiment tasks to find collaborators with similar interests using a prototype recommendation tool and GitHub. We reported a quantitative and qualitative evaluation of strategies and tool using the state of the practice as a baseline. As a result, we observed that recommender based on co–changed files can provide suitable collaborator recommendations to developers of a specific project. About 66% of the participants confirmed they would use or recommend this tool. https://sol.sbc.org.br/journals/index.php/jis/article/view/3728Open-Source Software ProjectsCollaborative Software DevelopmentDistributed CollaborationDeveloper Recommendation
spellingShingle Kattiana Constantino
Raquel Prates
Eduardo Figueiredo
A User Evaluation of a Collaborator Recommender based on Co-Changed Files
Journal on Interactive Systems
Open-Source Software Projects
Collaborative Software Development
Distributed Collaboration
Developer Recommendation
title A User Evaluation of a Collaborator Recommender based on Co-Changed Files
title_full A User Evaluation of a Collaborator Recommender based on Co-Changed Files
title_fullStr A User Evaluation of a Collaborator Recommender based on Co-Changed Files
title_full_unstemmed A User Evaluation of a Collaborator Recommender based on Co-Changed Files
title_short A User Evaluation of a Collaborator Recommender based on Co-Changed Files
title_sort user evaluation of a collaborator recommender based on co changed files
topic Open-Source Software Projects
Collaborative Software Development
Distributed Collaboration
Developer Recommendation
url https://sol.sbc.org.br/journals/index.php/jis/article/view/3728
work_keys_str_mv AT kattianaconstantino auserevaluationofacollaboratorrecommenderbasedoncochangedfiles
AT raquelprates auserevaluationofacollaboratorrecommenderbasedoncochangedfiles
AT eduardofigueiredo auserevaluationofacollaboratorrecommenderbasedoncochangedfiles
AT kattianaconstantino userevaluationofacollaboratorrecommenderbasedoncochangedfiles
AT raquelprates userevaluationofacollaboratorrecommenderbasedoncochangedfiles
AT eduardofigueiredo userevaluationofacollaboratorrecommenderbasedoncochangedfiles