Candoia: A Platform and an Ecosystem for Building and Deploying Versatile Mining Software Repositories Tools

TitleCandoia: A Platform and an Ecosystem for Building and Deploying Versatile Mining Software Repositories Tools
Publication TypeReport
Year of Publication2015
AuthorsTiwari, NM, Mills, DD, Upadhyaya, G, Lin, E, Rajan, H
Secondary TitleIowa State University Computer Science Technical Reports
Date Published11/2015
PublisherIowa State University
KeywordsAnalysis of software and its evolution, Application specific development environments, flossmole cited, msr, research to practice, software evolution, software repositories
Abstract

Research on mining software repositories (MSR) has shown great
promise during the last decade in solving many challenging software
engineering problems. There exists, however, a ‘valley of death’
between these significant innovations in the MSR research and their
deployment in practice. The significant cost of converting a prototype
to software; need to provide support for a wide variety of tools
and technologies e.g. CVS, SVN, Git, Bugzilla, Jira, Issues, etc,
to improve applicability; and the high cost of customizing tools to
practitioner-specific settings are some key hurdles in transition to
practice. We describe Candoia, a platform and an ecosystem that
is aimed at bridging this valley of death between innovations in
MSR research and their deployment in practice. We have implemented
Candoia and provide facilities to build and publish MSR
ideas as Candoia apps. Our evaluation demonstrates that Candoia
drastically reduces the cost of converting an idea to an app, thus
reducing the barrier to transitioning research findings into practice.
We also see versatility, in Candoia app’s ability to work with a variety
of tools and technologies that the platform supports. Finally,
we find that customizing Candoia app to fit project-specific needs
is often well within the grasp of developers.

Notes

" In terms of its focus,
the Candoia platform is closer to Bevan et al.’s Kenyon [9],
Bajracharya et al.’s Sourcerer [6], Gousios and Spinellis’s Alitheia
Core [32, 31], Howison et al.’s FLOSSMole [39] and different from
Boetticher et al.’s PROMISE Repository [69], González-Barahona
and Robles’s open-access data repositories [29], Black Duck OpenHub
(aka Ohloh) [13], GHTorrent [30, 33], Ossher et al.’s SourcererDB
[64], the SourceForge Research Data Archive (SRDA) [28],
and Boa [25].

URLhttp://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=1378&context=cs_techreports
Full Text