• DocumentCode
    731499
  • Title

    Lessons Learned from Building and Deploying a Code Review Analytics Platform

  • Author

    Bird, Christian ; Carnahan, Trevor ; Greiler, Michaela

  • Author_Institution
    Microsoft, Redmond, WA, USA
  • fYear
    2015
  • fDate
    16-17 May 2015
  • Firstpage
    191
  • Lastpage
    201
  • Abstract
    Tool-based code review is growing in popularity and has become a standard part of the development process at Mi-crosoft. Adoption of these tools makes it possible to mine data from code reviews and provide access to it. In this paper, we pre-sent an experience report for CodeFlow Analytics, a system that collects code review data, generates metrics from this data, and provides a number of ways for development teams to access the metrics and data. We discuss the design, design decisions and chal-lenges that we encountered when building CodeFlow Analytics. We contacted teams that used CodeFlow Analytics over the past two years and discuss what prompted them to use CodeFlow Ana-lytics, how they have used it, and what the impact has been. Fur-ther, we survey research that has been enabled by using the Code-Flow Analytics platform. We provide a series of lessons learned from this experience to help others embarking on a task of building an analytics platform in an enterprise setting.
  • Keywords
    data mining; program diagnostics; CodeFlow analytics; Microsoft; analytics platform; code review analytics platform; data mining; design decisions; enterprise setting; tool-based code review; Computational modeling; Data mining; Databases; Interviews; Measurement; Servers; Software;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Mining Software Repositories (MSR), 2015 IEEE/ACM 12th Working Conference on
  • Conference_Location
    Florence
  • Type

    conf

  • DOI
    10.1109/MSR.2015.25
  • Filename
    7180079