Software quality observatory for open source software

Información del proyecto

SQO-OSS

Identificador del acuerdo de subvención: 033331

Proyecto cerrado

Fecha de inicio 1 Septiembre 2006

Fecha de finalización 31 Agosto 2008

Financiado con arreglo a

• Information Society Technologies: thematic priority under the specific programme "Integrating and strengthening the European research area" (2002-2006).

Coste total € 2 469 800,00

Aportación de la UE € 1 638 800,00

Coordinado por

ATHENS UNIVERSITY OF ECONOMICS AND BUSINESS - RESEARCH CENTER
Greece

Exploitable results

Highlights of achievements for SQO-OSS

The SQO-OSS project is a holistic approach to software assessment, initially targeted at Open Source software. SQO-OSS uses as many sources of quality indicators as possible to create a set of metrics that can automatically, or with the least possible human intervention, be applied to a software project’s repository in order to extract quantifiable measurements of its quality. The project delivered a framework whose functionality can be extended by attaching plug-in metric components, and that can interoperate with web or application interfaces in order to allow user control.&L<he consortium’s main goal was to build a web site of quality assessed open source software, which would be based on a back-end engine that would run various process and product metrics on a continuous basis. As a by-product of this effort, the project also delivered novel quality assessment tools that combine data from various project data sources, along with various non-technical deliverables such a method for evaluating software quality using the SQO-OSS tools. The main target application set for SQO-OSS was OSS, due to the fact that all software development data is readily available. However, as many OSS software development and project management tools have proliferated into proprietary software development teams, the tools SQO-OSS produced are universally applicable. The project’s achievements include: - the SQO-OSS community web site created under http://www.sqo-oss.org/. - The consortium delivered a major version (0.8.1) of the SQO-OSS system. - Created collaboration environment with SVN repository and TRAC. - Defined the SQO-OSS consortium Quality plan. - Created custom reporting tools based on the SQO-OSS SVN repository. - Created the SQO-OSS software quality model. - Written 116,544 (on last count) lines of code and documentation, scattered around in 2310 source code files. - More than 700 open source projects are being actively mirrored (almost 100GB data), in terms of version control systems, bug databases, and mailing lists. - The http://demo.sqo-oss.org/ server currently processes more than 50 of those projects with more projects are added to the processing engine every day.&L - The consortium published 16 scientific papers in journals and conferences. - The consortium organised 38 SQO-OSS related events, to disseminate the knowledge gained from the project, and strengthen the creation of a community. - The consortium developed more functionality than specified in the system design deliverable. (CORBA service for running metrics in languages other than Java, source code parsing service for the Java language). - 5 new metrics that use novel data sources were introduced by the consortium: – Productivity Metric (SENSE/ISTLab) - Eyeball Metric (KDE) - Mean Developer Engagement (SIRIUS) – Commit Rate (SIRIUS) – Community Communication Compaction (SIRIUS) Quality Model Metric (AUTH) 7 metric plug-ins that include many measurements were implemented: – Developer Contribution Metric - 1 Measurement (SENSE/ISTLab) - WC Metric - 11 Measurements (SENSE/ISTLab) -Simulation (DB-NET) - Module Metrics - 3 Measurements (SENSE/ISTLab) - Quality Model - 1 Measurement (AUTH) - CLMT Plug-in - 18 Measurements (SENSE/ISTLab) - Mean Developer Engagement Metric - 3 Measurements (SIRIUS) Testability Metric - 1 Measurement (SENSE/ISTLab) - IXR (Intermediate XML Representation) was designed for neutral language representation, as main technology behind the CLMT plug-in and the parsing service.