ISESE2004 ymitani EASE/NAIST 1

 

An Experimental Framework for Japanese Academic-Industry Collaboration in Empirical Software Engineering Research

Yoshiki Mitani, Mike Barker, Koji ToriiNara Institute of Science and Technology

ymitani@empirical.jp, mbarker@is.naist.jp, torii@is.naist.jp

Seishiro TsuruhoInformation-Technology Promotion Agency/

Kochi University of Technologytsuruho@ipa.go.jp

AbstractEffective research in the field of empirical software engineering requires collaboration

between academia and industry. However, in Japan, the academic side has traditionally had relatively little access to real software developers for field tests and other studies. Nor has industry had access to expertise accumulated in the academic institutions such as empirical methods, software metrics, and theories and models of software engineering. The authors are field-testing a framework for such collaboration aimed at resolving this issue.The poster illustrates this framework in the EASE (Empirical Approach to Software

Engineering) project from the viewpoint of its conception, one year of initial use, and future expansion.

References1.        The EASE Project, http://www.empirical.jp/2.        Masao Ohira, Reishi Yokomori, Makoto Sakai, Ken-ichi Matsumoto, Katsuro Inoue, Koji Torii, “Empirical Project Monitor: A Tool for Mining Multiple Project Data”, International Workshop on Mining Software Repositories (MSR2004), Scotland, UK, 2004. (To be published)3.        Collins, S. & Wakoh, H. (2000) Universities and technology transfer in Japan: Recent reforms in historical perspective. Journal of Technology Transfer 25 p. 213-2224.        Kitchenham, B., and Budgen, D. (2002) Editorial Special issue: Empirical software engineering. IEE Proceedings-Software 149(5) 113.5.        Jeffery, R. & Scott, L. (2002) Has twenty-five years of empirical software engineering made a difference? Proceedings of the Ninth Asia-Pacific Software Engineering Conference

ISESE2004 ymitani EASE/NAIST 2

Little access between Academia and Industry

AcademiaExpertise accumulation Empirical methods Software metrics Theories of Software engineering

IndustryReal software developers & field tests, other studies.Real software development data.

ISESE2004 ymitani EASE/NAIST 3

Key factors in academic-industry collaboration

EASE Project

Leadership

Explicit Product

Budget and resources

Physical work place

Project evolution strategy

Academic side

Newly developedSoftware Development EnvironmentEPM: Empirical Project monitor

Seed money: Government (MEXT)Matching fund method

Collaboration laboratoryEmpirical SE Research Lab.

Step by step: Three steps

ISESE2004 ymitani EASE/NAIST 4

EPM: Empirical Project Monitor

Software Development Environment

ConfigurationManagementData

Mailing listManagementData

IssueTrackingData

Standardized Empirical SE data (XML)

Repository (RDB)

Analyzer

Visual

EPM: Empirical Project Monitor

ISESE2004 ymitani EASE/NAIST 5

Empirical Software Engineering Research Lab.

UniversityEmpirical Software EngineeringResearch Lab.

Industry

Collecting DataValidation of analysis result

Adoption of models and knowledge Feedback

ISESE2004 ymitani EASE/NAIST 6

Empirical Software Engineering Research Framework :Step 1

EPM: Empirical Project Monitor

tool

tool

Core Academic member

Core Industry member

Empirical SE Research Lab.

Matching Fund Method

International Advisor

Government(MEXT)Budget (trigger)

Supply resourceHuman powerTechnology

cooperation

R & D : EPM DevelopmentAcademic-Industry collaboration

*1

*1: Started 2003 April

University SoftwareIndustry

ISESE2004 ymitani EASE/NAIST 7

Empirical Software Engineering Research Framework :Step 2

EPM: Empirical Project Monitor

tool

tool

Core member=EPM user company

Core Academic member

Core Industry member

Empirical SE Research Lab.

Matching Fund Method

Supply EPM & its applied technology

Offer real field dataApply EPM to real field

R & D : EPM DevelopmentAcademic-Industry collaboration

*1

*2

*2: Started 2004 April

University SoftwareIndustry

ISESE2004 ymitani EASE/NAIST 8

Empirical Software Engineering Research Framework :Step 3

EPM: Empirical Project Monitor

tool

tool

Empirical SE Community

General member: Company, Individual

Core member=EPM user company

Core Academic member

Core Industry member

Empirical SE Research Lab.

Matching Fund Method

Research, Information exchange

PublicationCase study

Supply EPM & its applied technology

Offer real field dataApply EPM to real field

R & D : EPM DevelopmentAcademic-Industry collaboration

*1

*2

*3

*3: Under Planning

University SoftwareIndustry

ISESE2004 ymitani EASE/NAIST 9

 

Empirical Software Engineering Research Framework

EPM: Empirical Project Monitor

tool

tool

Empirical SE Community

General member: Company, Individual

Core member=EPM user company

Core Academic member

Core Industry member

Empirical SE Research Lab.

Matching Fund Method

International Advisor

Government(MEXT)Budget (trigger)

Supply resourceHuman powerTechnology

Research, Information exchange

PublicationCase study

Supply EPM & its applied technology

Offer real field dataApply EPM to real field

cooperation

R & D : EPM DevelopmentAcademic-Industry collaboration

*1

*2

*3

*1: Started 2003 April

*2: Started 2004 April

*3: Under Planning

University SoftwareIndustry

ISESE2004 ymitani EASE/NAIST 10

Academia Industry

.

.

Medium of EPM

Gain access to real data from industry.Learn collaboration with industry.

Get EPM platform for free.Learn to communicate with academia.Easier access to university research.

Merit of the EASE Project