Experience Using Web-Based Shotgun Measures for Large-System Characterization and Improvement

  • Authors:

  • Sam McLellan;Al Roesler;Zongming Fei;Savita Chandran;Clay Spinuzzi

  • Affiliations:

  • Schlumberger, Austin, TX;Schlumberger, Austin, TX;Georgia Institute of Technology, Atlanta, GA;HBO and Co., Atlanta, GA;Iowa State Univ., Ames

  • Venue:
  • IEEE Transactions on Software Engineering
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

This article discusses our experience in using a World Wide Web-based shotgun measurement approach for mining and characterizing large software systems. The approach recognizes that measurement information is essentially management information, that different levels and functions of the organizational hierarchy require different information to make decisions, and that a measurement program is typically a discovery process about an organization's current modes of operations. What we found was the usefulness of a measurement program that also allows managers to dynamically formulate new goals and get answers to questions not specifically related to original goals but raised nonetheless by metric data. We describe three specific cases of decisions that were made using this approach and data collected from one large system and accessed using the company's intranet over the past two years.