|
Addressing (Software) Crisis
|
Abstract
|
We all have problems to solve in software development. We all face some crisis at times. Some are easier to handle.
Others can be critical and they might lead to subsequent crisis. Crisis requires proper decision to make and remedy activity to follow.
It is the cumulative and aggregated arbitrary decisions and/or continuing erratic actions which might cause higher severity crisis. We call it failure or fiasco.
We need to address crisis systematically, although handling crisis is more of an art than science. It highly depends on the decision makers and their decision making process.
The systematic part of addressing crisis is given by (1) Detect exceptions – monitoring issue, (2) Expose them to responsible parties – transparency issue, (3) Make decision on them – action or no action issue, and, (4) Evaluate the outcomes – justification issue. The first two are easier. Any organization has tools and facilities to do them. The last two are much harder and might be very complex in terms of structure, functionality, behavior and unforeseen consequences. The art part is the decision model of the decision maker which is very personal and therefore different from the others. We need to look at what (elements which can be people, things/objects, activities or events), why (laddering up), and how (laddering down) to find all factors (constructs) which are involved in the exceptions. From there we build a decision grid (hopefully a full grid) on which we rate the importance or severity of the constructs (using tools and our own thought process) for analysis and evaluation.
|
Keywords
|
Crisis Management; Software Development and Management; Repertory Grid.
|
Aims and Learning Objectives
|
Share a method for understanding and self-evaluating decisions and decision making process during crisis.
|
Target Audience
|
Software developers and managers (or anyone who faces crisis one way or another)
|
Detailed Outline
|
- 1. Introduction
Quick review of past large projects with cost overruns in billions and time overruns in years
- 2. The problem
The Standish Group reports and other reports on success-failure, and in between: situation, crisis, turmoil, chaos, restart, collapse, failure
- 3. The framework
Based on the hierarchical biological spectrum consisting of components: (1) protoplasms, (2) cells, (3) humans, (4) institutions (collections of humans), (5) ecosystems (business ecosystems), (6) biosphere (economy) and analogy between components in terms of structure, functionality and behavior
- 4. The method
Focus on software exceptions and decisions rather than on success factors
- a. Detect exceptions via management by exceptions (dashboard, health checks, tracking systems, etc.)
- b. Expose exceptions for management attention and decision
- c. Understand decisions: Decision Repertory grid (psychological) as opposed to neurological decision making
- d. Measure decisions: Analysis and evaluation of repertory grid as opposed to preferences and inferences by mathematical tools: Bayesian, expected utility, etc.
- e. Remedy decisions aiming at maintaining stability, balance and equilibrium
- 5. The application
Software development management and other disciplines: Examples
- 6. Concluding remarks
|
Duration
|
1.5 hours
|
Biography of Dr. Thang Nguyen
|
Thang Nguyen holds a BS in Electrical Engineering from Laval University (Quebec, Canada), a MS in Information and Computer Science from Georgia Institute of Technology (Georgia, USA) and PhD in Information Technology and Engineering from George Mason University (Virginia, USA). He teaches MIS, E-business, Software engineering, Business Statistics and Quantitative Analysis courses. He authored a textbook on Business Statistics. His most recent journal article "A Different Approach to Information Management by Exceptions: Towards the prevention of another Enron" appeared in Information & Management (Vol 51, Issue 1), January 2014.
Thang Nguyen has served some thirty years at IBM branch and system center, Candle Corporation (now IBM Tivoli), SAIC, and others in various capabilities: senior systems engineer, senior process manager, senior developer and involved in IBM products and very large US military software development.
|
Contacts
|
e-mail: icsoft.secretariat@insticc.org
|