A software project management classic for the twenty-first century: in-the-trenches wisdom from legendary project leader Joe Marasco.
° Helps readers ship products they're proud of, on time and under budget, with real customer value
° Based on the authors' tremendously popular Franklin's Kite columns in IBM/Rational's popular e-zine: Teh Rational Edge
° Pre-publication promotion at IBM's Rational Software Development Conference (May 2005, Las Vegas)
"This set of articles captures decades of in-the trenches experience across a broad spectrum of software topics. Joe Marasco has the scars and the smarts to articulate patterns of success that can satisfy a broad audience. He uses mathematics, physics, common sense, and storytelling along with a no-candy-coating style to provide unique perspectives on significant problems in delivering software results as a business. Whether you are a computer science theoretician, a frustrated software project manager, a successful businessman, or a skeptical programmer, you will learn a lot from this compilation."
Walker Royce, Vice President, IBM Software Services-Rational, and author of Software Project Management (Addison-Wesley)
"Joe Marasco's readable essays on managing successful projects show that software development managersno different from all managersmust embrace the fundamentals of management if they are to succeed: working through people and process to be decisive, dealing with politics, keeping on schedule, and, yes, shipping a well-developed product. Marasco uses plain English to explain many integrated skills, ranging from estimating the time it will take to really do things, to negotiating effectively, even to eloquently describing three distinct phases of our personal development. He frequently uses a 'can we talk?' conversation with a fictional colleague, Roscoe Leroy, in a Socratic dialogue to illustrate the two sides to a point in many areas (reminiscent of Galileo's writings to explain his then-heretical views); in this case, Marasco's advice will help technology professionals escape the clutches of pervasive Dilbertian incompetence, and enable readers to be more effective in our ever-changing world."
Carl Selinger, author of Stuff You Don't Learn in Engineering School: Skills for Success in the Real World (Wiley-IEEE Press), and contributing editor of IEEE Spectrum magazine
The new software management classic: in-the-trenches wisdom from legendary project leader Joe Marasco
Over the course of a distinguished career, Joe Marasco earned a reputation as the go-to software project manager: the one to call when you were facing a brutally tough, make-or-break project. Marasco reflected on his experiences in a remarkable series of "Franklin's Kite" essays for The Rational Edge, Rational and IBM's online software development magazine. Now, Marasco collects and updates those essays, bringing his unique insights (and humor) to everything from modeling to scheduling, team dynamics to compensation. The result: a new classic that deserves a place alongside Frederick Brooks' The Mythical Man-Month in the library of every developer and software manager. If you want to ship products you're proud of... ship on time and on budget... deliver real customer value... you simply must read The Software Development Edge.
How software projects resemble other projectsand how they're different
The iterative problem-solving clock: ending the day with real solutions
The realities of scheduling: How late are you going to be?
Trade-offs, estimating, project rhythm, and getting products out the door
Understanding what you're seeing, hearing, and feeling as a software manager
The human element: politics, negotiation, compensation, culture, and growth
Avoiding crises before they happen... and mitigating them when they do
Thinking laterally: original ideas in software project management
© Copyright Pearson Education. All rights reserved.
Download the Sample
Chapter related to this title.
Download the sample pages (includes Chapter 8 and Index)
About the Author.
I. GENERAL MANAGEMENT.
1. Beginning at the Beginning.
The Importance of Good Software.
Hard Rocks in the Swamp.
The Iterative Problem-Solving Clock.
2. Computational Roots.
How This Program Worked.
Why Was This Generation of Engineers Special?
Getting to Know the Numbers by Their First Names.
So How About Those Computers?
Our Computational Heritage.
On Climbing Big Mountains.
Common Causes of Failure.
Ingredients for Success.
The Human Factor.
II. SOFTWARE DIFFERENCES.
5. The Most Important Thing.
Going Over the Waterfall.
The Other Extreme.
Roscoe’s First Picture.
Roscoe’s Second Picture.
Wait a Minute!.
Keeping the Vectors Short.
The Application to Software Development.
Applied Learning and Short-Vector Direction.
Have You Heard This One Before?
More on Applied Learning.
The Staffing Effect.
Just Plain Horse (shoe) Sense.
How to Explain the UML.
What Is the UML, and Why Is It Important?
A Second, Less Trivial Example.
The Third Example.
And Now for the Relevance to Software…
Raising the Level of Abstraction.
How Managers Can Learn a New Programming Language.
The Problem, Better Defined.
What Should the Standard Problem Contain?
The Animal Game.
Does the Animal Game Fit the Criteria?
Does It Pass the “So What?” Test?
It’s Your Game.
8. Getting It Out the Door.
If You Build It, They Will Come.
In the Beginning, There Was the Sandbox.
Why Should the Product Build Be Hard, Anyway?
What About Iterative Development?
III. THE PROJECT-MANAGEMENT VIEW.
The Project Pyramid.
Five, Not Four.
Enter the Pyramid.
The Altitude Variable.
The Pyramid’s Volume Is Constant.
A Statistical Interlude.
Right Idea, Wrong Distribution.
Implications for Real Projects.
What Does It Take to Get to a Coin Flip?
It’s All About Risk.
What If We Used Common Sense?
Chocolate Versus Vanilla.
Roscoe Goes Deeper.
Roscoe Gets into Software.
Roscoe Reports In.
Guess We Did Something Right.
Roscoe Sums It Up .
Roscoe Picks a Bone.
Guess We Did Something Right, Part Two.
Roscoe Admitted to Software Project Manager Fraternity.
Roscoe Poses the Problem: How Late Are You Gonna Be?
Joe Makes a Slight Comeback.
Roscoe’s Rogue’s Gallery.
One Last Objection.
Roscoe’s Parting Shot.
A Physicist Looks at Project Progress.
What About Iterative Development?
One Last Graph.
IV. THE HUMAN ELEMENT.
Politics Is Inevitable, But…
When Things Get Political.
The Engineering Mapping.
Other Variants of Bad Politics.
Communication Is Everything.
Roscoe Explains His Theory.
Are We Done Yet?
15. Signing Up.
Roscoe Gets His Nose Bloodied…
…And Immediately Cuts to the Chase.
How They Do It in Texas.
The Relevance to Software.
The Dog Ate My Homework.
The Three Most Common Excuses.
And Another Thing…
Thrust, Parry, and Riposte.
Large Project Chicken.
The End of Software Development as We Know It?
Elaboration Versus Construction.
Going for the Flow.
Flow and Software Development Performance.
Applying the Flow Model to Compensation.
Money Isn’t Always the Answer.
V. THINKING LATERALLY.
17. History Lesson.
Don’t Let the King Be Your Architect.
Things Aren’t Always as They Seem.
Checking the Design.
Knowing What You Don’t Know.
Continuity of Leadership.
In a Hurry, As Usual.
Focusing on the Wrong Features.
When the Design Is Bad…
The Relevance of Testing.
Prototype Versus Product.
18. Bad Analogies.
Houston, We Have a Problem.
19. The Refresh Problem.
Refreshing Embedded Software.
The Current Situation.
The Software Upgrade Game.
A Modest Proposal.
Software Upgrades, Revisited.
Some Nice Things Come for Free.
Why This Will Work.
What About Software Piracy?
Until the Sun Takes Over.
20. Not So Random Numbers.
Roscoe Sets the Stage.
Simulating the Batter.
Generating More Probabilities.
Of Course, We’ve Already Left the World of Baseball.
Reality Is Ugly.
VI. ADVANCED TOPICS.
The Five Days of the Fish.
The Fish Market.
Day 1: Unaware.
Day 2: Avoiding the Issue.
Day 3: Enter “The Fixer”.
Day 4: The Turning Point.
Day 5: Two Critical Paths.
Moral of the Story.
The Naïve Model.
Consequences of the Model.
An Illustrative Example.
Call to Action.
What Is a Culture?
Strong and Weak Cultures.
Defining Corporate Values.
And the Applicability to Software Is…
Building a Strong Culture.
When You’re Looking for a Job…
The Bottom Line.
24. Putting It All Together.
More on Mensches.
Some Final Thoughts on the Model.
My intention here is to not only to collect these articles but to sew them together in a form that makes them even more useful for software development managers and their managers. I have done that by reorganizing them thematically, instead of presenting them in the order they originally appeared. This has caused me to do some light editing in places where "forward-referencing" would otherwise take place. I have also paid attention to the footnotes, many of which appeared in the original as URLs and appear here as more formal citations where appropriate. Finally, I have added material at the beginning and end of each chapter so that the context of each article as part of the whole becomes clearer.
The reader will quickly note that the chapters have several different styles. Some of them are expository, some are fairly analytical, and some are folksy "Socratic dialogs" between the author and his avatar, one Roscoe Leroy. Roscoe is an invented character, a good technical general manager who initially knows little about software development. I use him as a foil, allowing his "naïveté" to force me to explain things without using technical jargon. My approach is ecumenical and subversive: I will use any technique that permits me to get the message across. Some of these chapters will appeal to some readers, and others will appeal to others. Whatever works is, by definition, good. I take my cue from Horace, who wrote in The Art of Poetry, "He has won every vote who has blended profit and pleasure, at once delighting and instructing the reader."1
I divide the work into six parts of four chapters each. Briefly:
General Management: These chapters deal with topics that are useful to managers in general and also expose the reader to my background and biases. I include them so that we have a common baseline for what follows.
Software Differences: In this section, we take a look at those things that distinguish software development from other management challenges.
The Project-Management View: I take the perspective that a software-development project is a variant of the generic project and, as such, amenable to classical project-management techniques. On the other hand, I strive to point out what is different about software development.
The Human Element: I turn around in this section and look at software development from the perspective of the people who do it. Once again, I try to compare and contrast that which is similar to that which is different for software-development projects.
Thinking Laterally: Software people come at problems from many different points of view. In this section, I expose the reader to some of the more speculative and original ideas that he or she may not have seen before.
Advanced Topics: The successful software-development manager is like a really good pinball player: His reward for high scoring is given in free games. This additional "stick time" leads to his becoming even more proficient. In this section, I talk about some of the challenges that come with success.
This book has 24 chapters.2 You can read it serially, or pick out a chapter at a time; they can stand on their own. This is a good "airplane book"; read a chapter and then think about it for the rest of the flight. If you get just one new idea from one of the chapters that covers the price of the book, I will have been successful.
With these prefatory remarks out of the way, let's get down to it.
1 Horace, Satires, Epistles, and Ars Poetica (Cambridge, Massachusetts: Harvard
University Press, 1999). The original Latin text is "Omne tulit punctum
qui miscuit utile dulci, lectorem delectando pariterque monenendo." It
can be found at line 343 of Ars Poetica.
2 Coincidentally, so does The Iliad.
Download the Index
file related to this title.