Reading Critique Lecture

1
Reading Critique Lecture

• EEE 492A
• September 2009
• Sandy Smith

2
Assignment

• Students are to prepare a point form summary of
  Brooks Chapter 1 for discussion in the lab
• What is Brooks purpose?
• What is the argument he uses?
• Is it logical?
• What facts does he use?
• Are they accurate?
• Do you understand the subject better?
• What would you do differently in presenting this
  argument?

3
Brooks Mythical Man Month

• Mythical Man Month
• Written originally in 1975
• Popular after 20 years
• Belief that large programming projects suffer
  management problems that are different from those
  in smaller projects
• Sees critical need to be preserving the
  conceptual integrity of the product itself

4
The Tar Pit

• Mortal struggle of great beasts from prehistory
  in the morass known as the tarpit
• Moral is the more you struggle, the more
  entangled you become
• No amount of skill or strength can save you from
  sinking into the morass

5
Software Products

• Estimate 7 to 10 lines of code per day per
  programmer
• Translates into 1000 LOC per year for an
  industrial team
• Small team or programmer developing software that
  exceeds the productivity of a large team
• Why isnt all software built this way?

6
Reality of Software

• What is the small team producing?
• A program
• Complete in itself
• Runs on the system it was built on
• Stand alone

The Program
7
The Evolution of the Program

• We have a program but we dont have a product
• Our options are to...
• Develop a software product
• Develop a programming system

Programming Product
Programming System
8
Evolution to a Product
Component of a larger system
The Program
Programming System
Generally usable product
Programming Product
9
Evolution to a Product - Cost
X 3
The Program
Programming System
X 3
Programming Product
10
Evolution to a Final Product
X 3
The Program
Programming System
X 3
X 9

• Intended product
• Truly useful object

Programming Product
Programming Systems Product
11
Joys of the Craft

• Making things is fun
• Making useful things is rewarding
• Fashioning complex objects is interesting
• Learning while doing
• Easily worked medium

12
Woes of the Craft

• Perfection
• Dependency on others and lack of formal authority
• Debugging is hard work with little reward
• Debugging is time consuming
• Obsolescence

13
The Challenge

• Find real solutions to real problems
• To do this on schedule with the resources
  available
• The Tar Pit