Software Management

Renaissance
Mekonnen Wagaw (PhD)
Outline
 Conventional Software Management

 Evolution of Software Economics

 Improving Software Economics

 The old way, and the new way

 The Old Way
Conventional Software Management

 Software crisis


“The best thing about software is its flexibility”

It can be programmed to do almost anything.


“The worst thing about software is also its flexibility”

The “almost anything” characteristic has made it difficult
to plan, monitor, and control software development.
 The Old Way The Waterfall
Model
System
requirements

Softwar
e
require
ments

Analysis Program
 Drawbacks design

 Protracted integration & late design breakage Coding

 Late risk resolution
 Requirements - driven functional decomposition
 Adversarial stakeholder relationships Testing
 Focus on document and review meetings
Maintenance
and reliance
 The Old Way Conventional Software Management
Performance
1) Finding and fixing a software problem after delivery
costs 100 times more than finding and fixing the
problem in early design phases.
2) You can compress software development schedules
25% of nominal, but no more.
3) For every $1 you spend on development, you will
spend $2 on maintenance.
4) Software development and maintenance costs are
primarily a function of the number of source lines of
code.
 The Old Way Conventional Software Management
Performance

5) Variations among people account for the biggest

differences in software productivity.
6) The overall ratio of software to hardware costs is still
growing.

In 1955 it was 15:85; in 1985, 85:15.
7) Only about 15% of software development effort is
devoted to programming.
8) Walkthroughs catch 60% of the errors.
9) 80% of the contribution comes from 20% of
contributors.
Software Economics
 Software economics is the study of how scarce
project resources are allocated for software
projects.
 Software economics helps software managers
allocate those resources in the most
efficient manner.
 Evolution of Software Economics


Most software cost models can be abstracted into a function of
five basic parameters:

Size (typically, number of source instructions)

Process (the ability of the process to avoid non-value-
adding activities)

Personnel (their experience with software development
issues and the applications domain issues of the project)

Environment (tools and techniques available to support
efficient software development and to automate process)

Quality (performance, reliability, adaptability…)
Three generations of software economics

Cost
Software size
1960s-1970s 1980s-1990s 2000 and on
Waterfall model Process improvement Iterative development
Functional design Encapsulation-based Component- based
Diseconomy of scale Diseconomy of scale Return to investment

Environments/tools: Environments/tools: Environments/tools:

Custom Off-the-shelf, separate Off-the-shelf, integrated
Size: Size: Size:
100% custom 3 0%component-based, 70% custom 7 0%component-based, 30% custom
Process: Process: Process:
Ad hoc Repeatable Managed/measured

Typical project performance

Predictably bad Unpredictable Predictable
Always: Infrequently: Usually:
-Over budget -On budget -On budget
-Over schedule -On schedule -On schedule
The predominant cost estimation process

Software manager,
software architecture manager,
software development manager,
software assessment manager

Cost modelers

Risks, options,
trade-offs,
alternatives

Cost estimate
 Pragmatic software cost estimation
 A good estimate has the following attributes:
 It is conceived and supported by the project manager,
architecture team, development team, and test team
accountable for performing the work.
 It is accepted by all stakeholders as ambitious but
realizable.
 It is based on a well defined software cost model with a
credible basis.
 It is based on a database of relevant project experience that
includes similar processes, technologies, environments,
quality requirements, and people.
 It is defined in enough detail so that its key risk areas are
understood and the probability of success is objectively
assessed.
 Improving Software Economics

 Five basic parameters of the software cost model:

1. Reducing the size or complexity of what needs to be developed

2. Improving the development process

3. Using more-skilled personnel and better teams (not necessarily

the same thing)

4. Using better environments (tools to automate the process)

5. Trading off or backing off on quality thresholds

Important Trends in Improving Software Economics
Cost model parameters Trends
Higher order languages
(C++, Java, Visual Basic, etc.)
Object-oriented
Size (Analysis, design, programming)
Abstraction and component Reuse
based development Commercial components
technologies Iterative development
Process Process maturity models
Methods and techniques Architecture-first
development Acquisition
reform
Training and personnel
Personnel
skill development
People factors
Teamwork
Win-win cultures

Environment Integrated tools

Automation technologies and tools (Visual modeling, compiler, editor, etc)
Open systems
Hardware platform performance
Automation of coding,
Quality documents, testing,
Performance, reliability, accuracy analyses

Hardware platform performance

Demonstration-based assessment
Statistical quality control
 Reducing Software Product Size

• “The most significant way to improve affordability and

return on investment is usually to produce a product
that achieves the design goals with the minimum
amount of human-generated source material.”

• Reuse, object-oriented technology, automatic code

production, and higher order programming
languages are all focused on achieving a given
system with fewer lines of human-specified source
directives.
 Reducing Software Product Size - Languages
Language SLOC per UFP

UFP -Universal Function Points

Assembly 320
The basic units of the function points
are external user inputs, C 128
external outputs, SLOC metrics
internal logic data groups, Fortran 77 105 are useful estimators for software
external data interfaces, after a candidate solution is formulated
and external inquiries. Cobol 85 91 and
an implementation language is known.
Ada 83 71
C++ 56
Ada 95 55
Java 55
Visual Basic 35
Reducing Software Product Size – Object-Oriented Methods

 An object-oriented model of the problem and its solution

encourages a common vocabulary between the end
users of a system and its developers, thus creating a
shared understanding of the problem being solved.
 The use of continuous integration creates opportunities
to recognize risk early and make incremental corrections
without destabilizing the entire development effort.
 An object-oriented architecture provides a clear separation
of concerns among disparate elements of a system,
creating firewalls that prevent a change in one part of the
system from rending the fabric of the entire architecture.
 Reducing Software Product Size – Reuse
1 Project Solution:
$N and M months Many-project
solution: Operating
with high value per unit
and Schedule Resources 2 Project Solution: investment, typical of
50% more cost and commercial products
Development Cost

100% more time

5 Project Solution:
125% more cost and
150% more time

Number of Projects Using Reusable Components

 Reducing Software Product Size – Commercial Components

APPROACH ADVANTAGES DISADVANTAGES

Commercial Predictable license costs Frequent upgrades
Broadly used, mature technology Up-front license fees
components Available now Recurring maintenance fees
Dedicated support organization Dependency on vendor
Hardware/software independence Run-time efficiency
Rich in functionality sacrifices
Functionality constraints
Integration not always
trivial
No control over upgrades and maintenance
Unnecessary features that consume extra resources
Often inadequate reliability and stability
Multiple-vendor incompatibility

Custom Complete change freedom

Smaller, often simpler implementations
Expensive, unpredictable development
Unpredictable availability date
development Often better performance Undefined maintenance model
Control of development and Often immature and fragile
enhancement
Single-platform dependency
 Three levels of processes and their
Attributes attributes
Metaprocess Macroprocess Microprocess
Subject Line of business Project Iteration
Objectives Line-of-business profitability Project profitability Resource management
Competitiveness Risk management Risk resolution
Project budget, schedule, Milestone budget,
quality schedule, quality
Audience Acquisition authorities, Software project Subproject managers
customers managers Software engineers
Organizational Software
management engineers
Metrics Project predictability On budget, on schedule On budget, on schedule
Revenue, market share Major milestone success Major milestone progress
Project scrap and rework Release/iteration scrap
and rework
Concerns Bureaucracy vs. Quality vs. financial Content vs. schedule
standardization performance
 Improving Team Effectiveness (1)
 The principle of top talent: Use better and fewer people.

 The principle of job matching: Fit the task to the skills and motivation of
the people available.
 The principle of career progression: An organization does best in the
long
run by helping its people to self-actualize.
 The principle of team balance: Select people who will complement
and harmonize with one another.
 The principle of phase-out: Keeping a misfit on the team doesn’t
benefit
anyone.
 Improving Team Effectiveness (2)
Important Project Manager Skills on improving team effectiveness:
 Hiring skills. Few decisions are as important as hiring decisions. Placing the
right person in the right job seems obvious but is surprisingly hard to achieve.
 Customer-interface skill. Avoiding adversarial relationships among stake-
holders is a prerequisite for success.
 Decision-making skill. The Jillion books written about management have failed
to provide a clear definition of this attribute. We all know a good leader when we
run into one, and decision-making skill seems obvious despite its intangible
definition.
 Team-building skill. Teamwork requires that a manager establish trust,
motivate progress, exploit eccentric prima donnas, transition average people into
top performers, eliminate misfits, and consolidate diverse opinions into a team
direction.
 Selling skill. Successful project managers must sell all stakeholders (including
themselves) on decisions and priorities, sell candidates on job positions, sell
changes to the status quo in the face of resistance, and sell achievements
against objectives. In practice, selling requires continuous negotiation,
compromise, and empathy.
 Achieving Required Quality
Key practices that improve overall software quality:
 Focusing on driving requirements and critical use cases
early in the life cycle, focusing on requirements
completeness and traceability late in the life cycle, and
focusing throughout the life cycle on a balance between
requirements evolution, design evolution, and plan
evolution
 Using metrics and indicators to measure the progress
and quality of an architecture as it evolves from a high-
level prototype into a fully compliant product
Achieving Required Quality
Key practices that improve overall software
quality:
 Providing integrated life-cycle environments that
support early and continuous configuration control,
change management, rigorous design methods,
document automation, and regression test automation
 Using visual modeling and higher level language that
support architectural control, abstraction, reliable
programming, reuse, and self-documentation
 Early and continuous insight into performance issues
through demonstration-based evaluations
The Old Way and the New
 The Principles of Conventional Software
Engineering
1. Make quality #1.
2. High-quality software is possible.
3. Give products to customers early.
4. Determine the problem before writing the requirements.
5. Evaluate design alternatives.
6. Use an appropriate process model.
7. Use different languages for different phases.
8. Minimize intellectual distance.
9. Put techniques before tools.
10. Get it right before you make it faster.
 The Principles of Conventional Software
Engineering
11. Inspect code.
12. Good management is more important than good technology.
13. People are the key to success.
14. Follow with care.
15. Take responsibility.
16. Understand the customer’s priorities.
17. The more they see, the more they need.
18. Plan to throw one away .
19. Design for change.
20. Design without documentation is not design.
 The Principles of
Conventional Software
Engineering
21. Use tools, but be realistic.
22. Avoid tricks.
23. Encapsulate.
24. Use coupling and cohesion.
25. Use the McCabe complexity measure.
26. Don’t test your own software.
27. Analyze causes for errors.
28. Realize that software’s entropy increases.
29. People and time are not interchangeable.
30. Expert excellence.
The Principles of Modern Software Management

Architecture-first approach The central design element

Design and integration first, then production and test

Iterative life-cycle process The risk management element

Risk control through ever-increasing function, performance, quality
Component-based development The technology element
Object-oriented methods, rigorous notations, visual modeling

Change management environment The control element

Metrics, trends, process instrumentation
Round-trip engineering The automation element
Complementary tools, integrated environments
 Modern Project Profiles
Top 10 Software Management Principles

1. Base the process on an architecture-first approach – rework

rates remain stable over the project life cycle.
2. Establish an iterative life-cycle process that confronts
risk early
3. Transition design methods to emphasize component-based
development
4. Establish a change management environment – the
dynamics
of iterative development, including concurrent workflows by
different teams working on shared artifacts, necessitate highly
controlled baselines
5. Enhance change freedom through tools that support
round-trip engineering
 Modern Project Profiles
Top 10 Software Management Principles

6. Capture design artifacts in rigorous, model-based notation

7. Instrument the process for objective quality control and

progress assessment

8. Use a demonstration-based approach to asses intermediate

artifacts
9. Plan intermediate releases in groups of usage scenarios with
evolving levels of detail
10. Establish a configurable process that is economically
scalable
 Modern Project
Software
Profiles
Management Best Practices
 There is nine best practices:
1) Formal risk management
2) Agreement on interfaces
3) Formal inspections
4) Metric-based scheduling and management
5) Binary quality gates at the inch-pebble level
6) Program-wide visibility of progress versus plan
7) Defect tracking against quality targets
8) Configuration management
9) People-aware management accountability
 Next-Generation Software Economics
Next-Generation Cost Models

 Software experts hold widely varying opinions about software

economics and its manifestation in software cost estimation models:

source lines of code function points

productivity quality
measures VERSUS measures

Java C++

object-oriented functionally
oriented
 It will be difficult to improve empirical estimation models while
the project data going into these models are
noisy
and highly uncorrelated, and are based on differing process
and technology foundations.
 Next-Generation Software Economics
Next-Generation Cost Models
 Some of today’s popular software cost models are not well matched to an
iterative software process focused an architecture-first approach
 Many cost estimators are still using a conventional process experience base
to estimate a modern project profile
 A next-generation software cost model should explicitly separate architectural
engineering from application production, just as an architecture-first process
does.
 Two major improvements in next-generation software cost estimation
models:
 Separation of the engineering stage from the production stage will force
estimators to differentiate between architectural scale and implementation
size.
 Rigorous design notations such as UML will offer an opportunity to define units of
measure for scale that are more standardized and therefore can be automated and
tracked.
Thank you