UNIT 2

Software Project Management Renaissance

Conventional Software Management :

Conventional software management practices are sound in theory, but practice is still tied to
archaic (outdated) technology and techniques.
Conventional software economics provides a benchmark of performance for
conventional software manage- ment principles.
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, monitors, and control software development.
Three important analyses of the state of the software engineering industry are
1. Software development is still highly unpredictable. Only about 10% of
software projects are delivered successfully within initial budget and schedule
estimates.
2. Management discipline is more of a discriminator in success or failure than are
technology advances.
3. The level of software scrap and rework is indicative of an immature process.
All three analyses reached the same general conclusion: The success rate for software
projects is very low. The three analyses provide a good introduction to the magnitude
of the software problem and the current norms for conventional software management
performance.

THE WATERFALL MODEL

Most software engineering texts present the waterfall model as the source of the
"conventional" software process.
IN THEORY
It provides an insightful and concise summary of conventional
software management Three main primary points are
1. There are two essential steps common to the development of computer programs:
analysis and

coding.

Waterfall Model part 1: The two basic steps to building a program.

Analysis Analysis and coding both involve creative work that

directly contributes to the usefulness of the end
Coding

2. In order to manage and control all of the intellectual freedom associated with
software development, one must introduce several other "overhead" steps,
including system requirements definition, software requirements definition,
program design, and testing. These steps supplement the analysis and coding
steps. Below Figure illustrates the resulting project profile and the basic steps in
developing a large-scale program.
3. The basic framework described in the waterfall and invites failure. The testing phase that
occurs at the end of the development cycle is the first event for which timing, storage, input/output
transfers, etc., are experienced as distinguished from analyzed. The resulting design changes are
likely to be so disruptive that the software requirements upon which the design is based are likely
violated. Either the requirements must be modified or a substantial design change is warranted.

Five necessary improvements for waterfall model are:-

1. Program design comes first. Insert a preliminary program design phase

between the software requirements generation phase and the analysis phase. By
this technique, the program designer assures that the software will not fail
because of storage, timing, and data flux (continuous change). As analysis
proceeds in the succeeding phase, the program designer must impose on the
analyst the storage, timing, and operational constraints in such a way that he
senses the consequences. If the total resources to be applied are insufficient or if
the embryonic(in an early stage of development) operational design is wrong, it
will be recognized at this early stage and the iteration with requirements and
preliminary design can be redone before final design, coding, and test
commences. How is this program design procedure implemented?

The following steps are required:

Begin the design process with program designers, not analysts or programmers.
Design, define, and allocate the data processing modes even at the risk of
being wrong. Allocate processing functions, design the database, allocate
execution time, define interfaces and processing modes with the operating
system, describe input and output processing, and define preliminary
operating procedures.
Write an overview document that is understandable, informative, and current
so that every worker on the project can gain an elemental understanding of
the system.

2. Document the design. The amount of documentation required on most software

programs is quite a lot, certainly much more than most programmers, analysts, or
program designers are willing to do if left to their own devices. Why do we need so
 much documentation? (1) Each designer must communicate with interfacing designers,
managers, and possibly customers. (2) During early phases, the documentation is the
design. (3) The real monetary value of documentation is to support later modifications by
a separate test team, a separate maintenance team, and operations personnel who are not
software literate.

3. Do it twice. If a computer program is being developed for the first time, arrange
matters so that the version finally delivered to the customer for operational
deployment is actually the second version insofar as critical design/operations are
concerned. Note that this is simply the entire process done in miniature, to a time scale
that is relatively small with respect to the overall effort. In the first version, the team
must have a special broad competence where they can quickly sense trouble spots in
the design, model them, model alternatives, forget the straightforward aspects of the
design that aren't worth studying at this early point, and, finally, arrive at an error-free
program.

4. Plan, control, and monitor testing. Without question, the biggest user of project
resources-manpower, computer time, and/or management judgment-is the test phase.
This is the phase of greatest risk in terms of cost and schedule. It occurs at the latest
point in the schedule, when backup alternatives are least available, if at all. The
previous three recommendations were all aimed at uncovering and solving problems
before entering the test phase. However, even after doing these things, there is still a
test phase and there are still important things to be done, including: (1) employ a team
of test specialists who were not responsible for the
original design; (2) employ visual inspections to spot the obvious errors like dropped
minus signs, missing factors of two, jumps to wrong addresses (do not use the computer to
detect this kind of thing, it is too expensive); (3) test every logic path; (4) employ the final
checkout on the target computer.

5. Involve the customer. It is important to involve the customer in a formal way so that
he has committed himself at earlier points before final delivery. There are three points
following requirements definition where the insight, judgment, and commitment of the
customer can bolster the development effort. These include a "preliminary software
review" following the preliminary program design step, a sequence of "critical software
design reviews" during program design, and a "final software acceptance review".

IN PRACTICE
Some software projects still practice the conventional software management approach.
It is useful to summarize the characteristics of the conventional process as it has
typically been applied, which is not necessarily as it was intended. Projects destined for
trouble frequently exhibit the following symptoms:

Protracted integration and late design breakage.

Late risk resolution.
Requirements-driven functional decomposition.
Adversarial (conflict or opposition) stakeholder relationships.
Focus on documents and review meetings.
Protracted Integration and Late Design Breakage

For a typical development project that used a waterfall model management process, Figure
1-2 illustrates development progress versus time. Progress is defined as percent coded, that
is, demonstrable in its target form.

The following sequence was common:

Early success via paper designs and thorough (often too thorough) briefings.
Commitment to code late in the life cycle.
Integration nightmares (unpleasant experience) due to unforeseen
implementation issues and interface ambiguities.
Heavy budget and schedule pressure to get the system working.
Late shoe-homing of no optimal fixes, with no time for redesign.
A very fragile, unmentionable product delivered late.
In the conventional model, the entire system was designed on paper, then implemented all at once, then
integrated. Table 1-1
1 provides a typical profile of cost expenditures across the spectrum of software
activities.

Late risk resolution A serious issue associated with the waterfall lifecycle was the lack of early risk
resolution. Figure 1.3 illustrates a typical risk profile for conventional waterfall model projects. It includes
four distinct periods of risk exposure, where risk is defined as the probability of missing a cost, schedule,
feature, or quality goal. Early in the life cycle, as the requirements were being specified, the actual risk
exposure was highly unpredictable.
Requirements-Driven
Driven Functional Decomposition: This approach depends on specifying requirements
com- pletely and unambiguously before other development activities begin. It naively treats all
requirements as equally important, and depends on those requirements remaining constant over the software
development life cycle. These conditions rarely occur in the real world. Specification of requirements is a
difficult and important part of the software development process.
Another property of the conventional approach is that the requirements were typically specified in a
functional manner. Built into the classic waterfall process was the fundamental assumption that the software
itself was decomposed into functions; requirements were then allocated to the resulting components. This
decomposition was often very different from a decomposition based on object
object-oriented
oriented design and the use of
existing components. Figure 1-4 4 ill
illustrates the result of requirements-driven
driven approaches: a software
structure that is organized around the requirements specification structure.

3
Adversarial Stakeholder Relationships:

The conventional process tended to result in adversarial stakeholder relationships, in large part because of
the difficulties of requirements specification and the exchange of information solely through paper
documents that captured engineering information in ad hoc formats.

The following sequence of events was typical for most contractual software efforts:

1. The contractor prepared a draft contract-deliverable document that captured an intermediate

artifact and delivered it to the customer for approval.
2. The customer was expected to provide comments (typically within 15 to 30 days).
3. The contractor incorporated these comments and submitted (typically within 15 to 30 days) a final
version for approval.
This one-shot review process encouraged high levels of sensitivity on the part of customers and contractors.

Focus on Documents and Review Meetings:

The conventional process focused on producing various documents that attempted to describe the software
product, with insufficient focus on producing tangible increments of the products themselves. Contractors
were driven to produce literally tons of paper to meet milestones and demonstrate progress to stakeholders,
rather than spend their energy on tasks that would reduce risk and produce quality software. Typically,
presenters and the audience reviewed the simple things that they understood rather than the complex and
important issues. Most design reviews therefore resulted in low engineering value and high cost in terms of
the effort and schedule involved in their preparation and conduct. They presented merely a facade of
progress.
Table 1-2 summarizes the results of a typical design review.

3
 CONVENTIONAL SOFTWARE MANAGEMENT PERFORMANCE

Barry Boehm's "Industrial Software Metrics Top 10

software development.
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.
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;
15:85 in 1985, 85:15.
7. Only about 15% of software development effort is devoted to programming.
8. Software systems and products typically cost 3 times as much per SLOC as individual software
programs. Software-system products (i.e., system of systems) cost 9 times as much.
9. Walkthroughs catch 60% of the errors
10. 80% of the contribution comes from 20% of the contributors.

2. Evolution of Software Economics

SOFTWARE ECONOMICS
Most software cost models can be abstracted into a function of five basic parameters: size, process,
personnel, environment, and required quality.
1. The size of the end product (in human
human-generated
generated components), which is typically quantified in
terms of the number of source instructions or the number of function points required to develop
the required functionality
2. The process used to produce the end product, in particular the ability of the process to avoid non-
value-adding activities (rework, bureaucratic delays, communications overhead)
3. The capabilities of software engineering personnel,, and particularly their experience with the
computer science issues and the applications domain issues of the project
4. The environment,, which is made up of the tools and techniques available to support efficient
software development and to automate the process
5. The required quality of the product, including its features, performance, reliability, and

adaptability The relationships among these parameters and the estimated cost can be written as

follows:

Effort = (Personnel) (Environment) (Quality) ( Sizeprocess)

One important aspect of software economics (as represented within today's software cost models) is
that the relationship between effort and size exhibits a diseconomy of scal
scale.
e. The diseconomy of scale of
software development is a result of the process exponent being greater than 1.0. Contrary to most
3
manufacturing processes, the more software you build, the more expensive it is per unit item.
Figure 2-1 shows three generations of basic technology advancement in tools, components, and
processes. The required levels of quality and personnel are assumed to be constant. The ordinate of the
graph refers to software unit costs (pick your favorite: per SLOC, per function point, per component)
realized by an organization.
The three generations of software development are defined as follows:
1) Conventional: 1960s and 1970s, craftsmanship. Organizations used custom tools, custom
processes, and virtually all custom components built in primitive languages. Project performance
was highly predictable in that cost, schedule, and quality objectives were almost always
underachieved.
2) Transition: 1980s and 1990s, software engineering. Organiz:1tions used more-repeatable processes and
off- the-shelf tools, and mostly (>70%) custom components built in higher level languages. Some of
the

components (<30%) were available as commercial products, including the operating system, database
management system, networking, and graphical user interface.
3) Modern practices: 2000 and later, software production. This book's philosophy is rooted in
the use of managed and measured processes, integrated automation environments, and
mostly (70%) off-the-shelf components. Perhaps as few as 30% of the components need to
be custom built
Technologies for environment automation, size reduction, and process improvement are not
independent of one another. In each new era, the key is complementary growth in all technologies. For
example, the process advances could not be used successfully without new component technologies and
increased tool automation.

3
Organizations are achieving better economies of scale in successive technology eras
eras-with
with very large projects
(systems of systems), long-lived products, and lines of business comprising multiple similar projects. Figure
2-2 provides an overview of how a return on investment (ROI) profile can be achieved in subsequent efforts
across life cycles of various domains.

4
 PRAGMATIC SOFTWARE COST ESTIMATION
One critical problem in software cost estimation is a lack of well
well-documented
documented case studies of projects that
used an iterative development approach. Software industry has inconsistently defined metrics or atomic
units of measure, the data from actual projects are highly suspect in terms of consistency and comparability.
It is hard enough to collect a homogeneous set of project data within one organization; it is extremely
difficult to homog- enize data across different organizations with different processes, languages, domains,
and so on.
There have been many debates among developers and vendors of software cost estimation models and tools.
Three topics of these debates are of particular interest here:

1. Which cost estimation model to use?

2. Whether to measure software size in source lines of code or function points.
3. What constitutes a good estimate?
4
There are several popular cost estimation models (such as COCOMO, CHECKPOINT, ESTIMACS,
KnowledgePlan, Price-S,
S, ProQMS, SEER, SLIM, SOFTCOST, and SPQR/20), CO COMO is also one of
the most open and well-documented
documented cost estimation models. The general accuracy of conventional cost
models
(such as COCOMO) has been described as "within 20% of actuals, 70% of the time."
Most real-world
world use of cost models is bottom
bottom-upup (substantiating a target cost) rather than top-down
top
(estimating the "should" cost). Figure 22-33 illustrates the predominant practice: The software project manager
defines the target cost of the software,e, and then manipulates the parameters and sizing until the target cost
can be justified. The rationale for the target cost maybe to win a proposal, to solicit customer funding, to
attain internal corporate funding, or to achieve some other goal.
The process described in Figure 2-3 is not all bad. In fact, it is absolutely necessary to analyze the cost risks
and understand the sensitivities and trade-offs objectively. It forces the software project manager to
examine the risks associated with achieving the target costs and to discuss this information with other
stakeholders.
A good software cost 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, similar
technologies, similar environments, similar quality requirements, and similar people.
It is defined in enough detail so that its key risk areas are understood and the probability of success
is objectively assessed.
Extrapolating from a good estimate, an ideal estimate would be derived from a mature cost model with an
experience base that reflects multiple similar projects done by the same team with the same mature
processes and tools.

3. ImprovingSoftware Economics
Five basic parameters of the software cost model are

4
 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.

These parameters are given in priority order for most software domains. Table 3
3-1
1 lists some of the
technology developments, process improvement efforts, and management approaches targeted at
improving the economics of software development and integration.

REDUCING SOFTWARE PRODUCT SIZE

The most significant way to improve affordability and return on investment (ROI) is usually to produce a
product that achieves the design goals with the minimum amount of human-generated
generated source material.
Component-based development is introduced as the general term for reducing the "source
source" language size
to achieve a software solution.
Reuse, object-oriented
oriented technology, automatic code production, and higher order programming languages
are all focused on achieving a given system with fewer lines of human-specified
specified source directives
(statements).
size reduction is the primary motivation behind improvements in higher order languages (such as C++, Ada
95, Java, Visual Basic), automatic code generators (CASE tools, visual modeling tools, GUI builders), reuse
of commercial components (operating systems, windowing environments, database management systems,
middleware, networks), and object-oriented
oriented technol
technologies
ogies (Unified Modeling Language, visual modeling
tools, architecture frameworks).
The reduction is defined in terms of human
human-generated
generated source material. In general, when size-reducing
size
technologies are used, they reduce the number of human-generated source lines.

4
 LANGUAGES
Universal function points (UFPs1) are useful estimators for language-independent,
independent, early life-cycle
estimates. The basic units of function points are external user inputs, external outputs, internal logical data
groups, external
nal data interfaces, and external inquiries. SLOC metrics are useful estimators for software
after a candidate solution is formulated and an implementation language is known. Substantial data have
been documented relating SLOC to function points. Some of these results are shown in Table 3-2.

LANGUAGES SLOC per UFP

1 Function point metrics provide a standardized method for measuring the various functions of a software
application.

The basic units of function points are external user inputs, external outputs, internal logical data groups,
external data interfaces, and external inquiries.

Assembly 320
C 128
FORTAN77 105
COBOL85 91
Ada83 71
C++ 56
Ada95 55
Java 55
Visual Basic 35
Table 3-2

OBJECT-ORIENTED METHODS AND VISUAL MODELING

Object-oriented
oriented technology is not germane to most of the software management topics discussed here, and
books on object-oriented
oriented technology abound. Object
Object-oriented
oriented programming languages appear to benefit
both software productivity and software quality. The fundamental impact of object-oriented
oriented technology is
in reducing the overall size of what needs to be developed.
People like drawing pictures to explain something to others or to themselves. When they do it for
software system design, they call these pictu pictures
res diagrams or diagrammatic models and the very
notation for them a modeling language.
These are interesting examples of the interrelationships among the dimensions of improving software eco-
nomics.

1. An object-oriented
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.
2. The use of continuous integration creates opportunities to recognize risk early and make incremental

4
 corrections without destabilizing the entire development effort.
3. 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.

Booch also summarized five characteristics of a successful object-oriented project.

1. A ruthless focus on the development of a system that provides a well understood collection of essential
minimal characteristics.
2. The existence of a culture that is centered on results, encourages communication, and yet is not
afraid to fail.
3. The effective use of object-oriented
oriented modeling.
4. The existence of a strong architectural vision.
5. The application of a well-managed
managed iterative and incremental development life cycle.

REUSE
Reusing existing components and building reusable components have been natural software engineering
activities since the earliest improvements in programming languages. With reuse in order to minimize
development costs while achieving all the other required attributes of performance, feature set, and quality.
Try to treat reuse as a mundane part of achieving a return on investment.
Most truly reusable components of value are transitioned to commercial products supported by
organizations with the following characteristics:

They have an economic motivation for continued support.

They take ownership of improving product quality, adding new features, and transitioning to new
technologies.
They have a sufficiently broad customer base to be profitable.
The cost of developing a reusable component is not trivial. Figure 3-1 examines the economic trade-offs.
The steep initial curve illustrates the economic obstacle to developing reusable components.
Reuse is an important discipline that has an impact on the efficiency of all workflows and the quality of most
artifacts.

COMMERCIAL COMPONENTS
A common approach being pursued today in many domains is to maximize integration of commercial
components and off-the-shelf products. While the use of commercial components is certainly desirable as a
4
means of reducing custom development, it has not proven to be straightforward in practice. Table 3-3
3
identifies some of the advantages and disadvantages of using commercial components.

IMPROVING SOFTWARE PROCESSES

Process is an overloaded term. Three distinct process perspectives are.
Metaprocess: an organization's policies, procedures, and practices for pursuing a software-
intensive line of business. The focus of this process is on organizational economics, long-term
strategies, and software ROI.
Macroprocess: a project's policies, procedures, and practices for producing a complete software
product within certain cost, schedule, and quality constraints. The focus of the macro process is on
creating an adequate instance of the Meta process for a specific set of constraints.
Microprocess: a project team's policies, procedures, and practices for achieving an artifact of the
software process. The focus of the micro process is on achieving an intermediate product baseline
with adequate quality and adequate functionality as economically and rapidly as practical.
Although these three levels of process overlap somewhat, they h have
ave different objectives, audiences,
metrics, concerns, and time scales as shown in Table 3-4

In a perfect software engineering world with an immaculate problem description, an obvious solution

4
space, a development team of experienced geniuses, adequate resources, and stakeholders with common
goals, we could execute a software development process in one iteration with almost no scrap and rework.
Because we work in an imperfect world, however, we need to manage engineering activities so that scrap
and rework profiles do not have an impact on the win conditions of any stakeholder. This should be the
underlying premise for most process improvements.

IMPROVING TEAM EFFECTIVENESS

Teamwork is much more important than the sum of the individuals. With software teams, a project manager
needs to configure a balance of solid talent with highly skilled people in the leverage positions. Some
maxims of team management include the following:
A well-managed project can succeed with a nominal engineering team.
A mismanaged project will almost never succeed, even with an expert team of engineers.
A well-architected system can be built by a nominal team of software builders.
A poorly architected system will flounder even with an expert team of builders.

Boehm five staffing principles are

1. The principle of top talent: Use better and fewer people

2. The principle of job matching: Fit the tasks to the skills and motivation of the people available.
3. The principle of career progression: An organization does best in the long run by helping its
people to self-actualize.
4. The principle of team balance: Select people who will complement and harmonize with one another
5. The principle of phase-out: Keeping a misfit on the team doesn't benefit anyone

Software project managers need many leadership qualities in order to enhance team effectiveness. The
following are some crucial attributes of successful software project managers that deserve much more
attention:

1. 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.
2. Customer-interface skill. Avoiding adversarial relationships among stakeholders 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

IMPROVING AUTOMATION THROUGH SOFTWARE ENVIRONMENTS

The tools and environment used in the software process generally have a linear effect on the
4
productivity of the process. Planning tools, requirements management tools, visual modeling tools,
compilers, editors, debuggers, quality assurance analysis tools, test tools, and user interfaces provide crucial
automation support for evolving the software engineering artifacts. Above all, configuration management
managemen
environments provide the foundation for executing and instrument the process. At first order, the isolated
impact of tools and automation generally allows improvements of 20% to 40% in effort. However, tools and
environments must be viewed as the primary delivery vehicle for process automation and improvement, so
their impact can be much higher.
Automation of the design process provides payback in quality, the ability to estimate costs and
schedules, and overall productivity using a smaller team.
Round-trip engineering describe the key capability of environments that support iterative development. As
we have moved into maintaining different information repositories for the engineering artifacts, we need
automation support to ensure efficient and ererror-free
free transition of data from one artifact to another. Forward
engineering is the automation of one engineering artifact from another, more abstract representation. For
example, compilers and linkers have provided automated transition of source code into executable code.
Reverse engineering is the generation or modification of a more abstract representation from an existing
artifact (for example, creating a .visual design model from a source code representation).
Economic improvements associated with tools and environments. It is common for tool vendors to make
rela- tively accurate individual assessments of lifelife-cycle
cycle activities to support claims about the potential
economic impact of their tools. For example, it is easy to find statements such as the following from
companies in a particular tool.
Requirements analysis and evolution activities consume 40% of life-cycle costs.
Software design activities have an impact on more than 50% of the resources.
Coding and unit testing activities consume about 50% of software development effort and schedule.
Test activities can consume as much as 50% of a project's resources.
Configuration control and change management are critical activities that can consume as much as
25% of resources on a large-scale
scale project.
Documentation activities can consume more than 30% of project engineering resources.
Project management, business administration, and progress assessment can consume as much as
30% of project budgets.

ACHIEVING REQUIRED QUALITY

Software best practices are derived from the development process and technologies. Table 3-5 summarizes
some dimensions of quality improvement.
Key practices that improve overall software quality include the following:
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
Providing integrated life-cycle
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 languages that support architectural control, abstraction,
reliable programming, reuse, and self
self-documentation
Early and continuous insight into performance issues through demonstration-based
based evaluations

4
Conventional development processes stressed early sizing and timing estimates of computer program
resource utilization. However, the typical chronology of events in performance assessment was as
follows
Project inception. The proposed design was asserted to be low risk with adequate performance
margin.
Initial design review. Optimistic assessments of adequate design margin were based mostly on
paper analysis or rough simulation of the critical threads. In most cases, the actual application
algorithms and database sizes were fairly well understood.
Mid-life-cycle design review. The assessments started whittling away at the margin, as early
benchmarks and initial tests began exposing the optimism inherent in earlier estimates.
Integration and test. Serious performance problems were uncovered, necessitating fundamental
changes in the architecture. The underlying infrastructure was usually the scapegoat, but the real
culprit was immature use of the infrastructure, immature architectural solutions, or poorly
understood early design trade-offs.

PEER INSPECTIONS: A PRAGMATIC VIEW

Peer inspections are frequently over hyped as the key aspect of a quality system. In my experience, peer
reviews are valuable as secondary mechanisms, but they are rarely significant contributors to quality
compared with the following primary quality mechanisms and indicators, which should be emphasized in
the management process:

Transitioning engineering information from one artifact set to another, thereby assessing the
consistency, feasibility, understandability, and technology constraints inherent in the engineering
artifacts
Major milestone demonstrations that force the artifacts to be assessed against tangible criteria in
4
 the context of relevant use cases
Environment tools (compilers, debuggers, analyzers, automated test suites) that ensure
representation rigor, consistency, completeness, and change control
Life-cycle testing for detailed insight into critical trade-offs, acceptance criteria, and requirements
compliance
Change management metrics for objective insight into multiple-perspective change trends and
convergence or divergence from quality and progress goals
Inspections are also a good vehicle for holding authors accountable for quality products. All authors of
software and documentation should have their products scrutinized as a natural by by--product of the
process. Therefore, the coverage of inspections should be across all authors rather than across
acros all
components.

THE OLD WAY AND THE NEW

THE PRINCIPLES OF CONVENTIONAL SOFTWARE ENGINEERING

1.Make quality #1. Quality must be quantified and mechanisms put into place to motivate its achievement
2.High-quality software is possible. Techniques that have been demonstrated to increase quality include
involving the customer, prototyping, simplifying design, conducting inspections, and hiring the best people
3.Give products to customers earlyearly.. No matter how hard you try to learn users' needs during the
requirements phase, the most effective way to determine real needs is to give users a product and let them
play with it 4.Determine
Determine the problem before writing the requirements
requirements.. When faced with what they
believe is a problem, most engineers rush to offer a solution. Before you try to solve a problem, be sure to
explore all the alternatives and don't be blinded by the obvious solution
5. Evaluate design alternatives.. After the requirements are agreed upon, you must examine a variety of
architectures and algorithms. You certainly do not want to
the requirements specification.
6. Use an appropriate process model
model.. Each project must select a process that makes ·the most sense for that
project on the basis of corporate culture, willingness to take risks, application ation area, volatility of
requirements, and the extent to which requirements are well understood.
7. Use different languages for different phases phases.. Our industry's eternal thirst for simple solutions to
complex problems has driven many to declare that the best development method is one that uses the same
notation throughout the life cycle.
8. Minimize intellectual distance.. To minimize intellectual distance, the software's structure should be as
close as possible to the real-world structure
9. Put techniques before tools.. An undisciplined software engineer with a tool becomes a dangerous,
undisciplined software engineer
10. Get it right before you make it faster
faster. It is far easier to make a working program run faster than it is to
make a fast program work. Don't worry about optimization during initial coding
11. Inspect code. Inspecting the detailed design and code is a much better way to find errors than testing
12.Good management is more important than good technology. Good management motivates people to
do their best, but there are no universal "right" styles of management.
13. People are the key to success. Highly skilled people with appropriate experience, talent, and training are
key.
14. Follow with care.. Just because everybody is doing something does not make it right for you. It may be
right, but you must carefully assess its applicability to your environment.
15. Take responsibility.. When a bridge collapses we ask, "What did the engineers do wrong?" Even when
software fails, we rarely ask this. The fact is that in any engineering discipline, the best methods can be used
to produce awful designs, and the most antiquated methods to produce elegant designs.
16. Understand the customer's priorities
priorities. It is possible the customer would tolerate 90% of the
5
functionality delivered late if they could have 10% of it on time.
17. The more they see, the more they need. The more functionality (or performance) you provide a user,
the more functionality (or performance) the user wants.
18. Plan to throw one away. One of the most important critical success factors is whether or not a product is
entirely new. Such brand-new applications, architectures, interfaces, or algorithms rarely work the first
time.
19. Design for change. The architectures, components, and specification techniques you use must
accommodate change.
20. Design without documentation is not design. I have often heard software engineers say, "I have
finished the design. All that is left is the documentation. "
21. Use tools, but be realistic. Software tools make their users more efficient.
22. Avoid tricks. Many programmers love to create programs with tricks constructs that perform a function
correctly, but in an obscure way. Show the world how smart you are by avoiding tricky code
23. Encapsulate. Information-hiding is a simple, proven concept that results in software that is easier
to test and much easier to maintain.
24. Use coupling and cohesion. Coupling and cohesion are the best ways to measure software's inherent
maintainability and adaptability
25. Use the McCabe complexity measure. Although there are many metrics available to report the inherent
complexity of software, none is as intuitive and easy to use as Tom McCabe's
26. Don't test your own software. Software developers should never be the primary testers of their
own software.
27. Analyze causes for errors. It is far more cost-effective to reduce the effect of an error by preventing it
than it is to find and fix it. One way to do this is to analyze the causes of errors as they are detected
28.Realize that software's entropy increases. Any software system that undergoes continuous change will
grow in complexity and will become more and more disorganized
29. People and time are not interchangeable. Measuring a project solely by person-months makes little
sense 30.Expect excellence. Your employees will do much better if you have high expectations for them.

THE PRINCIPLES OF MODERN SOFTWARE MANAGEMENT

Top 10 principles of modern software management are. (The first five, which are the main themes of
my definition of an iterative process, are summarized in Figure 4-1.)

Base the process on an architecture-first approach. This requires that a demonstrable balance be
achieved among the driving requirements, the architecturally significant design decisions, and the
life- cycle plans before the resources are committed for full-scale development.
Establish an iterative life-cycle process that confronts risk early. With today's sophisticated
software systems, it is not possible to define the entire problem, design the entire solution, build the
software, and then test the end product in sequence. Instead, an iterative process that refines the
problem understanding, an effective solution, and an effective plan over several iterations encourages
a balanced treatment of all stakeholder objectives. Major risks must be addressed early to increase
predictability and avoid expensive downstream scrap and rework.
Transition design methods to emphasize component-based development. Moving from a line-of-
code mentality to a component-based mentality is necessary to reduce the amount of human-
generated source code and custom development.

4. Establish a change management environment. The dynamics of iterative development,

including concurrent workflows by different teams working on shared artifacts, necessitates
5
objectively controlled baselines.

5. Enhance change freedom through tools that support round-trip engineering.

engineering Round-
trip engineering is the environment support necessary to automate and synchronize
engineering information in different formats(such as requirements specifications, design
models, source code, executable code, test cases).
6. Capture design artifacts in rigorous, model-based notation. A model based approach (such as
UML) supports the evolution of semantically rich graphical and textual design notations.
7. Instrument the process for objective quality control and progress assessment.
assessment Life-cycle
assessment of the progress and the quality of all intermediate products must be integrated into the process.
8. Use a demonstration-based approach to assess intermediate artifacts.
9. Plan intermediate releases in groups of usage scenarios with evolving levels of detail. It is
essential that the software management process drive toward early and continuous
demonstrations within the operational context of the system, namely its use cases.
10. Establish a configurable process that is economically scalable. No single process is
suitable for all software developments.

Table 4-1 maps top 10 risks of the conventional process to the key attributes and principles of a modern
process

5
 TRANSITIONING TO AN ITERATIVE PROCESS

Modern software development processes have moved away from the conventional waterfall model, in
which each stage of the development process is dependent on completion of the previous stage.
The economic benefits inherent in transitioning from the conventional waterfall model to an iterative
development process are significant but difficult to quantify. As one benchmark of the expected economic
impact of process improvement, consider the process exponent parameters of the COCOMO II model.
(Appendix B provides more detail on the COCOMO model) This exponent can range from 1.01 (virtually
no diseconomy of scale) to 1.26 (significant diseconomy of scale). The parameters that govern the value
of the process exponent are application precedentedness, process flexibility, architecture risk resolution,
team cohesion, and software process maturity.
The following paragraphs map the process exponent parameters of CO COMO II to my top 10
principles of a modern process.

Application precedentedness. Domain experience is a critical factor in understanding how to plan

and execute a software development project. For unprecedented systems, one of the key goals is to
confront risks and establish early precedents, even if they are incomplete or experimental. This is one
of the primary reasons that the software industry has moved to an iterative life-cycle process. Early
iterations in the life cycle establish precedents from which the product, the process, and the plans can
be elab- orated in evolving levels of detail.
Process flexibility. Development of modern software is characterized by such a broad solution
space and so many interrelated concerns that there is a paramount need for continuous incorporation
of changes. These changes may be inherent in the problem understanding, the solution space, or the
plans. Project artifacts must be supported by efficient change management commensurate with
project needs. A configurable process that allows a common framework to be adapted across a
range of projects is necessary to achieve a software return on investment.
Architecture risk resolution. Architecture-first development is a crucial theme underlying a
5
 successful iterative development process. A project team develops and stabilizes architecture before
developing all the components that make up the entire suite of applications components. An architecture-
first and component-based development approach forces the infrastructure, common mechanisms, and
control mechanisms to be elaborated early in the life cycle and drives all component make/buy decisions into
the architecture process.
Team cohesion. Successful teams are cohesive, and cohesive teams are successful. Successful teams
and cohesive teams share common objectives and priorities. Advances in technology (such as
programming languages, UML, and visual modeling) have enabled more rigorous and
understandable notations for communicating software engineering information, particularly in the
requirements and design artifacts that previously were ad hoc and based completely on paper
exchange. These model- based formats have also enabled the round-trip engineering support
needed to establish change freedom sufficient for evolving design representations.
Software process maturity. The Software Engineering Institute's Capability Maturity Model
(CMM) is a well-accepted benchmark for software process assessment. One of key themes is that
truly mature processes are enabled through an integrated environment that provides the appropriate
level of automa- tion to instrument the process for objective quality control.
Life-Cycle Phases and Process artifacts

Life cycle phases

Characteristic of a successful software development process is the well-defined separation between

"research and development" activities and "production" activities. Most unsuccessful projects exhibit one
of the following characteristics:
An overemphasis on research and development
An overemphasis on production.
Successful modern projects-and even successful projects developed under the conventional process-tend
to have a very well-defined project milestone when there is a noticeable transition from a research attitude
to a production attitude. Earlier phases focus on achieving functionality. Later phases revolve around
achieving a product that can be shipped to a customer, with explicit attention to robustness, performance,
and finish.
A modern software development process must be defined to support the following:
Evolution of the plans, requirements, and architecture, together with well defined
synchronization points
Risk management and objective measures of progress and quality
Evolution of system capabilities through demonstrations of increasing functionality

5
ENGINEERING AND PRODUCTION STAGES

To achieve economies of scale and higher returns on investment, we must move toward a software
manufacturing process driven by technological improvements in process automation and component-
component
based development. Two stages of the life cycle are:

1. The engineering stage, driven by less predictable but smaller teams doing design and
synthesis activities
2. The production stage, driven by more predictable but larger teams doing construction, test,
and deployment activities

The transition between engineering and production is a crucial event for the various stakeholders.
The production plan has been agreed upon, and there is a good enough understanding of the problem and
the solution that all stakeholders can make a firm commitment to go ahead with production.
Engineering stage is decomposed into two distinct phases, inception and elaboration, and the production
stage into construction and transition. These four phases of the life
life-cycle
cycle process are loosely mapped to
the conceptual framework of the spiral model as shown in Figure 5-1

INCEPTION PHASE
The overriding goal of the inception phase is to achieve concurrence among stakeholders on the life-
cycle objectives for the project.

5
PRIMARY OBJECTIVES

Establishing the project's software scope and boundary conditions, including an operational
concept, acceptance criteria, and a clear understanding of what is and is not intended to be in the
product
Discriminating the critical use cases of the system and the primary scenarios of operation that
will drive the major design trade-offs
Demonstrating at least one candidate architecture against some of the primary scenanos
Estimating the cost and schedule for the entire project (including detailed estimates for
the elaboration phase)
Estimating potential risks (sources of unpredictability)

ESSENTIAL ACTMTIES
Formulating the scope of the project. The information repository should be sufficient to define
the problem space and derive the acceptance criteria for the end product.
Synthesizing the architecture. An information repository is created that is sufficient to
demonstrate the feasibility of at least one candidate architecture and an, initial baseline of
make/buy decisions so that the cost, schedule, and resource estimates can be derived.
Planning and preparing a business case. Alternatives for risk management, staffing, iteration
plans, and cost/schedule/profitability trade-offs are evaluated.
PRIMARY EVALUATION CRITERIA
Do all stakeholders concur on the scope definition and cost and schedule estimates?
Are requirements understood, as evidenced by the fidelity of the critical use cases?
Are the cost and schedule estimates, priorities, risks, and development processes credible?
Do the depth and breadth of an architecture prototype demonstrate the preceding criteria? (The
primary value of prototyping candidate architecture is to provide a vehicle for understanding the
scope and assessing the credibility of the development group in solving the particular technical
problem.)
Are actual resource expenditures versus planned expenditures acceptable

ELABORATION PHASE

At the end of this phase, the "engineering" is considered complete. The elaboration phase activities must
ensure that the architecture, requirements, and plans are stable enough, and the risks sufficiently
mitigated, that the cost and schedule for the completion of the development can be predicted within an
acceptable range. During the elaboration phase, an executable architecture prototype is built in one or
more iterations, depending on the scope, size, & risk.
PRIMARY OBJECTIVES
Baselining the architecture as rapidly as practical (establishing a configuration-managed snapshot in
which all changes are rationalized, tracked, and maintained)
Baselining the vision
Baselining a high-fidelity plan for the construction phase
Demonstrating that the baseline architecture will support the vision at a reasonable cost in a
reasonable time
5
ESSENTIAL ACTIVITIES
Elaborating the vision.
Elaborating the process and infrastructure.
Elaborating the architecture and selecting components.

PRIMARY EVALUATION CRITERIA

Is the vision stable?
Is the architecture stable?
Does the executable demonstration show that the major risk elements have been addressed and
credibly resolved?
Is the construction phase plan of sufficient fidelity, and is it backed up with a credible basis of
estimate?
Do all stakeholders agree that the current vision can be met if the current plan is executed to develop
the complete system in the context of the current architecture?
Are actual resource expenditures versus planned expenditures acceptable?

CONSTRUCTION PHASE
During the construction phase, all remaining components and application features are integrated
into the application, and all features are thoroughly tested. Newly developed software is integrated
where required. The construction phase represents a production process, in which emphasis is placed
on managing resources and controlling operations to optimize costs, schedules, and quality.

PRIMARY OBJECTIVES
Minimizing development costs by optimizing resources and avoiding unnecessary scrap and
rework
Achieving adequate quality as rapidly as practical
Achieving useful versions (alpha, beta, and other test releases) as rapidly as practical

ESSENTIAL ACTIVITIES
Resource management, control, and process optimization
Complete component development and testing against evaluation criteria
Assessment of product releases against acceptance criteria of the vision

PRIMARY EVALUATION CRITERIA

Is this product baseline mature enough to be deployed in the user community? (Existing defects
are not obstacles to achieving the purpose of the next release.)
Is this product baseline stable enough to be deployed in the user community? (Pending changes
are not obstacles to achieving the purpose of the next release.)
Are the stakeholders ready for transition to the user community?
Are actual resource expenditures versus planned expenditures acceptable?

5
 TRANSITION PHASE
The transition phase is entered when a baseline is mature enough to be deployed in the end-user domain.
This typically requires that a usable subset of the system has been achieved with acceptable quality
levels and user documentation so that transition to the user will provide positive results. This phase
could include any of the following activities:

1. Beta testing to validate the new system against user expectations

2. Beta testing and parallel operation relative to a legacy system it is replacing
3. Conversion of operational databases
4. Training of users and maintainers
The transition phase concludes when the deployment baseline has achieved the complete vision.

PRIMARY OBJECTIVES
Achieving user self-supportability
supportability
Achieving stakeholder concurrence that deployment baselines are complete and consistent with
the evaluation criteria of the vision
Achieving final product baselines as rapidly and cost-effectively as practical

ESSENTIAL ACTIVITIES
Synchronization and integration of concurrent construction increments into consistent
deployment baselines
Deployment-specific engineering (cutover, commercial packaging and production, sales
rollout kit development, field personnel training)
Assessment of deployment baselines against the complete vision and acceptance
criteria in the requirements set

EVALUATION CRITERIA
Is the user satisfied?
Are actual resource expenditures versus planned expenditures acceptable?

Artifacts of the process

THE ARTIFACT SETS

To make the development of a complete software system manageable, distinct collections of information
are organized into artifact sets. Artifact
rtifact represents cohesive information that typically is developed and
reviewed as a single entity.
Life-cycle
cycle software artifacts are organized into five distinct sets that are roughly partitioned by the
underlying language of the set: management (ad hoc textual formats), requirements (organized text and
models of the problem space), design (models of the solution space), implementation (human-readable
(human
programming language
anguage and associated source files), and deployment (machine
(machine-process
process able languages and
associated files). The artifact sets are shown in Figure 6-1.

5
THE MANAGEMENT SET
The management set captures the artifacts associated with process planning and execution. These
artifacts use ad hoc notations, including text, graphics, or whatever representation is required to
capture the "contracts" among project personnel (project management, architects, developers, testers,
marketers, administrators), among stakeholders (funding authority, user, software project manager,
organization manager, regulatory agency), and between project personnel and stakeholders. Specific
artifacts included in this set are the work breakdown stru structure
cture (activity breakdown and financial
tracking mechanism), the business case (cost, schedule, profit expectations), the release
specifications (scope, plan, objectives for release baselines), the software development plan (project
process instance), the release descriptions (results of release baselines), the status assessments
(periodic snapshots of project progress), the software change orders (descriptions of discrete baseline
changes), the deployment docu-- ments (cutover plan, training course, sales rollout kit), and the
environment (hardware and software tools, process automation, & documentation).
Management set artifacts are evaluated, assessed, and measured through a combination of the following:
Relevant stakeholder review
Analysis of changes between the current version of the artifact and previous versions
Major milestone demonstrations of the balance among all artifacts and, in particular, the
accuracy of the business case and vision artifacts

THE ENGINEERING SETS

The engineering sets consist of the requirements set, the design set, the implementation set, and
the deployment set.
Requirements Set

Requirements artifacts are evaluated, assessed, and measured through a combination of the following:
Analysis of consistency with the release specifications of the management set
Analysis of consistency between the vision and the requirements models
Mapping against the design, implementation, and deployment sets to evaluate the consistency
and completeness and the semantic balance between information in the different sets
Analysis of changes between the current version of requirements artifacts and previous
versions (scrap, rework, and defect elimination trends)
Subjective review of other dimensions of quality

Design Set

UML notation is used to engineer the design models for the solution. The design set contains
varying levels of abstraction that represent the components of the solution space (their identities,
attributes, static relationships, dynamic interactions). The design set is evaluated, assessed, and
measured through a combination of the following:
Analysis of the internal consistency and quality of the design model
Analysis of consistency with the requirements models
Translation into implementation and deployment sets and notations (for example, traceability,
source code generation, compilation, linking) to evaluate the consistency and completeness and
the semantic balance between information in the sets
Analysis of changes between the current version of the design model and previous versions
(scrap, rework, and defect elimination trends)
Subjective review of other dimensions of quality

Implementation set

The implementation set includes source code (programming language notations) that represents the tangible
implementations of components (their form, interface, and dependency relationships)
Implementation sets are human-readable formats that are evaluated, assessed, and measured
through a combination of the following:
Analysis of consistency with the design models
Translation into deployment set notations (for example, compilation and linking) to evaluate the
consistency and completeness among artifact sets
Assessment of component source or executable files against relevant evaluation criteria through
inspection, analysis, demonstration, or testing
Execution of stand-alone component test cases that automatically compare expected results with
actual results
Analysis of changes between the current version of the implementation set and previous
versions (scrap, rework, and defect elimination trends)
Subjective review of other dimensions of quality
Deployment Set

The deployment set includes user deliverables and machine language notations, executable software, and
the
build scripts, installation scripts, and executable target specific data necessary to use the product in its
target environment.
Deployment sets are evaluated, assessed, and measured through a combination of the following:

Testing against the usage scenarios and quality attributes defined in the requirements set to
evaluate the consistency and completeness and the~ semantic balance between information in
the two sets
Testing the partitioning, replication, and allocation strategies in mapping components of the
implementation set to physical resources of the deployment system (platform type, number,
network topology)
Testing against the defined usage scenarios in the user manual such as installation, user-
orienteddynamic reconfiguration, mainstream usage, and anomaly management
Analysis of changes between the current version of the deployment set and previous versions
(defect elimination trends, performance changes)
Subjective review of other dimensions of quality
Each artifact set is the predominant development focus of one phase of the life cycle; the other sets take
on check and balance roles. As illustrated in Figure 6-2, each phase has a predominant focus:
Requirements are the focus of the inception phase; design, the elaboration phase; implementation, the
construction phase; and deployment, the transition phase. The management artifacts also evolve, but at a
fairly constant level across the life cycle.
Most of today's software development tools map closely to one of the five artifact sets.
1. Management: scheduling, workflow, defect tracking, change
management, documentation, spreadsheet, resource management, and
presentation tools
2. Requirements: requirements management tools
3. Design: visual modeling tools
4. Implementation: compiler/debugger tools, code analysis tools, test coverage analysis tools, and
test management tools
5. Deployment: test coverage and test automation tools, network management tools, commercial
components (operating systems, GUIs, RDBMS, networks, middleware), and installation tools.
Implementation Set versus Deployment Set

The separation of the implementation set (source code) from the deployment set (executable code) is
important because there are very different concerns with each set. The structure of the information
delivered to the user (and typically the test organization) is very different from the structure of the source
code information. Engineering decisions that have an impact on the quality of the deployment set but are
relatively incomprehensible in the design and implementation sets include the following:

Dynamically reconfigurable parameters (buffer sizes, color palettes, number of servers, number
of simultaneous clients, data files, run
run-time parameters)
Effects of compiler/link optimizations (such as space optimization versus speed optimization)
Performance under certain allocation strategies (centralized versus distributed, primary and
shadow threads, dynamic load balancing, hot backup versus checkpoint/rollback)Virtual
machine constraints (file descriptors, garbage collection, heap size, maximum record size, disk
file rotations)
Process-level
level concurrency issues (deadlock and race conditions)
Platform-specific differences in performance or behavior

ARTIFACT EVOLUTION OVER THE LIFE CYCLE

Each state of development represents a certain amount of precision in the final system description.
Early in the life cycle, precision is low and the representation is generally high. Eventually, the precision
of representation is high and everything is specified in full detail. Each phase of development focuses on a
particular artifact set. At the end of each phase, the overall system state will have progressed on all sets, as
illustrated in Figure 6-3.
The inception phase focuses mainly on critical requirements usually with a secondary focus on an
initial deployment view. During the elaboration phase, there is much greater depth in requirements,
much more breadth in the design set, and further work on implementation and deployment issues. The
main focus of the construction phase is design and implementation. The main focus of the transition
phase is on achieving consistency and completeness of the deployment set in the context of the other
sets.

TEST ARTIFACTS
The test artifacts must be developed concurrently with the product from inception through
deployment. Thus, testing is a full-life-cycle activity, not a late life-cycle activity.
The test artifacts are communicated, engineered, and developed within the same artifact sets as
the developed product.
The test artifacts are implemented in programmable and repeatable formats (as
software programs).
The test artifacts are documented in the same way that the product is documented.
Developers of the test artifacts use the same tools, techniques, and training as the
software engineers developing the product.
Test artifact subsets are highly project
project-specific, the following example clarifies the relationship between
test artifacts and the other artifact sets. Consider a project to perform seismic data processing for the
purpose of oil exploration. This system has three fundamental subsystems: (1) a sensor subsystem that
captures raw seismic data in real time and delivers these data to (2) a technical operations subsystem
that converts raw data into an organized database and manages queries to this database from (3) a
display subsystem that allows workstation operators to examine seismic data in human-readable
readable form.
Such a system would result in the following test artifacts:

Management set. The release specifications and release descriptions capture the objectives,
evaluation criteria, and results of an intermediate milestone. These artifacts are the test plans
 and test results negotiated among internal project teams. The software change orders capture
test results (defects, testability changes, requirements ambiguities, enhancements) and the
closure criteria associated with making a discrete change to a baseline.
Requirements set. The system-level use cases capture the operational concept for the system
and the acceptance test case descriptions, including the expected behavior of the system and its
quality attributes. The entire requirement set is a test artifact because it is the basis of all
assessment activities across the life cycle.
Design set. A test model for nondeliverable components needed to test the product baselines is
captured in the design set. These components include such design set artifacts as a seismic event
simulation for creating realistic sensor data; a "virtual operator" that can support unattended,
after- hours test cases; specific instrumentation suites for early demonstration of resource usage;
transaction rates or response times; and use case test drivers and component stand-alone test
drivers.
Implementation set. Self-documenting source code representations for test components and test
drivers provide the equivalent of test procedures and test scripts. These source files may also
include human-readable data files representing certain statically defined data sets that are
explicit test source files. Output files from test drivers provide the equivalent of test reports.
Deployment set. Executable versions of test components, test drivers, and data files are provided.

MANAGEMENT ARTIFACTS
The management set includes several artifacts that capture intermediate results and ancillary
information necessary to document the product/process legacy, maintain the product, improve the
product, and improve the process.

Business Case

The business case artifact provides all the information necessary to determine whether the project is
worth investing in. It details the expected revenue, expected cost, technical and management plans,
and backup data necessary to demonstrate the risks and realism of the plans. The main purpose is to
transform the vision into economic terms so that an organization can make an accurate ROI
assessment. The financial forecasts are evolutionary, updated with more accurate forecasts as the life
cycle progresses. Figure 6-4 provides a default outline for a business case.

Software Development Plan

The software development plan (SDP) elaborates the process framework into a fully detailed plan.
Two indications of a useful SDP are periodic updating (it is not stagnant shelfware) and
understanding and acceptance by managers and practitioners alike. Figure 6-5 provides a default
outline for a software development plan.
Work Breakdown Structure

Work breakdown structure (WBS) is the vehicle for budgeting and collecting costs. To monitor and
control a project's financial performance, the software project man1ger must have insight into project
 costs and how they are expended. The structure of cost accountability is a serious project planning
constraint.

Software Change Order Database

Managing change is one of the fundamental primitives of an iterative development process. With greater
change freedom, a project can iterate more productively. This flexibility increases the content, quality,
and number of iterations that a project can achieve within a given schedule. Change freedom has been
achieved in practice through automation, and today's iterative development environments carry the
burden of change management. Organizational processes that depend on manual change management
techniques have encountered major inefficiencies.

Release Specifications

The scope, plan, and objective evaluation criteria for each baseline release are derived from the vision
statement as well as many other sources (make/buy analyses, risk management concerns, architectural
considerations, shots in the dark, implementation constraints, quality thresholds). These artifacts are
intended
d to evolve along with the process, achieving greater fidelity as the life cycle progresses and
requirements understanding matures. Figure 6-6 provides a default outline for a release specification

Release Descriptions

Release description documents describe the results of each release, including performance against each
of the evaluation criteria in the corresponding release specification. Release baselines should be
accompanied by a release description document that describes the evaluation criteria for that
configuration baseline and provides substantiation (through demonstration, testing, inspection, or
analysis) that each criterion has been addressed in an acceptable manner. Figure 6
6--7 provides a default
outline for a release description.

Status Assessments

Status assessments provide periodic snapshots of project health and status, including the software project
manager's risk assessment, quality indicators, and management indicators. Typical status assessments
should include a review of resources, personnel staffing, financial data (cost and revenue), top 10 risks,
technical progress (metrics snapshots), major milestone plans and results, total project or product scope &
action items
Environment

An important emphasis of a modern approach is to define the development and maintenance environment
as a first-class artifact of the process. A robust, integrated development environment must support
automation of the development process. This environment should include requirements management,
visual modeling, document automation, host and target programming tools, automated regression testing,
and continuous and integrated change management, and feature and defect tracking.

Deployment

A deployment document can take many forms. Depending on the project, it could include several
document subsets for transitioning the product into operational status. In big contractual efforts in which
the system is delivered to a separate maintenance organization, deployment artifacts may include
computer system operations manuals, software installation manuals, plans and procedures for cutover
(from a legacy system), site surveys, and so forth. For commercial software products, deployment artifacts
may include marketing plans, sales rollout kits, and training courses.

Management Artifact Sequences

In each phase of the life cycle, new artifacts are produced and previously developed artifacts are updated
to incorporate lessons learned and to capture further depth and breadth of the solution. Figure 6-8
identifies a typical sequence of artifacts across the life-cycle phases.
 ENGINEERING ARTIFACTS
Most of the engineering artifacts are captured in rigorous engineering notations such as UML,
programming languages, or executable machine codes. Three engineering artifacts are explicitly
intended for more general review, and they deserve further elaboration.

Vision Document

The vision document provides a complete vision for the software system under development and.
supports the contract between the funding authority and the development organization. A project
vision is meant to be changeable as understanding evolves of the requirements, architecture, plans,
and technology. A good vision document should change slowly. Figure 6-9 provides a default outline
for a vision document.

Architecture Description

The architecture description provides an organized view of the software architecture under
development. It is extracted largely from the design model and includes views of the design,
implementation, and deployment sets sufficient to understand how the operational concept of the
requirements set will be achieved. The breadth of the architecture description will vary from project
to project depending on many factors. Figure 6 6-10
10 provides a default outline for an architecture
description.

Model based software architecture

ARCHITECTURE: A MANAGEMENT PERSPECTIVE

 The most critical technical product of a software project is its architecture: the infrastructure,
control, and data interfaces that permit software components to cooperate as a system and
software designers to cooperate efficiently as a team. When the communications media
include multiple languages and intergroup literacy varies, the communications problem can
become extremely complex and even unsolvable. If a software development team is to be
successful, the inter project communications, as captured in the software architecture, must
be both accurate and precise
From a management perspective, there are three different aspects of architecture.
1. An architecture (the intangible design concept) is the design of a software system
this includes all engineering necessary to specify a complete bill of materials.
2. An architecture baseline (the tangible artifacts) is a slice of information across the
engineering artifact sets sufficient to satisfy all stakeholders that the vision
(function and quality) can be achieved within the parameters of the business case
(cost, profit, time, technology, and people).
3. An architecture description (a human-readable representation of an architecture,
which is one of the components of an architecture baseline) is an organized subset
of information extracted from the design set model(s). The architecture
description communicates how the intangible concept is realized in the tangible
artifacts.
The number of views and the level of detail in each view can vary widely.
The importance of software architecture and its close linkage with modern software
development processes can be summarized as follows:
Achieving a stable software architecture represents a significant project milestone
at which the critical make/buy decisions should have been resolved.
Architecture representations provide a basis for balancing the trade-offs between
the problem space (requirements and constraints) and the solution space (the
operational product).
The architecture and process encapsulate many of the important (high-payoff or
high-risk) communications among individuals, teams, organizations, and
stakeholders.
Poor architectures and immature processes are often given as reasons for project
failures.
A mature process, an understanding of the primary requirements, and a
demonstrable architecture are important prerequisites for predictable planning.
Architecture development and process definition are the intellectual steps that map
the problem to a solution without violating the constraints; they require human
innovation and cannot be automated.

ARCHITECTURE: A TECHNICAL PERSPECTIVE

An architecture framework is defined in terms of views that are abstractions of the UML
models in the design set. The design model includes the full breadth and depth of information.
An architecture view is an abstraction of the design model; it contains only the
architecturally significant information. Most real-world systems require four views: design,
process, component, and deployment. The purposes of these views are as follows:
Design: describes architecturally significant structures and functions of the design
model
Process: describes concurrency and control thread relationships among the design,
component, and deployment views
Component: describes the structure of the implementation set
Deployment: describes the structure of the deployment set
Figure 7-1 summarizes the artifacts of the design set, including the architecture views and
architecture description.
The requirements model addresses the behavior of the system as seen by its end users,
analysts, and testers. This view is modeled statically using use case and class diagrams, and
dynamically using sequence, collaboration, state chart, and activity diagrams.
The use case view describes how the system's critical (architecturally significant)
use cases are realized by elements of the design model. It is modeled statically
using use case diagrams, and dynamically using any of the UML behavioral
diagrams.
The design view describes the architecturally significant elements of the design
model. This view, an abstraction of the design model, addresses the basic structure
and functionality of the solution. It is modeled statically using class and object
diagrams, and dynamically using any of the UML behavioral diagrams.
The process view addresses the run-time collaboration issues involved in executing
the architecture on a distributed deployment model, including the logical software
network topology (allocation to processes and threads of control), interprocess
communication, and state management. This view is modeled statically using
deployment diagrams, and dynamically using any of the UML behavioral
diagrams.
The component view describes the architecturally significant elements of the
implementation set. This view, an abstraction of the design model, addresses the
software source code realization of the system from the perspective of the project's
integrators and developers, especially with regard to releases and configuration
management. It is modeled statically using component diagrams, and dynamically
using any of the UML behavioral diagrams.
The deployment view addresses the executable realization of the system, including
the allocation of logical processes in the distribution view (the logical software
topology) to physical resources of the deployment network (the physical system
topology). It is modeled statically using deployment diagrams, and dynamically
using any of the UML behavioral diagrams.
Generally, an architecture baseline should include the following:
Requirements: critical use cases, system-level quality objectives, and priority
relationships among features and qualities
Design: names, attributes, structures, behaviors, groupings, and relationships of
significant classes and components
Implementation: source component inventory and bill of materials (number, name,
purpose, cost) of all primitive components
Deployment: executable components sufficient to demonstrate the critical use
cases and the risk associated with achieving the system qualities
Software User Manual

The software user manual provides the user with the reference documentation necessary to support
the delivered software. Although content is highly variable across application domains, the user
manual should include installation procedures, usage procedures and guidance, operational
constraints, and a user interface description, at a minimum. For software products with a user
interface, this manual should be developed early in the life cycle because it is a necessary
mechanism for communicating and stabilizing an important subset of requirements. The user manual
should be written by members of the test team, who are more likely to understand the user's
perspective than the development team.

PRAGMATIC ARTIFACTS
People want to review information but don't understand the language of the artifact. Many
interested reviewers of a particular artifact will resist having to learn the engineering language in
which the artifact is written. It is not uncommon to find people (such as veteran software managers,
veteran quality assurance specialists, or an auditing authority from a regulatory agency) who react as
follows: "I'm not going to learn UML, but I want to review the design of this software, so give me a
separate description such as some flowcharts and text that I can understand."
People want to review the information but don't have access to the tools. It is not very common
for the development organization to be fully tooled; it is extremely rare that the/other stakeholders
have any capability to review the engineering artifacts on-line. Consequently, organizations are
forced to exchange paper documents. Standardized formats (such as UML, spreadsheets, Visual
Basic, C++, and Ada 95), visualization tools, and the Web are rapidly making it economically
feasible for all stakeholders to exchange
information electronically.
Human-readable engineering artifacts should use rigorous notations that are complete,
consistent, and used in a self-documenting manner. Properly spelled English words should be
used for all identifiers and descriptions. Acronyms and abbreviations should be used only where they
are well accepted jargon in the context of the component's usage. Readability should be emphasized
and the use of proper English words should be required in all engineering artifacts. This practice
enables understandable representations, browse able formats (paperless review), more-rigorous
notations, and reduced error rates.
Useful documentation is self-defining: It is documentation that gets used.
Paper is tangible; electronic artifacts are too easy to change. On-line and Web-based artifacts
can be changed easily and are viewed with more skepticism because of their inherent volatility.