Classic Model Part II

Verification and Validation of Requirement Specifications (I)

• Completeness and Consistency (C&C)

• Internal -- self C&C

• External -- w.r.t. external documents

• Prototyping

• Feasibility

• User interface

• Performance

Verification and Validation of Requirement Specifications (II)

• Traceability

internal traceability

external traceability

• Test case generation

functional testing

requirement review or walkthroughs

• Inspection and Walkthroughs

Other Points

• At the end of each phase, verification and validation should be performed.

• The Waterfall model is never linear as presented in some text books.

• It is always an iterative process.

• Embedded in 2167A or many standard software development processes.

No Silver Bullet

• Silver Bullet is sometime that will solve all the problem

• One of the most famous article in software engineering even though it is negative.

• It was mainly used to argue AI or knowledge-based approach.

• "The essence of a software entity is a construct of interlocking concepts: data sets, relationships among data items, algorithms, and invocations of functions. This essence is abstract in that such a conceptual construct is the same under many different representation. It is nonetheless highly precise and richly detailed."

• "I believe the hard part of building software to be the specification, design, and testing of this conceptual construct, not the labor of representing it and testing the fidelity of the representation."

• F. Brooks, "No Silver Bullet-- Essence and Accidents of Software Engineering", IEEE Computer, Vol. 20, No. 4, April 1987, pp. 10-19.

Two kinds of Problems

• Essence -- inherent

• Accidents -- incur during production

• Past breakthrough solved accidental problems:

high-level programming languages

unified programming environments

time-sharing machines

False Silver Bullets

• Ada or any high-level languages (how about Java?)

• Object-oriented programming

• AI

• Expert systems

• Automatic programming

• Program verification

• Graphical user interface

• Environments and tools

• Workstations

What about Java and the Internet/Intranet ?

• A lot of Startup companies think so

• Only time can tell

• It is the same thing as "Is mainframe dead ?"

• Time tells that it is far from being replaced

Stuffs Worth Repeating

• Important concepts in software engineering

• Waterfall as meta-model

• Waterfall shows inherent problems in software development

• Techniques to address inherent problems:

divide and conquer

abstraction

processes

explicit representation of complex relationships

verification and validation

reusability

Boehm’s Top Ten List

1. Finding and fixing a software problem after delivery is 100 times more expensive than finding and fixing it during the requirement and design phases.

2. You can compress a software development schedule upto 25% of nominal but no more.

3. For every dollar you spend on software development you will spend two dollars on software maintenance.

4. Software development and maintenance techniques are primarily a function of the number of source instruction in the product.

5. Variation between people account for the biggest difference in software productivity.

6. The overall ratio of computer software to hardware costs has gone from 15:85 in 1955 to 85:15 in 1985, and it is still growing.

7. Only about 15% of software product-development effort is devoted to programming.

8. Software systems and software products each typically cost three times as much per instruction to fully develop as does an individual software program. Software systems products cost nine times as much.

9. Walkthroughs catch 60% of the faults.

10. Many software phenomena following a Pareto distribution: 80% of the contribution comes from 20% of the contributors.

20% of modules contribute 80% of faults

20% of modules contribute 80% of costs

20% of faults consume 80% of the cost to fix them

20% of modules consume 80% of execution time

20% of tools experience 80% of tool usage

Others

80% of code was done by 20% of programmers

80% of faults was made by 20% programmers

20% of engineers contribute 80% of tasks

Recent Trends

• Internet/Intranet and Java

• Component technology -- including ActiveX, CORBA, Java Beans, eSuite DevPack

• Frameworks - SanFranciso Project

Internet and Intranet Again

• Internet is becoming extremely popular with many new technologies coming every day.

• Most of Fortune 500 companies if not all are doing Internet projects.

• This is considered one of the biggest technological breakthrough since IBM PC -- said by Bill Gates.

Applications of Internet/Intranet

• Stock transactions, Internet commerce

• Phone books

• Publications

• Restaurant

• Internal communication

• Workflow tools

• Teleconferencing

• International calls

• One of your Project :)

Java and Java

• We will have Java chips, just-in-time Java compilers, Java beans.

• Java Enterprise API, Electronic commerce API, Server API, Media API, Beans API, Security API, Embedded API, Management API.

• Platform independence -- still a dream.

• It will provide dynamic software components and dynamic software distribution

More Ideas

• Introspection -- a builder tool can analyze how a bean works.

• Customization -- customers can pick their preferences.

• Events -- Simple communication metaphor can be used to connect up beans.

• Properties -- layout, color for customization. Enable or disable properties.

• Persistence -- ready for database.

Component Engineering/Frameworks

• The idea of Object Factory, Just like Hardware

• Microsoft ActiveX, DNA

• Sun’s Java Beans

• CORBA

• Include "software bus", software architecture, "software glue", reusable components, and conversion with other component/framework technology.

Let's Talk About Our Projects !

Reference

All materials are from work experience, software engineering books and ideas & notes from my advisor in University of Minnesota