a) Explain dual role of software.   (4p)
b) What are basic ways to deal with complexity? (4p)
c) In the following description, explain when the term “account” is used as an application domain concept and when as a solution domain concept:

“Assume you are developing an online system for managing bank accounts for mobile customers. A major design issue is how to provide access to the accounts when the customer cannot establish an online connection. One proposal is that accounts are made available on the mobile computer, even if the server is not up. In this case, the accounts show the amounts from the last connected session.”  (4p)

(May 2005)

a) What is a Principle of Falsification in modeling? Why and how it is applied in software design? (5p)
b) Which types of decomposition do you know? Briefly explain and give their advantages and disadvantages. (4p)

(October 2007)

a) A programming language is a notation for representing algorithms and data structures. List two advantages and two disadvantages of using a programming language as sole notation throughout the development process. (5p)
b) A passenger aircraft is composed of several millions of individual parts and requires thousands of persons to assemble. A four-lane highway bridge is another example of complexity. The first version of Word for Windows, a word processor released by Microsoft in November 1989, required 55 person-years, resulted into 249,000 lines of source code, and was delivered 4 years late. Aircraft and highway bridges are usually delivered on time and below budget, whereas software is often not. Discuss what are, in your opinion, the differences between developing an aircraft, a bridge, and a word processor, which would cause this situation. (5p)

(October 2008)

a) Briefly explain difference between functional and object oriented decomposition. What are pros and cons of each of them? (5p)
b) Why do we apply principle of Falsification in software engineering? (4p)

(October 2009)

Software Life Cycle and UML

a) What is difference between activity-centered and entity-centered life cycles? Which of them is  dominating now? (4p)
b) Draw Boehm’s spiral model as a UML activity diagram. Does the activity diagram improve  readability compare to the original figure? (5p)
(May 2005)

a) Unified Software Development Process (UP) considers phases, iterations and workflows in software life cycle. Explain how they are related. (5p)
b) Specify which of the following decisions were made during requirements or system design:
• “The ticket distributor is composed of a user interface subsystem, a subsystem for computing tariff,
and a network subsystem managing communication with the central computer.”
• “The ticket distributor will use PowerPC processor chips.”
• “The ticket distributor provides the traveler with an on-line help.”
(October 2007)

a) Give an explanation of Spiral Model? What are main problems of waterfall model that Spiral model overcomes? (5p)
b) Software life-cycles may be different for different projects. Why? (4p)
(October 2008)

a) Draw a UML class diagram representing Unified Software Development Process (UP)? (6p)
b) We considered 3 formal software development models: inductive, deductive and transformational. Briefly (without technical details) explain main features of each of these models. (6p)
(October 2009)

Object Design – Interface Design

a) Which different roles play developers in Object design and how it is reflected in UML diagrams? (4p)
b) Consider the List interface in the java.util package for ordered collections of objects. Write preconditions and post conditions in OCL for the following operations:

· int size() returns the number of elements in the list.

· void add(Object e) adds an object at the end of the list.

· void remove(Object e) removes an object from the end of the list.

· boolean contains(Object e) returns true if the object is contained in the list.

· Object get(int idx) returns the object located at index idx, 0 being the index of the first object in the list.

(May 2005)

a) How does contract inheritance work? Explain for each type of constrains. (5p)
b) Assume that we have a Rectangle class that inherits from a Parallelogram class. Write invariant(s) in OCL for the Rectangle class. (5p)
(October 2007)

a) Consider a sorted binary tree data structure for storing integers. Write invariants in OCL denoting that
• All nodes in the left subtree of any node contain integers that are less than or equal to the current  node, or the subtree is empty.
• All nodes in the right subtree of any node contain integers that are greater than the current tree, or the subtree is empty.
(October 2008)

a) Assume that you have the following 3 classes: parallelogram, rhombus and square. Assume that rhombus inherits parallelogram and square inherits from rhombus. Write invariants in OCL for each of these classes taking into account the inheritance of contracts.

(October 2009)

Testing

a) What are guidelines for selecting equivalence classes in testing? (5p)
b) What is path testing? Is it white-box or black-box technique? Does it allow to identify all faults? Explain your answer. (5p)

(May 2005)

a) Explain modified sandwich testing strategy. What are advantage(s) and disadvantage(s) of this
strategy?

(October 2007)

a) Briefly explain which different types of tests are performed during testing. Who performs each of them? (5p)

(October 2008)

a) What is a goal of partitioning into equivalence classes in black-box testing? What are guidelines for selection of equivalence classes? (5p)

(October 2009)

(May 2005)

(October 2007)

(October 2008)

(October 2009)

Moving to the code

a) Explain how a unidirectional, one-to-one association can be realized in source code. Give and  example (6p)

(May 2005)

a) Explain the following mapping concepts: model transformation, forward engineering, reverse engineering and refactoring. Draw a figure showing their relations to model space and source code space. (5p)
b) Explain different ways of mapping inheritance into relational database schema. Give their pros and cons. (6p)
(October 2007)

a) Explain realization of a bidirectional, many-to-many association in source code. (5p)
(October 2008)

There are 2 different ways of mapping inheritance into the relational tables, with a separate table and by duplicating columns. Explain these ways with example and compare them in terms of modifiability and response time.

(October 2009)

Agile Software Development

a) What are main ideas of Manifesto of the Agile Software Development? (4p)
(May 2005)

a) Explain the following mapping concepts: model transformation, forward engineering, reverse   engineering and refactoring. Draw a figure showing their relations to model space and source code   space. (5p)
b) Explain different ways of mapping inheritance into relational database schema. Give their pros and cons. (6p)

(October 2007)

a) What is test-driven (test-first) software development? Briefly explain. (5p)

(October 2008)

a) What is metaphor in Extreme programming? Give an example                (5 p)

(October 2009)

1) The availability of TV channels is greatly extended in the information society. Media experts forecast that every household will be able to receive between 300 and 500 television programs in the near future. A quick estimate shows that a systematic search for interesting TV programs would take so much time that there would be hardly any time left to view the selected broadcasts. Assume that you are supposed to help in solving this problem and to create an intelligent agent for that. List ALL possible agent properties the intelligent TV agent will have and illustrate these properties by examples. (DEC 2008)

Answer:
· The user informs the agent of his TV preferences: delegation Manchester United games and
X-files serial, agent searches in available digital TV programs on behalf of the user and then
informs user what will be shown on TV concerning Manchester United and X-files.
· Agent acts autonomously – without intervention from user, the agent continuously follows
the new broadcasts and television programs and could provide the user with new information
on his areas of interest.
· Pro-activity – the agent may not only inform user about TV broadcast of Manchester United
games or X-files but also about programs with interview with Manchester United players (Xfiles
actors) or inform about games of Manchester United competitors.
· Reactivity – the agent can react to changes in TV schedule.
· Social ability – the agent can communicate to other (TV) agents and ask their opinion about
some programs or knowledge of related subjects or help – if I can’t find a program for some
channel then I can ask for other agents or trying to contact brokers.
· Mobility – the agent can move from one computer to another when traversing the TV
programs.
· Benevolence – the agent will always try to do what is asked of it.
· Veracity – the agent will not reply with false information.
· Rationality – the agent has a purposeful behavior – to inform user about interested programs.
Its rationality is oriented to optimal satisfaction of user’s preferences, f/e scheduling trying to
avoid collisions in watching favorite programs (by finding the same program on different
channels in different time).

2) A simple elevator (lift) is a small room that carries people one floor to another. It can be assumed to have two doors, a display indicating the floor it’s on, buttons one the outside for the user to call it and buttons inside for the user to indicate the floor that he wants to move to.

Assumes that the elevators are to be considered as intelligent agents. List all possible properties the intelligent elevator agents will have and illustrate these properties by examples (DEC 2005)

Agent theory

1)

a) In the class we considered the Wise Man puzzle (see Appendix) and we also considered possible worlds. What are the possible worlds in this puzzle at the beginning and after each man answer? (7p)
b) Which axiom(s) from logic of knowledge are not valid in the logic of belief? Explain (DEC 2008)

2)

a) Formalize the following statements in the logic of knowledge (we consider logic of knowledge as
a modal logic with Ki operator):
i) Agent B knows that Agent C wants to find out the cost of football tickets.

Answer: Kbw

ii) If Agent A does not know the cost of football tickets then Agent B knows that Agent A does not know the cost

Answer:

¬Kap –>  Kb¬Kap

(DEC 2004)

3)

a) Is it true or not that Model Logics are more suitable than Classical Logic for modeling intelligent agents? Explain your answer.
b) Explain positive introspection axiom, negative introspection axiom and knowledge axiom. Which of them are appropriate for logic of knowledge and which for logic of belief? Explain why?
c) Formalize in the logic of knowledge the following problem:
“Agent A wants to find out cost of football ticket. Agent A doesn’t know the cost but Agent A knows that Agent B exists. Agent B doesn’t know the cost either but Agent B knows that Agent C exists. Agent C knows the cost.” (DEC 2005)

Agent Architectures

1)

a) What are basic types of layered agent architectures for hybrid agents? Briefly explain their advantages and disadvantages (5p)
b) Give abstract implementation of the function “action” for symbolic agents and for reactive agents? (6p)
c) What is a meaning of introduction the “see” function in an abstract agent architecture? Give an example (DEC 2008)

2)

a) What are basic types of layered agent architectures for hybrid agents? Briefly explain their advantages and disadvantages

b) What kind of a layered architecture is Brooks subsumption architecture? Explain.

Answers:

a) Horizontal (all layers are connected with input and output) and vertical (not all layers connected to input and output) layers. Vertical can be one-pass or two-pass control.

b) Horizontal – all layers are connected with inputs and outputs

(DEC 2004)

Coordination

1)

a) Describe common coordination techniques and compare them in terms of predictability, reactivity
and amount of information exchange.

b) What are fundamental coordination processes? Give a brief explanation.

2)

a) Give examples of a coordinated action that is cooperative and a coordinated action that is not
cooperative.
b) What is a difference between a norm and a social law in agent coordination?
c) What are fundamental coordination processes? Give a brief explanation.
d) Let us consider the Partial Global Planning. Why this planning is partial? And why this planning
is global?

(DEC 2008)
3)

a) Compare different organization structures according to the number of messages required to assign a task, the results of product manager failure and the production costs.
b) What are main steps of Partial Global Planning? Briefly explain them.

(DEC 2005)