CSC 223 Grinnell College Fall, 2005
 
Software Design
 

Discussion Questions for Riel, Chapter 9

Persistence

  1. Can you describe the concept 'persistence in space' in more detail? (David D)
  2. Could you explain the theory of presistence and its two sub-types a little more? (Saugar)
  3. Can an Object be truly persistent? I assumed that when using a database, as the book mentioned, for persistence this was just saving the data in the object not the object itself. (Mike L.)
  4. Would objects that use persistence in space still be able to utilize persistence in time? (Alex)

Shallow and Deep Copies

  1. What are the key differences between Deep copy and shallow copy? is one more advatageous than the other, or does its usability just depend on the circumstanaces or the situation? (Gaurav)
  2. What is the advantage of shallow copies over deep copies asides from conservation of memory space? (monica)
  3. Could shallow copying be implemented in such a way (as a wrapper) that the developer would not have to keep track of objects using the memory address? Is this a part of garbage collection? (Elijah)

Logical and Physical Design

  1. In software design, Should we start with logical design first, then make changes according to physical design or We start with physical design, then generate logical design? What's the difference?(Sheng)
  2. Hueristic 9.1 says "Object-oriented designers should not allow physical design criteria to corrupt their logical designs. However, physical design criteria are often used in the decision-making process at the logical design time." How does the second part of the heuristic not contradict the first part? Is using the physical design criteria in decision-making process not corrupting the logical design time? (Saugar)
  3. Riel writes that when caught between 2 equivalent designs, physical design considerations may be useful in making a decision. But shouldn't the physical design be a factor right from the time of initial logical design desicion making? (monica)

Wrappers

  1. What is the difference between a wrapper and an interface? (Jonathan)
  2. Are there any inherent dangers to using wrappes? (Eric)
  3. How does one manage the program that is wrapped by a wrappter? (that is, if the program needs to be updated, what changes have to be made tothe wrapper thats the go-between?) (Eric)

Programming Language (e.g., Java, C++, Smalltalk) Issues

  1. Does Java have a template feature like C++? What are the differences between strong and weak type checking? (Mike C)
  2. Is there a one-to-one correspondence between meta-classes and normal classes in SmallTalk, or can there be multiple instances (classes) of a single meta-class? (John)
  3. What kind of additional behavior can SmallTalk meta-classes take on that C++ template classes can't? (John)
  4. Does Java have the notion of meta-classes? (John)
  5. Can you give an example of a Destructor in Java? (David D)

Concurrency

  1. Are "active objects" the only way to do concurrent programming in object-oriented programming languages? How are they implemented? (Alex)
  2. What's the difference bewteen concurrent object-oriented programming and parallel programming? (Sheng)
  3. How does parallel programming work and what is an example of a program that does it successfully? How does concurrent object-oriented programming relate to problems with parallel programming? (Mike C)
  4. I was under the impression that certain types of programming languages (functional, and maybe declarative) actually reduce the complexity of programming for concurrency. Is this true? If so, will such programming languages become more popular as concurrency becomes necessary for keeping up with Moore's Law? (John)

Other Questions

  1. Riel mentions an author who argued against "minimal public interfaces" arguing that there were classes "whose semantics seem to break". What is an example of this? (Elijah)
  2. When Riel talks about constructors as minimal public interface he says that it should not be possible for users of a class to create objects in an invalid state (all methods should be ready after creation). Does this mean that after an object has been created every method in that object should be able to be called? (Mike L.)
  3. Is there a difference between a metaclass used to capture class-specific data, and class that is part of an inheritance hierarchy? (Jonathan)
  4. From chapter 8, what are the advantages and disadvantages of using Metaclasses and Language-level keyword to deal with Class-specific and Object-specific behavior? How are they different in other languages and design patterns? (chayapan)
  5. I did not understand how we implement polymorphism in a nonobject oriented language? (Gaurav)
  6. How does the last topic in chapter 8, "a Useful Abstract Class That Is Not a Base Class," relates to the Class-specific and Object-specific behavior issue? (chayapan)

This document is available on the World Wide Web as 

http://www.cs.grinnell.edu/~walker/courses/223.fa05/readings/riel-9.shtml
created 8 November 2005
last revised 9 November 2005 		Valid HTML 4.01! Valid CSS!
For more information, please contact Henry M. Walker at walker@cs.grinnell.edu.